Hey. I’m Andy Baldacci. I’m the CEO of SaberSim, and in this video, I’m going to show you everything you need to know to build winning lineups with our optimizer. We have built a one of a kind algorithm that simulates every single game, play by play, thousands of times. And this gives you the best projections and optimizer on the market. The simulations power absolutely everything that we do, but it also makes our three-step lineup building process a little bit different than the way the traditional, outdated optimizers work. But don’t worry. It is easy to understand, and once you give it a try, you’re not going to want to go back to the old way of doing things.
The first step is reviewing and adjusting projections. Second is choosing your build settings, and third is dialing in your exposures. We have a free course over on our website, under the resources section, that walks through each part of the process in a lot more detail than I’m going to cover right now, and I definitely recommend checking that out. But over the next few minutes, I’m going to share everything you need to know to at least get started and comfortable with the process.
The biggest thing I want to stress is that if you want to win money at DFS and you can’t spend all day building lineups, then you have to focus your limited time where you can add the most value. And with SaberSim and our three-step process, that’s going to be the first and last step of it. Our team of data scientists is constantly working to improve our models, but no model is ever going to be perfect, and that’s why it’s always important to review and adjust the projections.
And with this first step, there’s actually a couple of approaches you can take. If you are comfortable just working directly with projections, you know what to look for, then you can just review them right here on this screen, and make your adjustments without doing anything else. But if you’re like me and maybe you don’t really know where to even begin as far as narrowing things down, or you just don’t have the time to dig through an entire slate of players, you should start with something we call a test build, and see what impact these projections actually have on the lineups.
And so I’ll talk a little bit more about these settings in just a second, but for the test build, all we have to do is just pick something middle of the road, as far as the contest goes. And I like to build 150 lineups just to get a bigger sample, but this isn’t some exact science. What I’m trying to do right now is just quickly get a feel for where SaberSim is at on the slate, and narrow down the players and teams that I should focus on.
And this is really where the power of SaberSim shines, is because with a traditional optimizer, if you just went in and hit build, you wouldn’t actually learn anything from the result, because the lineups that they build wouldn’t be representative of anything that you or your opponents would enter into a contest. To get good lineups from a traditional optimizer, you have to set dozens of rules and create all kinds of groups. Otherwise, your lineups would just be dead in the water. But by running thousands of simulations, we get data that allows our optimizer to understand correlations and variants, and incorporate that in your lineups, along with ownership, so you can get a foundation of high upside lineups without spending hours programming exactly what you want.
And let’s jump over here into this stack tab, and I’ll show a little bit of what I’m talking about. I didn’t put in any rules or change any of the default settings, other than to tell the optimizer that I’m playing a GPP. And look at the stacks that I’m automatically getting. No other optimizer does this, because no other optimizer actually understands what it takes to win a contest, and helps you build lineups that do that. And that’s why you have to fight them to get the kind of lineups that you need if you want to actually win.
And so you still have to put in some work of your own if you want to win consistently, but because of our powerful data, SaberSim acts as your advisor. We’re not just a dumb spreadsheet. So instead of just forcing in whatever players you’re looking for and doing all of the work by yourself, let us help you by building you a strong foundation that’s in line with what you’re looking for.
That being said, let’s just jump in and talk about what we’re actually looking for in the test build. And the short of it is, you’re trying to find places where you aren’t as confident in SaberSim’s suggestions, and you want to do a little bit more digging. So these are typically the places where we’re going to be much higher or lower on a team or player than you’d expect. And if you don’t have strong expectations, you don’t have a strong opinion on a slate, that’s fine. What you should do is focus on the players and teams that simply we have a lot of, as well as players where we have significantly more exposure to them than the field does. So this is what you can check by looking at the projected ownership. That’s where we’re saying, “We think the field in an average contest, 18.7% of the lineups in the field will have Cory Seager in them.” You can look at that and see if you are way over or under the field.
And also, I like to look at the higher value players, especially in a sport like baseball, which is not like basketball, in that in basketball, you’re going to have some backup players that are actually going to start. They’re going to get a lot of minutes, and they are going to be a clear, obvious value play. That’s not even close to anything else out there. But for hitters in baseball, there’s just a ton of variants involved, and so you just always want to at least give those high value plays a second look, to make sure that you actually agree with why we’re assigning them that much value.
On DraftKings, to actually see your lineups, you do have to make two or more changes, but right now we’re not going to dig through all the lineups. That takes a ton of time and doesn’t give you the picture that we’re really looking for here. Right now we’re just getting a feel for the exposures.
So what I’ll usually do is jump into the team stacks first. And this shows me just the team stacks. It says what percent of our lineups has at least two players from each of these teams. And so right away, we are incredibly heavy on the Dodgers and the Rockies game. 90% of our lineups have a Dodger stack in it. 36% have a Colorado stack in it. I’m definitely going to look into that game.
And something else that jumped out at me, this isn’t a huge deal, and just the more kind of you get familiar with the sport, the more you’re following things, some teams might surprise you that you don’t see more of them or any of them at all. And so I just had expected to see a little bit more of San Diego. So I’m going to just look into that as well. And what I’m doing is just off screen, I’ve got a little Notepad file where I just keep track of a handful of players or teams that I want to look into. And the more time I would have in this build, the deeper I would go, and the more players and teams I would actually look into.
Now, let’s talk a little bit about the individual players. And so when I’m looking through this list, I’ll actually start with the pictures. So I’m looking to see where we’re out of line with the ownerships, and that’s not always a bad thing. Oftentimes you have to take some stance on players that are a bit more under the radar, or just more than the field, if you really want to stand out in these big contests. So it’s not inherently a bad thing, but that’s just where I’ll usually start looking. And in general, yeah, we’re getting a ton of Pablo Lopez, and he’s also really high value. So I’m just going to make a note that that’s a player I’d want to just be a little bit more cautious of when we have two times the field on him.
Going through, yeah, we have a bit more Heaney than the field, but it’s only 40% overall, nothing too crazy similar with Kelly. Yeah. The more time I have, the more I’m going to look into these players, but to keep things simple for this video, I’m just going to focus on Pablo Lopez. And let’s jump over at the batters.
So for this, like we’ve already talked about, we have a ton of Dodgers and Rockies stacks, so I’m not going to necessarily look into the players individually. I want to look at that in game on the whole, and make an opinion on that game. But what I’m going to do is sort by value and then just see, “Okay, who is jumping out at me?” And there’s a couple specifically, primarily on Seattle. So I see Jake Fraley, highest value. Not an insane amount, but at least twice the field. We’ve got Taylor Trammell. I’m just making a note of that one as well. And then let’s also check out … Yeah. Sam Hilliard, we have a lot more than the field, and Jon Burr too. Okay.
And so again, there are more players that I would look into if I had more time, but what I’m looking for is, okay, who are the players that would surprise me to be getting this much of? And I just want to do a little bit of work to make sure that I should be getting it. Because this is, again, where having SaberSim as an advisor to you is helpful, because we’re not just randomly trying to fit as many people in your alliance as we can. We are looking at the players’ upside. We’re looking at their range of outcomes. We’re looking at how they correlate to everyone else. So there are going to be a lot of under the radar plays that we put in there that you wouldn’t have naturally thought of.
You can kind of think of us as an architect. You’re building your house, you tell the architect what you want, and they come back and they’ve got a blueprint for you. And yeah, sure. You’re going to have to make a good amount of changes. Especially in that first round, you’re going to have to give feedback, have them throw out a few things, add a few other things. But along the way, just drawing on their expertise, that architect is going to think of some things that you wouldn’t have thought of on your own, and you want to make sure you don’t just immediately discount them. So that’s why I’m not just going in here and being like, “Oh, I don’t like this guy. I don’t like that guy.” And just mess with all the exposures. I’m really focused on giving the instructions to the architect to build the right blueprint for my lineups.
If really all you have is 15, 20 minutes, then your process is probably going to look pretty similar to what I’m doing right here. It’s finding those biggest outliers and focusing on them, because it all goes back to the 80-20 principle, where 80% of the outcome comes from 20% of the effort. Finding those biggest outliers is where you’re going to get the biggest reward for your research. If you have more time, you can do more research into some of the marginal plays, and this and that, but that just won’t move the needle nearly as much as those bigger outliers.
And I’m not going to walk through my exact research process for how I verify all this. We have a lot of videos on that subject. And one I would recommend for anyone watching this is a new one we put out, The Six Secrets to Beating Daily Fantasy Baseball. And that’s where DFS Pro and SaberSim partner Max Steinberg walks through his research process and talks about the types of things he’s looking for. We also had Danny Steinberg, Max’s brother, and Matt Hunter, a SaberSim founder on that video, and we went really deep into the research process. So if you have questions about that, watch that video.
For now, I’ll just kind of talk about what I would do with this. And in short, it’s not that complicated. If you feel like we are too high on a team or player, lower their projection. If you think we’re too low, increase their projection. Let me start with making adjustments to teams, because this is actually one of our coolest features, where over on the left, you have our game projections, which shows the number of runs we expect each team to score on average from our simulation data. And if you think we’re too low on the Padres, which right now I do, rather than trying to figure out how much to boost each player on that team and then lower the opposing pitcher and all that, I can just increase the average projected run total for them right here, and we’re going to use our simulation data to accurately adjust all the players in that game accordingly. So I will add 0.7 runs to them, and that’s actually a pretty big adjustment.
And this isn’t an exact science. You’re not trying to get these adjustments exactly right. Your goal is to move things in the right direction. For game adjustments, I don’t recommend changing a team’s total by more than a run, while for players, which we’ll talk about in just a second, you can use their value as a guideline.
A good place to look for this is just compare our projected totals to Vegas. Vegas is obviously very good at setting odds. We would put our numbers against theirs any day on average, but there are going to be spots that each of us excel at, and so it’s always helpful to leverage different points of data and use that as ways to figure out, “Okay, do I want to get closer to Vegas, or do I think SaberSim is right on with this one, where they are a little bit different, but maybe there’s a reason to it?”
And one other important thing to note is, we saw that we were getting a ton of the Dodgers. I don’t actually think that I should lower their projected run total here, because when I looked into it, we were already lower than Vegas on them. So if I was using Vegas as kind of the gold standard, which I don’t, but I do use it as a good point of information, I would increase their run total rather than decrease it. And so if I don’t disagree with the projection after doing it, I’m not going to change it here. That doesn’t mean that in my final lineups, I’m going to just have 90% of the Dodgers as a stack. That’s something that we can account for in the last step of the process.
What we really want to do here is adjust the projections in the right direction, if we think that is actually what the expected outcome is going to be. Otherwise, if it’s more just managing exposures, we’ll save that for the last step of the process. And so for pitchers, if I think Pablo Lopez is too high, which I do, what I will do is look at all the pitchers, and to see the range of values. So the top plays seem to be right around two. Pablo Lopez is significantly above that. And then there’s a handful of guys that are much closer to one, and then you get into like the relief pitchers below one.
What I’m going to do is lower Pablo’s projection based on my thoughts here. So if I thought he was just truly a horrendous choice here, I may lower him to a 1.5 value, but to go even to that or lower than that, I would need pretty strong evidence before making that big of a change. Because SaberSim is saying this is clearly the best value play for the slate. And so to completely disregard that, it’s not to say we’re always right, because we aren’t, but to completely disregard that, I would need a good amount of evidence.
On the other side of the coin, if I thought that they were an even better value, I would just want to be cautious, because they’re already the highest one by a good margin. So theoretically you’d go to three, but that’s 50% higher value than anyone else on the slate. So this to me seems like the top end that I would ever leave them that. But if it was someone you were trying to raise, you thought we were actually way too low on them, keep that top end in mind. I do think going to a little bit higher than the other top values on the slate is okay, but you just don’t want to get out of whack. Because again, these aren’t random numbers. We put a lot of effort into our models. So you want to tweak them, move them in the right direction, but don’t completely throw out what we’re giving you.
So right here, what I’ll do is put him at 14. I don’t think he is a horrible play. I do think he should be a contender for one of those top picks, but I don’t think he’s the standout play. So I’m going to lower him to 14, which is two times value. And I should have mentioned this earlier, but the way you calculate value is you figure out how many thousands of dollars does a player cost, so in this case, it’s seven. To get two times value, you multiply seven by two. That gets us to projection, which is 14. And again, though, what I’m really trying to emphasize is that we’re not trying to get this perfect. Projections are obviously a huge part of DFS, but they aren’t the only part, and we do a lot of those other parts, like looking at correlations, looking at ownership, and looking at upside in general, we do those very well. So don’t obsess here, because this is not the only work you’re doing. You’re touching all parts of the process to consistently add value. And really here, you’re trying to just move things in the right direction.
So a few of those other hitters that we talked about, typed out Fraley, same thing here. Three is the top end, but a lot of players are a little bit below three. Fraley, near the bottom of the order. I don’t want to, again, completely discount it, but I don’t think he necessarily should be as high up there. Trammell a little bit higher in the order. A little bit cheaper, though, but maybe we could put him at … Just call it. We want to get them to three. So that’ll be 6.6. He’s actually leading off, or projected to lead off, so I’m not going to touch that one. And Hilliard, similar to Trammell. Let’s go 6.8, so a little below three. And if I had more time, I would probably just look at all of these high value players, but I just want to show you the basics and how to get started. And I’m really just focused on the ones actually came up in the build.
So now let’s actually go to the next step, which is choosing your build settings. This step, to be completely honest, is almost a formality. Like I said before, because we’re simulating every game thousands of times, we’re able to understand correlations and variants, and incorporate that information, along with ownership, into the lineups that we build you, and we automatically adjust how we’re doing this based on the contests that you’re playing using these sliders right here. So you just simply tell us the contest that you’re building lineups for, and we will automatically adjust the sliders accordingly, based on what we have predetermined to be the best settings on average for that contest.
But it also goes beyond that. We have different defaults for every sport, and even within the sports, we have different defaults for all the contest sites, obviously, but they change based on the slate size. So a six-game slate will have different settings than a 12-game slate, than a two-game slate. We have spent a lot of resources into testing needs, and we’re always working to continue improving them. But this is just a spot where you can let us handle this part of the process, and then you can focus your time on those other steps, where you can actually add significant amounts of value.
And I’m not saying just completely ignore these. If you do want to know more about what’s going on behind the scenes, check out this video, and feel free to play around with it. But like those projections, you don’t want to make massive departures without really understanding what’s going on behind the scenes. And this video will do a great job of helping show what is going behind the scenes.
When it comes to actually building our lineups, there are generally two approaches that we recommend. One is just doing a single build for all of your contests. And in that case, I would just select, sort of like we did for actually the test build, I would just select something that’s based on all of my entries, about middle of the road. And the other alternative, though, is doing a separate build for each entry limit you’re in, meaning doing one build for your single entry contest, another for three max, and so on. I typically will do that second approach, and in that course I talked about, I walk through how I do that and why, but the reason for this is that the strategy for a single entry contest is going to be noticeably different than a 150 max.
And so I want to capitalize on those differences by not treating all of the lineups the same. But there is no right or wrong approach here. I probably wouldn’t recommend going all the way in that direction, of just doing a separate build for every single contest you’re playing, because the differences won’t be that big. You’re trying to, again, utilize 80-20. What is a way where I can quickly get a lot of impact for my efforts? And this is really where it’s going to be. For this video, though, what I’m going to do just for sake of time is a single build. But again, if you want to see me do a build for each entry limit, check out that course over on our website.
And another point that I want to touch on is, I’m going to build a unique lineup for every entry that I have, in order to spread out my risk, and also just give myself more chances at a first place prize. If you’re not as concerned about risks, though, and you just want to bet more on your top lineups, you can definitely take a more focused approach, and enter some of your lineups multiple times. There is no right or wrong approach here, again. So try not to overthink this, but if you are trying to grind out wins more consistently, and you’re not trying to go as boom or bust, I would recommend doing a unique lineup for every entry that you have. I’ve already reserved some entries, so I have 240 out there. This is pretty middle of the road for what all my lineups look like in terms of the contest, not messing with any of the settings, and hitting build.
As we get to this last step, dialing in your exposures, another thing I want to point out is that rather than just building you the exact number of lineups that was requested, we build you a pool of lineups much larger than that, so that you can make these exposure adjustments immediately rather than having to build an entirely new set of lineups like with traditional optimizers. If you say you want a different exposure of, say, the Dodgers, we will then sort through this lineup pool and find the best set of lineups that match the exposures that you’re looking for right away. You don’t have to go back and start from scratch over and over again.
We often refer to this step as quality control, because what you’re first doing is checking to make sure the adjustments that we just made worked as planned. And then we’re making some fine tuned adjustments from there to get everything dialed in the way that we want. If you find yourself making major changes here, then it’s really best to go back to the home screen and make your adjustments there rather than trying to fight what SaberSim has given you. Because again, if the foundation that we give you doesn’t match what you’re ultimately trying to build, start over rather than trying to build on top of it anyways.
So let’s just kind of quickly go through what we got. We’ll start with the teams. I’m still getting a lot of the Dodgers, and that makes sense. I didn’t adjust their projections at all. This is the type of larger adjustment that it makes sense to do here, because I’m not saying I disagree with the projections. It’s that I just don’t want this much exposure to a single stack. If this were an under the radar play that I really had a lot of confidence in, I could see keeping it to around maybe 50%, 60%, but it’s the Dodgers at Coors. This isn’t a shocker. This is going to be a really popular stack. And while we are accounting for ownership, I wouldn’t be surprised if especially this early in the season, people are just purely following rules of thumb and heuristics and just stacking cores really heavily. So I do still want exposure to them, because I think there’s a reason people like this situation, but I’m going to lower it to 40%.
We can also see that I am getting a lot of San Diego, and this, it’s a bit more. This actually makes me think that I made too big of an adjustment on the home screen. So I don’t need to start it from scratch, because I’m not bringing him back down to like the 10%, 15% they were originally at, but let’s just put them to 20%. And again, what we’re doing behind the scenes is sorting through this pool of 1,500 lineups to find the 240 that best match what exposures I’m asking for here.
Now, let’s look at the players. We’ll go back to the pitchers. And so for Lopez, yeah, we’re getting a lot less of him. We’re getting less of him than the field is. That’s what I expected. I could see potentially going back in and doing another round of edits here, maybe on San Diego, maybe on some of these other pictures. I didn’t look to it much, but for right now, what I’m doing is just doing a double check and seeing if this is close enough for me to just dial it in, or if I should go back in and work on some of those fundamental assumptions.
Let’s look over at the batters. And yeah, we’re getting a lot of Berti, which we’re okay with after looking into it. He’s supposed to be leading off, and that’s all right. But maybe we’re like, okay, this could still be a little bit too much. We’re three times the field on him, so maybe we just want to keep it to two times the field. We’re still saying SaberSim likes this play. I like this play, but I don’t necessarily like it three times as much as the field does.
Then who else did we have? We had Fraley. Now, we are a little bit below the field on him, which again, we weren’t trying to remove him entirely, just wanted to get something we felt was a little bit more realistic. And then Hilliard and Tremmell. Actually both these guys. Yeah. So for each of these cases, I’m fine with this. We have a little bit more than the field. We’re not going nuts. This is where I will trust it. Maybe there is more to it than I would have initially thought, and so I’ll run with these ones.
That’s really the process. Like I said, if you have a lot more time, you’re going to follow these same steps, but maybe you do another round of those initial adjustments. Maybe you start doing some research into some of the other players that are popping up. Whatever it may be, the process isn’t really changing. You’re still going through these steps. You’re just spending more time on those spots where you can add the most value, specifically reviewing and adjusting projections, but also dialing in your exposures.
And now the last part is, we just need to get these lineups into DraftKings. I will save them here and they’ll get loaded up into that. And this used to be a real pain in the ass. It required copying and pasting from different Excel files, and just kind of praying that you didn’t make a mistake. With our new Entry Editor with Late Swap, that is all a thing of the past. I already have my entries loaded in here and they just had a random lineup I put in to reserve those entries. And what I’m going to do is fill the contests.
We’re going to pick this bill that I just saved, and I’ve got videos that explain all this in more detail. But I had said I want to put a unique lineup in each of my contests, so I made sure to build enough to cover all of my entries, and I’m going to fill unique. Hit fill. You can see that this all got applied. My exposures got updated. Over here, download and open DraftKings. Click “Upload CSV.” Just find the file that just went in my download folder automatically, and we’re done. Now, when I go to my lineups, I can see that all of these lineups are in there. Really it just took seconds.
This is the process. Like I said, you’re going to adjust how much time you spend where, based on just how much time you have available, based on your experience, where you think you personally add the most value. But this is the process. There is no more spending hours and hours fighting with outdated optimizers. Instead, you can spend your limited amount of time where it makes the biggest impact, and that means you can get an edge on the field. You can get an edge on the competition without spending all day on DFS.
If you haven’t started a free trial yet, just head over to sabersim.com and you can get signed up completely free. If you don’t like it, you can cancel your account in just a couple of clicks, and you will not have to pay us a single cent. But because we’re so confident in our tools, that’s why we offer this free trial. We want you to get in there and see for yourself. And if you have any questions at all along the way, you can always contact us using this green question mark in the bottom corner of the screen, or by emailing me personally. We’re, again, always happy to help. So please do not hesitate to reach out. Thank you and good luck.
When do you use Late Swap? Is it only if one of your players is injured?
Matt Hunter:
Not necessarily. You can do it that way. If you want to just use the Late Swap feature, when you have a player that’s out. That’s totally reasonable. You’re still getting edge over not doing that, obviously. You’re certainly getting edge over just doing a global swap on DraftKings. So if you’re short on time, you’re on your phone or you’re out… away from your laptop and you just want to do a quick plates off if there’s just major news, you can do that. For me, I generally will run Late Swaps before every single set of games locked. Tonight, I think there’s some games at 4:30 and then there’s some at 5:00 Pacific and then a couple of late ones as well.
I’ll probably run Late Swap before every set of games, assuming that projections update. You can tell on our Slack channel, we have notifications when Sims start and stop. You can also tell on this last updated timestamp if the projections have updated. For me, I think any sort of edge that I can get from Late Swap, I want to take advantage of. Even if it’s just… there were unexpected starters or there’s just some minor change in the projections, even if it’s just a couple points here and there, I like to run it every time because I want to take advantage of the up-to-date news and it’s really easy to run it late. So opposite.
There’s really no reason not to, unless… I think the reasons not to actually are if you’re again short on time, and you just don’t have the time to run the Late Swap after… before every lock, or if you’ve done a lot of changes to your builds, kind of post build managing exposures, or if you’ve used our filter tool and unselected certain lineups, and you’ve done a lot of tinkering to your builds to really kind of get them to where you want to be.
Late Swap will still account for your exposure adjustments. We’ll still respect the min to max exposure, but if there’s not huge changes, you don’t necessarily need to run Late Swap before every game locks. It is extra time that you’re having to stand during the evening to do that. Even though it’s quick, it’s still… I wouldn’t say it’s hugely necessary, but if you want to get the most out of the tool, I would say it’s worth it to run before every set of games lock, assuming that our projections have updated.
Max Steinberg:
I sort of see it as a priority ladder, right? There are certain situations where you definitely want a Late Swap. This is when there’s an injured player that’s unexpected, that’s actually in your lineups. That means you literally need to get that player out and you’re going to Late Swap or do what are the features that we have that’s called quick swap.
Then if there’s a significant injury, that’s a little below, but you’re probably are going to want a Late Swap. When there’s minor injuries, you could test it out. You can do a Late Swap. You can actually see how much your lineups have changed from it, and then you can say, okay, do I want to use this? Do I not want to use it? You don’t have to use the Late Swap, but you knew this sort of cast it out and just see, okay, how much are my lineups changing? Do I actually want to do this? But there’s certain situations… There’s actually an injured player in your lineup. So we’re just going to definitely want to do it instead of just going to the drafting stage and trying to manually do it yourself.
Danny Steinberg:
I mean, just adding onto what Max said. I think basketball’s a little unique in the sense that if you’re in the game, you’re going to get some stats and minutes tend to have like a much stronger… or playing time tends to have much stronger correlation with fantasy points than another sports. If there’s some guys announced the surprise starter or there’s a surprise injury, often the projections will change enough to where you’re going to want to Late Swap. Now for me personally, I tend to use Late Swap about 50% of the time, and I don’t run it before every game. But if there’s a surprise starter or a surprise injury, then I think it absolutely makes sense to do it when that happens, because oftentimes you will have significantly different optimal lineups than you did before the lock.
Does Late Swap change your initial lineup building process?
Max Steinberg:
I definitely do. If there’s an early game, especially if there’s an early value play that might not be as good of a value play, if there’s some injury later, I’m going to just use these max and min ownership buttons that are buttons [inaudible 00:00:42] in the [inaudible 00:00:44] process to actually suppress some of these players’ exposures in my lineups and increase it with players who are in later games, because that’s going to give me the opportunity later to make a better Late Swap. If you use a guy who’s a pretty good value and use him a lot, and then he fails and then there’s this huge value opens up later, that’s a huge missed opportunity that I want to be able to not miss. I want to take advantage of it. So, absolutely.
Matt Hunter:
Yeah. I think I do probably a little bit less of the adjustments in my initial build for Late Swap. I think the main thing for me is I just don’t care that much about my exposures for the later games so I don’t need to… Because I know that I’m going to be late swapping, I’m not going to be worrying too much about my exposure or projections for those later-game players which obviously it’s not… I probably could be doing a little bit more strategy, like Max said, of having maybe more of the later-game players or trying to take advantage of it.
But I think a lot of… Especially newer players, might see, for example, a questionable tag on LeBron James and lower his exposure because you don’t want to… if he gets rolled out and you don’t want to get caught with a bunch of LeBron James’ that you’re trying to swap out.
But with Late Swap, there’s really not much of a reason to do that, unless it’s really like that’s the only game at the end of the slate, there’s no other options in that price point, then you might want to think about, “Okay, maybe I’m going to be stuck with no other options.” But especially for a slate like tonight where there’s, I think, two games late and there’s probably other players that we can swap to, especially with Portland, sure, we can do some swaps there. I wouldn’t care about having 50% LeBron James in my initial ability, even if he’s questionable, even if he is unlikely to play. That might make me even more want to have a lot of him just because his ownership is going to be so deflated. And I think, just anecdotally, anytime I see that question, we’ll tag on a player and then I look at their ownership later on in the slate, it’s always way lower than it should be. So I think for me, it’s just more I don’t make the changes that I might have without Late Swap.
Do you ever use Late Swap before lock?
Matt Hunter:
No, I don’t, unless you’re really tight on time. So right now, I don’t have any exposure caps, but if I had set some min and max exposures on my build, those would carry over here. And so if I had set a bunch of those and I don’t want to redo, I just want to swap with my same exposures based on the new news, I might do that. I might also do a quick swap if, say, I have my lineups and somebody from an early game is a late scratch, and so two minutes before lock, somebody, Embid or something, gets ruled out and I just need to really quickly get them out of my lineups, that’s when I would use maybe quick swap. But, for the most part, I would just run a new build and apply it, just use the field contents with lineups and apply it rather than using Late Swap, if it’s before a lock.
How do the different Swap methods work? Which should I use?
Matt Hunter:
First of all, quick swap is basically like a better version of DraftKings or FanDuel’s global swap. So essentially we will look for all out players. If you just go by default, we’ll take all out players that are in your lineup and we will replace them with the next best available player for whatever position they’re in. So again, say LeBron James gets ruled out. I would just do this quick swap. It would just replace LeBron in every lineup with the next best player for that lineup. It might not be the same player in all of them, it might be a different player in half lineups and another one in the other half, but it’s basically a really quick way so we don’t have to rebuild all the lineups. It’s only affecting the lineups that have that player.
Andy Baldacci:
And it’s also going lineup by lineup. With the global swap tools on DraftKings, FanDuel, I’m assuming Yahoo, the way it works is you pick a replacement for it and you have to pick a player typically who’s under, who has less salary than the player you’re replacing, or if you pick someone that’s over, again, either way, it’s going to blanket, replace that person in all of your lineups. Either you’re going to be leaving a lot of money on the table, or it’s not going to work on a lot of lineups. What we’ll do is lineup by lineup, figure out how much salary of left and maximize that use, and then show you who got in, who got out and all of that.
Danny Steinberg:
I have a question. How quickly do we get the out status updates? And is it possible to designate a player as out if they don’t have an out status?
Matt Hunter:
We get those status updates very quickly. The SIM itself that updates the projections can take a little bit longer because a player being out will affect the rest of the game or rest of the team, certainly. But in terms of the actual out status, we’ll get that very quickly. If a player gets ruled out within a minute or two, really, you can go to quick swap and just use the all-out players, but if I just uncheck LeBron here, or if I set his projection to zero, and then I go to quick swap and I just apply it, it’ll actually remove him from all of your lineups.
Andy Baldacci:
If you set the projection to zero.
Matt Hunter:
Yeah, I think it’s setting it to zero. It’s not the unchecking because that’s the max exposure. If you set them to zero, you run the quick swap, it’ll swap him out of all your lineups. But generally, if he was ruled out, if I just clicked refresh projections, or even if you’re on another page and you’d go to entries, it’ll update him, mark him as out, and then you just have to do the quick swap. Again, that’s generally not the last swap that I’ll do. I’ll do that right after the news comes out just to make sure I get him out of the lineups. And then once the SIMs update, then I’ll do a late swap, which will actually rebuild all the lineups so that you can get more of the other Lakers that benefit from him being out and get the full benefit of that tool rather than just the quick swap, which I think the value is more just having the really quick changes there.
Andy Baldacci:
Yeah. Before we jump into the full late swap, a couple of things I just want to touch on. To Danny’s point, we’re always investing in faster and better sources of data. We’re very happy with where our injury news setup is right now. We’ll react very quickly by throwing up the out status tag as soon as we get that information. However, because the way we create our projections is by simulating them out thousands of times, it’s going to take sometimes five or 10 minutes to get fully updated projections. In those cases, it’s really tight. We do speed up the process. You can trust us that we’ll update the out as fast as possible.
One related question we got on this is that if I saw a player out, does it only change that player at that position? Or does it optimize the entire lineup and switch out other players in the lineup as well? That was from Jeff. For that question, if you’re using quick swap, while we will use as much salary from each individual lineup, it is only handling that specific player and it will not touch the rest of your lineup. That’s where the full late swap functionality really comes in. And Matt, if you want to just touch on that a little bit.
Matt Hunter:
Yeah, like Andy said, the quick swap will only swap out that one out player. With late swap, what we’re doing is we send all of those lineups to the builder and the builder is going to lock in the players whose games have already started, so they’re locked into your lineups. Then, basically the rest of the lineup will just rebuild based on the new projections and all of your other settings. We do account for correlation and ownership, and we use the smart diversity with the SIMs in the same way.
If you run late swap, all we’re going to do is just lock in those players that obviously can’t be moved because the games have started and then the rest of the lineup will change. Then we’ll do that individually with every single lineup that you have and make sure that they’re unique so that you’re not going to have duplicate lineups, if the best possible lineup is the same for two of them. We’ll make sure that they’re unique just like with your original builds. You end up with 150 or however many lineups you swapped, you’ll end up with a new set of lineups that account for all of the changes and projections. We’re essentially re-optimizing, rebuilding those lineups.
Andy Baldacci:
What we always recommend when it comes between quick swap and late swap is if someone was ruled out, always do a quick swap. It’s the same reason why we replaced dummy lineups right away is you just want it as insurance, make sure you’re not going to be screwed if you don’t have enough time to get the full swap in or anything else. Always do quick swap, always upload those lineups as soon as you know someone is out and then see what you have time for.
Should I do a different Late Swap for each contest I’m playing?
Matt Hunter:
Yeah, that’s a really good question. So, a lot of it depends on your time and then it also depends on the nature of the swap and the players that are… That changed. So if you have time, if you have… However long it, I mean it doesn’t take that long to run the slots, but so if you have time and you want to do a different swap for each of your contests, especially if you had done different builds for each of them, then I think that’s great. And that’s probably going to give you even better lineups because we’ve set these default settings based on the size of the contest and the entry limit. So if you want to really get the most out of Late Swap, I think you can run a different one for each contest.
For me personally, I, because I generally entering a lot of different contests and because, there are scenes running all the time and stuff. Sometimes there’s only five minutes left before I need to get the lineups in. And so I usually will just run a single Late Swap.
And I think really especially when there’s big news, if there’s a major player injury and you have suddenly new value that’s popping up, I don’t care too much about those differences between the slider settings, because that’s so much less important than the value you’re getting from the new projections and the new players that suddenly you have a lot of leverage from. So, again it really just depends if you’re doing a different bill for each of your contests initially and you have the time to do that for Late Swap, by all means, do that. Just keep in mind that there is a time constraint and if you’re short on time and you just want to do a single Late Swap for all your contests, there’s nothing wrong with that. And you’ll still get a lot of edge if there’s that big news and you’re counting for that.
Andy Baldacci:
And one thing we should have touched on in the last question, but it does apply here as well is a lot of this, whether it’s just Late Swap or just how you’re actually building your lineups comes down to time. If you are a professional and have all day and are playing a sport like NFL or baseball, where you typically have a good amount of time between when you know the final confirmed lineups and the game starts, yeah you should be built doing separate builds for the different contests types and getting as precise as you can, because all those customization will add edge to your strategy. But if you’re doing this on the side, if you’re doing it for support like NBA, where you just simply don’t have that much time, you’ve got to make compromises. And so what we always recommend when it comes between quick swap and Late Swap is if someone was ruled out, always do a quick swap.
It’s the same reason why we replaced dummy lineups right away, is you just want as insurance make sure you’re not going to be screwed if you don’t have enough time to get the full swap in or anything else. So always you quick swap, always upload those lineups as soon as you know someone is out and then see what you have time for. And I would say, do a Late Swap build across all of your contests, then get that in if you have time. And then if there’s still a little bit more time left, you could pick say your higher dollar buy in contests and do a specific swap for those. And it’s really just fitting the process into to how much time you have available based on minimizing risk. And then based on just getting the most customization and effort into those higher dollar contests. Yep.
Max Steinberg:
So one thing that the Entry Editor can do as well is, and I think we touched on this a little bit, but maybe we didn’t is you actually just do a Late Swap for specific contests, right? You could say, I just want to do a Late Swap for the minimax, or I just want to do a Late Swap for my single entry or whatever. Right? And so the question is, A, can you do that? And B, do we want to do that? And when would you do that? Do you get what I’m asking?
Andy Baldacci:
Yeah. Matt, do you want to, so I think we touched that a little bit where just to summarize what we were saying before is that if you had an unlimited amount of time, you would want to be as precise with your Late Swaps as possible and do them for specific contests at that in-lab different [inaudible 00:04:31] them, but it’s finding that balance based on how much time you have available, whether it’s because lock is, not lock, but tip-off is coming right up or because you just simply don’t have that much time, maybe dinner’s coming up, but Matt do you want to show how people actually do that?
Max Steinberg:
Right. I just, I would love to just demonstrate, how do you do it for just a specific contest? Or how do we make that?
Matt Hunter:
Yeah. So, right. So it’s pretty easy to [inaudible 00:04:58] Late Swap just based on whatever contest is checked here. So I can uncheck and just check the minimax and then run a Late Swap just for that. And that’ll just swap the minimax, maybe I’ll check all of them and, or I’ll just want to run it for a single entry or something. I can check my single entry contests and swap that way. And then I’m just swapping for the single entry contest. So it’s pretty easy to do that way. And so yeah, if you have the time and then you can change your settings and only run it based on the single entry settings, which are more conservative with smart diversity and ownership because they’re smaller contests. So yeah, that’s how I would do that.
Max Steinberg:
And we also categorize your entries for you as well. Right. You can say just [crosstalk 00:05:54] the satellite, just do it for GPP. Just do it for you, right?
Matt Hunter:
Yep. So you can choose it from the dropdown here. I generally don’t play cash, but if you play cash, you can just select cash and then run your Late Swap just for that contest type as well. That’s probably, that would be the simplest way of doing it. So that’s kind of the in-between between just doing one Late Swap for everything and doing one for each individual contest, let’s just base it off of the type of contest. It’s cash especially you really want to have different settings there. So if you’re running GPP, lineups and cash, I would recommend doing different Late Swaps unless you really have no time. I would separate those out because obviously for cash, you just want the best projected lineup period. And for GPP, you want to care more about those other factors.
What sliders should I use for Late Swap?
Andy Baldacci:
Before you jump in, Matt, I just want to give some context in that, what is going to happen when you open this, is that it will pull in the sliders from your most recent build to give you as a starting point. You can make adjustments, but the starting point of what you see will be the sliders from your most recent build. And then from there, Matt, are you making any adjustments? Or how do you approach that?
Matt Hunter:
Yeah, so it’s an interesting question. I think, by default, I’m not making too many adjustments to the sliders for Late Swap if there hasn’t been big news. So I will usually make minor changes at the slate progresses. As there becomes fewer and fewer games left, you don’t need the smart diversity as much because we have fewer players to choose from and any changes to projections that there have been, we want to take advantage. The whole point of Late Swap is to get the edge from taking advantage of those changes. And so the lower the smart diversity, the more you’re accounting for those for projection. So especially if there are big changes where there’s a major player that’s ruled out, or a surprise starter, I will lower smart diversity, especially a lot, because I want to make sure that I’m getting as much leverage out of that player as I can.
But even without those major changes, I’ll usually lower sliders a little bit as the slate progresses, just because the diversity doesn’t matter as much when there’s fewer players to diversify. And then with ownership fade because ownership is important as a whole, but as we get later into the slate, the ownership of those players is less important, I think, than the edge you’re getting from the projection changes. So if I would recommend personally, if there’s a big news, LeBron gets ruled out and you want to account for that change, I would just set ownership fade to zero and set a lowest smart diversity because… I mean, I might even just set them all to the left because you want to really just get as much of the projection edge as possible from the new starters, rather than trying to diversify too much and trying to get too cute with the sims. You just want to really get that leverage.
Max Steinberg:
Yeah. I’d love to add onto this because I think, one thing is, just to clarify, your lineups probably are not going to be tremendously different whether you kept the settings that Matt was just at, or you do it to zero in a Late Swap scenario. Because, let’s say it’s seven o’clock Pacific time, there’s two more games left, and we’re swapping your lineups. It’s the difference between, if someone’s way better, he’s going to be way better than majority of the sims, and using our simulation data, it’s going to come to mostly the same conclusion. And so, I don’t think it’s a huge deal personally, but this is more advanced stuff of how much you want to consider our simulations versus how much you want to consider average projection when doing a Late Swap. If you want to consider average projection more, you’re going to move the sliders to the left. If you want to consider our simulations data more, you’re going to move the sliders to the right. And so it just depends on what you want to do. And sometimes it makes sense, right? What Matt is saying is, in the later games where you might just want to do a Late Swap based mostly on average projection, and that’s fine. In that case, you want to move the sliders more to the left.
Are the same slider settings used for all of my contests when Late Swapping?
Yes. Unless you select each contest individually, or by style, and do a separate swap for each of them.
Do the projections update automatically or do you need to manually refresh?
Each time you load the site, we will automatically update the projections, but if you’ve left the site open you will need to manually refresh the projections by clicking the “refresh” button in the top right corner of the app (below the account management link).
How do min and max exposures work with Late Swap?
Matt Hunter:
For normal builds, the way that our min/max exposure works, is that we build a pool of 500 to 1500 lineups and then we sort those and filter those down. So when you set min-max exposures, what you’re really doing for a normal build, is you’re starting from that pool of 1500 and then you’re filtering those so that you’re pulling different sets of lineups from that pool to get your final 150 or 20 or however many you want. For Late Swap, we don’t build that whole pool. We’d swap the however many lineups you actually have in your entries, and so when you set this min/max exposure, we don’t have a pool to draw from, we’re just sending those exposure settings to the builder when you run the light swap and then the builder is accounting for that as it builds. So it’s just…
I just wanted to clarify that part of it because it might not be fully apparent off the bat that it’s a little bit of a different process where Late Swap is not just pulling from that big pool of lineups like a normal build.
Is it normal for almost all of my lineups to change when I Late Swap?
The earlier in the slate you are when you late swap and the higher your Smart Diversity, the more your lineups will change. In most cases the changes themselves won’t be major, but most lineups still will be different than they originally were and that isn’t cause for concern.
Does Late Swap use Projected Score or Saber Score?
Matt Hunter:
So essentially how the builder works, normally, not the Late Swap, but just for normal builds is that you’ll set your settings for correlation, ownership and smart diversity, and then the builder will take those settings and it’ll account for those settings as it’s building. So it’s going to incorporate correlation into your lineups and it’ll incorporate ownership fade into the lineups and the smart diversity kind of controls how the sample of Sims that we’re pulling from it’s when we’re creating those. So that’s already accounting for those sliders and that’s sort of in a way that is accounting for Saber score.
The other way that we accompany sliders is post build. After we build those 1500 lineups for your pool, we then we’ll rate each lineup based on, we’ll just have their Projected score, which is just the sum of the projections for the players. And then we’ll have the Saber score, which accounts for these three sliders and the Saber score and Project score are used to sort and filter your full lineup pool to determine which of your 1500 lineups is actually going to be in your selected lineup set. That’s kind of the background into how the Saber score works.
For Late Swap, because we don’t build the whole lineup pool, Saber score and Project score don’t really matter. So it’s actually, the answer is sort of neither because we’re not doing any sorting of a lineup pool with Late Swap. All that matters is how they’re being built. So we are building the lineups based on ownership and correlation and based on those slider settings. So in a way, I guess it’s closer to Saber score because we do account for those settings, but we’re not sorting them in any different way. We’re literally just taking each lineup and rebuilding it based on the settings that you provide and the players that are lots.
Andy Baldacci:
And just to reiterate too, if you want control of that, don’t want any of those factors to play a role. You can just turn on cash mode or just manually put all of the sliders to off like that. And then it’ll just use projections from there.
How do I get Late Swap to match the exposures I’m looking for?
Matt Hunter:
So for me, one thing you can do is just focus more on projection than on the exposure. Sometimes it’s easier if you’re just getting way too much of a player, more than you’re comfortable with, it might be easier to look at the projection and see, “Why am I getting so much of this player? Do I need to lower the projection to help with that?”
The other thing is that sometimes… If there aren’t that many games to swap, it can be hard to get the exact exposures that you want because some of your lineups just aren’t very swappable, or it’s hard to get those exact exposures. One thing you can do is make your settings less restrictive so that when you do your Late Swap, maybe lower your main salary a bit. Maybe you had set it to $49,500. And so we’re not able to swap some of your lineups to get within the exposures that you want because there’s not enough salary. You can lower your mean salary, maybe lower your mean projection, try to get a bigger pool of players to pull from. That’s something that I would try. Messing with the player projections and see if you can maybe get a little bit more or less of a player that way. I don’t know if Max and Danny, you have any thoughts?
Danny Steinberg:
Yeah, I think the thing is if a game has already started a player is going to be locked, so we’re working within restrictions and sometimes you can’t get exposure as you want.
If everyone’s game has started in a lineup, we’re not going to be able to Late Swap your favorite player into that lineup because there’s literally nowhere to put them if there’s only like a few places. If you really… Like a center that is benefited from some injury and you already have all your centers where you played their game started, we’re not going to be able to get exposure to that person or get a lot. There’s just a lot of constraints. So it’s going to be hard to get exposure exactly right in a Late Swap, for sure. And I think you have to worry about that.
Matt Hunter:
Especially on FanDuel where there’s not the multi-position eligibility, so it’s a lot harder to fit in exactly what you want.
Andy Baldacci:
Right. And one thing that is important to point out is that if you do find it important in your build process to have a lot of rules and really control exactly how the lineups are built, that’s fine. But you have to understand that as the remaining options become fewer and fewer is less to do, but what makes SaberSim’s Late Swappers so different than the other tools is that with traditional Late Swap tools, they say you have 150 lineups, and they’re going to try to swap all 150 following the rules that you put in. If they can’t, you are not going to get any line of spec, it’ll say, “We couldn’t do it.” Whereas what we’ll do is we will rebuild and Late Swap as many of the lineups as we can while honoring your rules and your salary ranges and everything else.
And then when we hit a wall and there’s no more left for us to build, following that, we’ll throw up a message that says, “Hey, we were able to rebuild 70 out of those, 150. Do you want us to, for this last 80, rebuild your lineups without those restrictions?” And so what that will do is let you get as close as mathematically possible to the rules you had in place before throwing them out. And so it’s not an all or nothing proposition like it is with other tools. And that’s one of the big things we really do want to stress with this is that if you want that extra control, we’ll let you get as much of that as you can. But when Matt just says, “That’s it,” then we’ll build the rest in whatever way possible.
Andy Baldacci:
Hey everyone, how’s it going? My name’s Andy Baldacci, I’m the CEO of SaberSim and today, I have Max Steinberg, DFS professional, and one of the partners at SaberSim here with me. How’s it going, Max?
Max Steinberg:
It’s going great, very excited to shoot this video.
Andy Baldacci:
Yes. So this is a fun one. We’re kind of going in a bit of a different direction than we have in some of the other videos and talking about something we really haven’t covered much, but that we think is a strong part of the DFS process. And that is choosing your core group of players, but the way a lot of other touts, a lot of other people really talk about this, frankly, is wrong and can be misleading and can actually hurt your DFS results and waste a lot of time in the process. So what we’re going to do today is just cover the right and wrong way to choose your core plays for NFL. And Max, why don’t we just jump right into this one and just talk about choosing the core plays, how you’re doing it, how you shouldn’t do it, just how you think about this in general.
Max Steinberg:
Yeah. So I think this is what quite a few, I think DFS players, whether they’re using our product or another product will choose core or what they call core plays for one reason or another. And it’s something that I will actually do sometimes, but I think there’s a right way to do it and a wrong way to do it. And it has to do with how the players interact with each other, how excluding one play might affect another play. And I think what’s really good about SaberSim is, as we say, everything we do is built on our simulator, right? So we have a simulator. We simulate every game, thousands of times. And what we get is a lot of great data and that great data helps us build your lineups for you. Is when you go to the bill process, if you look under the hood, we have these correlations smart diversity in our ownership fade settings, and all these settings are based on the actual data we got from the simulations and those help us build you really great lineups as a mathematical process.
Max Steinberg:
I’m not going to say anything that you don’t know already. Everyone knows that the quarterback has correlation with his receivers. Patrick Mahomes has correlation with Tyreek Hill with Travis Kelce, with Mecole Hardman, all his receivers. And also something that’s common knowledge is players have correlations with opposing players. Patrick Mahomes is going to have some correlation with some receivers on the other team or if you look at, let’s say DeAndre Hopkins, this is the concept of game stacking, is you choose a wide receiver on one team, and then you may choose a wide receiver on another team, Tyler Lockett. There’s some correlation there.
Max Steinberg:
And one overlooked one as well is maybe running back in defense, we’d use Kareem Hunts. We’re probably going to see that he’s correlated with defense of the grounds, which he actually has a pretty strong correlation and that’s because running backs as a team gets really up in a game, they’re just going to run a lot. They’re going to pound the rock and that’s going to be good for a player like Grandmont. And so the reason that this is important when if you’re actually pruning some plays, if you’re actually choosing core players, is you want to choose core plays that are conscious of how they might affect other situations.
Max Steinberg:
For example, let’s say you were attempting to say, “I really don’t like Kyle Allen,” or something like that or, “I really don’t like Logan Thomas,” that’s a lot of targets. I think he’s washed, whatever. I don’t want to use them. Well, the issue with that is he has some correlation with some players on Dallas. And if you use Dallas players, it might be advantageous for you to use Logan Thomas. And so even if you have some take like that, you want to make sure that you’re not just getting rid of Logan Thomas, even though in some of your Dallas lineups, he’s going to be the optimal play.
Andy Baldacci:
Right because to get upside in lineups and just even putting upside aside, I guess players don’t exist in a vacuum. This is not something where you can just think about the impact they have on a lineup individually because correlations are really what’s driving the upside and how you’re going to stand apart in a lot of GVP. And while you might not like one player, you have to think about kind of the ripple effects that can come from that is removing him is going to remove other players accordingly and maybe those are ones that you wanted more of. And so that’s where just kind of looking at those correlations and making sure not to take too much of a narrow view of an individual player really comes into play.
Max Steinberg:
Right. I actually want to talk about how you should not choose core place first and so, I actually already did a build before this video. And I think if you’ve used our product before, then you’re going to know what I’m talking about here because we let you see the team stacks and the game stacks that you get and most of the time in the lineup seat belt, you’re going to have some sort of game stack in basically every lineup. In this pool of 150 lineups that I’ve created, I had just a rough estimate, it seems like 75% of lineups have game stacks. If you look at the actual lineups, we have this Houston onslaught and we have Davante Adam’s coming back. If we go down further, it looks like we actually really liked Deshaun Watson. So if you have Matt Ryan, oh wait, suddenly we don’t have anyone from this Green Bay, Houston game.
Max Steinberg:
And so what I think is really cool about the quarterback position is if you get diversity through the quarterback position, you end up creating optimal lineups that have wildly different players. If we have Matt Ryan, this lineup is going to look completely different than a lineup with Deshaun Watson and it’s going to look completely different than a lineup with Josh Allen. And so if you end up pruning some of these players at quarterback or wide receiver at tight end, I think what ends up happening is you end up missing out on some of those really what I call good diversity. It’s diversity that you’ve got a diversity in a wide array of players while keeping upside and correlation, all that good stuff you want and the GPP lineup. So in general, I think that when you’re looking at core players, I don’t even touch wide receiver, tight end quarterback or flax. I think that usually it’s really bad idea because it’s really going to be the best way to get that diversity.
Andy Baldacci:
Yeah. And one of the thing to just add is that when we talk about correlations, it cuts both ways where the more correlations you stack upon each other, the higher upside you have, but it just also adds a lot of risk because while the performance of one player will positively impact another, if one of them does poorly, it means the others do as well. So it makes your lineups more boom or bust.
Andy Baldacci:
And so I feel like just naturally, this is something that I don’t think not people talk about, but when it comes to diversity is that people might say, “Oh, I set my cap for all positions at 70% exposure,” or whatever it may be, but not all positions are the same. They don’t all have the same variants, especially when you start building out those stacks. And it sounds like a lot of what you’re getting at is that one you’re able to get diversity here. But when you talk about the good diversity, it’s diversity where you’re not sacrificing much and to take it, the other side of it, if you really focus on that, it might not be like a bad EV decision, but your swings dramatically increase and you’re not really gaining anything. If you just have a few core stacks and in all of them, they’re all built around that, that could be fine, but your swing is going to be massive and it’s not necessarily better than building out the wider array of stacks. Is that in line with what you’re thinking?
Max Steinberg:
Yeah, I think every week I’m thinking about how do I mitigate my risk as much as possible without sacrificing the EV of my last because I don’t want to run the risk. You don’t want to make a diverse lineup that’s going to be an unprofitable lineup. And so, figuring out how to get that diversity without risking having a negative expectation lineup is going to be really valuable. I think the best way to do that is through the quarterback position through these game stacks because it’s just naturally going to happen in the SaberSim built here and there’s a reason for it.
Andy Baldacci:
Can you walk through a little bit of how you would actually do this? Cause it sounds like what you’re saying is that the positions for football you’re focusing on for choosing those core plays are for the running back and the defensive specialties specifically, can you just kind of walk through how you would navigate narrowing these down?
Max Steinberg:
Yeah. You just got what I was getting at is running back, I think is a really interesting position because a lot of times there’s going to be entries throughout the week where you’re going to get the desk value place. these are going to be a lot of times the highest projected players on the slate. They’re going to be the best value players on the slate. And for me, sometimes I really want to take a stand on some of these running back players is going to be a play that I really like, and I haven’t done much research for this week so I don’t have that play for you in week seven. But usually, there’s going to be three or four players that I really, really like and I want to make sure that I get good chunks of pieces of them in my lineups.
Max Steinberg:
And so in that case, I might actually start by excluding some players and just checking the boxes of some of these running back players that I really, I usually start by just actually sorting by value. So I make sure I just like looking at the best value players and not missing anything or overlooking anything, not just deciding beforehand. And then I’m just going to sort of check the boxes of some players and maybe it’s a projection thing, maybe it’s something else, but for example, maybe I think Houston is going to get killed by Green Bay. And so despite the fact that the projection, the average projection for David Johnson is good and that SaberSim is going to quantify some games where Houston is going to do well against Green Bay, I have some take that I think Houston’s not going to do it all.
Max Steinberg:
If Houston doesn’t do well, Duke Johnson, who’s more of the third down back is going to get a lot of play. So maybe instead of having David Johnson in 10% of my lineups, I am going to just want to make sure that I get more of my core players to guys that I actually want to get leverage on because some of these guys, I might want to get 60 or 70% of the time in my lineups. And the reason I’m okay with that is running back as a lower and variance play, you can have more confidence they’re going to do well, because for most of these guys, they’re going to get 15 to 20 rushes. They’re going to get three or four targets at the very least. And so they’re never going to have those like zero games that you might see from a wide receiver.
Max Steinberg:
So I think there’s some value in choosing some of these running backs as core plays. I’m not giving you specific bets but I’m sure as you’re watching this video, you’re looking at this being like, “Yeah, I really like Aaron Jones. I really like Antonio Gibson.” Okay, well you can exclude some players and make sure that you get these guys more in your player pool for the post-build process.
Andy Baldacci:
You’ve mentioned before the correlation between running back and D, how should that impact the decision to kind of prune the pool a bit? Is that something you’re thinking about?
Max Steinberg:
Yeah, absolutely. So I think these are going to be related to each other because the running back almost always has correlation with their defense. Mike Davis have been a pretty strong pass catcher, pass correlation with the Panthers’ defense. So I think you want to keep in mind in both ways, you say, okay, “What running backs do I like?” You look at defense, you say, “Okay, what defenses do I like?” And then you want to actually go back and say, oh okay, so Washington’s football team offense is really cheap. Well, maybe I didn’t like Antonio Gibson, but obviously if Washing’s defense does well, I want to have Antonio Gibson in my lineups. He is sort of the lead back and he is going to be a good player. So maybe I didn’t like Antonia Gibson, but suddenly I think it would be a massive mistake if you didn’t include him because the Washington D is cheap and I liked the Wash team defense.
Max Steinberg:
And so you want to go through these like this and say like Browns’ defense. You say, “I don’t really like the Browns’ defense. I think they’re overrated.” Well, if you were like Kareem Hunt, then I don’t think that’s a smart move to exclude the Browns’ defense because they’re going to be related. Even if you think, “Oh, Joe burrow, people think he’s overrated or underrated in terms of, if he doesn’t actually have that many turn on for worthy plays,” then you say, “I don’t like the Browns’ D.” Well, then you have to think about, “Well, then do I want to exclude Kareem Hunt too?” And so you really need to think about both of these plays before just taking a stand on one of them because they’re related to each other. So that’s something that I’m going to be going back and forth with when I’m making these core plays and looking at who to play.
Andy Baldacci:
And so let’s go through and just do a quick build and just see what happens. And then I want to hear your take on just what to expect, like what impact does this have in that last step of the build process where you’re doing your quality control and just getting exposures dialed in?
Max Steinberg:
Well, let’s make sure for a second that I actually practice what I preach. So we have New Orleans, I’m using Alvin Kamara. So in New Orleans’ D, Jerick McKinnon definitely want to use them. Let’s see Green Bay D. I think they’re viable. They’re not a very good defense, but Deshaun Watson does turn it over. So they could have a big game. I’m just more worried about what GPP is. Is there a way that this defense has a good game and I think [inaudible 00:14:24]. So maybe I missed a thing or two here, but we’ll just go with it for now and do a quick belt.
Max Steinberg:
And yeah, I think what you’re going to notice actually is we’re probably going to have the builder hone in on some defense running back stacks. We’re going to get a lot of a defense. My guess is it’s going to be Washington and we’re going to get more Antonio Gibson than we probably think because he has that correlation, even when we don’t like him, the SaberSim builder is actually giving you good feedback. But they’re saying if this defense is such good value and Antonio Gibson is such good value, you don’t really have to love either of them that much. What is going to happen is if the defense does well, Antonio Gibson’s going to do well. And that’s going to open up a lot of value for you for the rest of your lineup, which is going to be really, really good.
Max Steinberg:
So there we go. We have Washington D we have Browns’ D. And if you actually look at these lineups, we have Kareem Hunt and we have the Browns’ D. We have Antonio Gibson and we have Washing defense. These are our top lineups. So if we excluded Antonio Gibson, what SaberSim is telling us is we would have excluded actually like one of the top lineups that we could have built. And so, I think it’s really important to consider both running back and defense together when you’re picking this core players, because you might actually be sacrificing more than you intuitively think.
Max Steinberg:
And I just want to talk about the last thing because thinking about running back and why we’re doing these core players is I think it’s all about leverage, is you see our projected ons for some of these players and throughout the week, I think as the slate shapes up, these are going to change and you’re going to see some players who are projected at 20, 25, 30% ownership. With that, if we like a player a lot and we want to have him and be a positive EV play for us, that means we have to get him either in more optimal lineups than the field, which we’re going to do at SaberSim. But we also want to get him more than the field and so let’s say I really like Kareem Hunt. Well, his ownership by the end of the week might be projected at 20%.
Max Steinberg:
If I really, really like him and really wanted to take a stand on him, if I limit my running back player pool, it’s really easy to just raise as many exposure and suddenly get him 70% and have a lot of leverage on him. Maybe I really like Antonio Gibson, it’s really easy to raise him an exposure and get 40% Antonio Gibson. If we didn’t limit our player pool, it’s possible that we might have to build more lineups in order to do that. But with this, we’ve built a great pool of lineups that allows us to do what we want to do in the post-build process very, very easily. So I think in general, the reason that I’m picking these core running back players to get that leverage and it’s a really easy in the post-build process to get it while building some really good lineups.
Andy Baldacci:
Awesome. And we covered a lot today on this specific concept, but if you want to dig into any of our other videos or strategy guides on football strategy in general, we have a ton of content @sabersim.com and also on our YouTube channel, so be sure to check that out. And if people have questions, Max, what’s the best way for them to ping you or reach out?
Max Steinberg:
You can find me on the SaberSim slack channel, or you can tweet me @MaxJSteinberg and happy to answer any questions.
Andy Baldacci:
Awesome. And if you guys want to try this out for yourself, if you want to play around with SaberSim and see how it works and really dig into some of these concepts today, we have a free three-day trial, which you can check out at SaberSim.com. If you don’t like it, cancel your account. You will not get charged a single cent. So you really have nothing to lose there. Go in, check it out, play around with it and see if it’s right for you. But I appreciate the time. Max, it was great chatting today and I’ll see you guys later. Good luck.
Andy Baldacci:
Hey, what’s going on everyone? Thank you so much for joining us as we walk through the six secrets of daily fantasy football. My name is Andy Baldacci. I’m the CEO of SaberSim and I’m joined by DFS pro and SaberSim partner, Max Steinberg. How’s it going, Max?
Max Steinberg:
It’s going great, Andy. Excited for the football season. Let’s get into this.
Andy Baldacci:
This has been a long time coming as well. We weren’t sure if it would be coming, but it is here and we are very excited about it. And so this video is kind of a highlight reel that we’re putting together from our much more thorough guide that was called How to Beat Daily Fantasy Football in 2020. And in that video, we broke down the fundamentals of daily fantasy football and really just walked step by step through the build process from researching to actually putting all that together and building those lineups.
Andy Baldacci:
And if you want a deep dive, go to YouTube, go to SaberSim.com, wherever it is, go check out that guide and you will have everything you need to know to really hit the ground running. But for today in this video, we’re trying to keep things a little bit higher level to cut right to the chase and just give you the six secrets that we found to beating daily fantasy football. So watch this one first, if you haven’t watched either of them. And then if you’re ready for that deeper dive, head over to our site, head over to YouTube and watch the full one. But right now let’s just jump in.
Andy Baldacci:
So the first secret is find the upside. And really if there’s only one secret today, it would be this because to win at GPPs, you absolutely need to build high upside lineups. And by high upside, what we’re really talking about is just finding ways to create leverage against the field so you can take down these huge top heavy tournaments. And when it comes to upside, there are three components that we think are the most important. It’s correlations, ownership and variance. Correlations are just the relationship between two players’ performances. So when two players are positively correlated, that means that when one of them does well, it’s very likely that the other will also do well. And the opposite is true for negative correlations. Ownership just means the percent of lineups in a contest that have a specific player. And the reason this is so important is because, just as an example, say everyone in a tournament, literally 100% of people, have the same player in their lineup. And if that’s the case, it literally does not matter at all what that player does because everyone gains or loses the same amount.
Andy Baldacci:
And so what you’re doing is you’re looking for the right spots to do what’s called fading the chalk, or just avoiding highly owned players. And you’re also looking for under the radar players who can have a big game and immediately separate you from the field because very few other people have those players.
Andy Baldacci:
And then when it comes to variance, we’re just talking about how a player’s performance can vary in an individual game, because projections are going to give you an average, but performances in sports are anything but average. And you need to find the players who have a higher ceiling than others who may have similar projections. And if you want to deeper dive into how to apply those specific examples, check out that other video I mentioned, How to Beat Daily Fantasy Football in 2020. But ultimately what matters most is finding smart ways to get leverage on the field and give your lineups a real shot at taking down these big tournaments.
Andy Baldacci:
The second secret is perfecting your projections. And kind of working with a lot of players, talking to our customers, talking to just people in the space, the most common mistake that I see from experienced players is having almost an obsession with exposures. And by that, I just mean how frequently a player is appearing in their own lineups. And it’s not that exposures don’t matter, but by focusing on exposures above all else, you’re really putting the cart before the horse, because ultimately everything comes down to point projections and upside. And without considering all of that, without considering correlations and variance and ownership and the actual projections themselves, you can’t know what is too much or too little of a player. So exposures matter, all those things matter, but they all come from the projections, they all come from kind of those original inputs. And that’s why it’s important to focus on perfecting those.
Andy Baldacci:
And another thing that a lot of newer players ask is should I try to create my own projections? And honestly, if you’re not doing this professionally, if you don’t have a data science background, if you don’t have a ton of time, I’d say no. There are a lot of strong projection systems for both points and ownership and SaberSim is one of those. But unless you have a ton of free time, I wouldn’t worry about creating them from scratch. I would find a system that you like that that works for you and use that as a baseline to then get it dialed in because no model is ultimately ever going to be perfect. It is incredibly challenging to properly balance historical data with all the unique considerations that can come up each week. So while a model can be strong on average, there are still going to be things that it will miss, even the best model is going to miss.
Andy Baldacci:
And so the highest leverage thing you can do is start with a strong projection system, but to get that edge, use your own research and really get it dialed in. The more time you have available, the more research you can do, but even if it’s just 30 minutes, you can still add a ton of value to the process by focusing your research on the outliers, the ones that are jumping out, the ones that rather than trying to go through the entire list of players, it’s finding those spots that really stick out. And we’ll talk to you about a way to do this pretty quickly in a little bit. But what you’re, again, looking for are those players, those teams that really stick out and focusing your research on that.
Andy Baldacci:
And in these next two secrets that Max is going to cover, we’re going to give you some practical tips on what to look for in your research so you can actually make good adjustments because that’s another thing we’ve run into a lot is, all right, where do I even begin with us? So the third secret is pay attention to the news. So Max, when it comes to the news, when it comes to just kind of staying on top of things, what are you looking for? And why are these the kinds of things that are really difficult, if not impossible, for models to automatically adapt to?
Max Steinberg:
Yeah, I think this is going to be the number one way that you can add value because there’s a lot of data out there that you could misapply or apply doubly, something that we already do already. But things that have to do with what coaches say, something to do with subtle injury stuff like a player is playing, but he might be a little, snap counts, things like that. Those are things that if you can find good information on, that’s going to add a lot of value to our projection systems or a projection system that you’re using. And so when I’m perfecting my projections for my fantasy lineups, I’m really, really trying to be in tune with the news. And there’s a lot of easy ways to be in tune with it is there’s some sites that sort of just have a newsfeed.
Max Steinberg:
And Rotoworld is a really, really good one. If you go to their site and just go to their player news page, you’ll just see a feed of a bunch of tweets, a bunch of articles and basically excerpts that have really good information. You can also have good Twitter follows. One person I follow religiously is Adam Levitan. He’ll have a lot of tweets throughout the week that are coach’s quotes, just general injury stuff as you see here. And there’s just a lot of people you can follow. I’m sure you have some as well where they’re just going to give you really good bite-sized knowledge that’s not going to be quantified in any data. It’s going to be stuff that you can interpret and change projections for.
Max Steinberg:
And so one thing that’s standing out to me in week one, which I just saw today, has to do with the Colts’ running back situation. So the Colts have a really good offensive line. They also have three running backs. They have a rookie draft pick who’s a high draft pick in Jonathan Taylor. They have Marlon Mack who was good last season. And they have this guy named Nyheim Hines. So they have three running backs. And we’re trying to figure out, it’s week one of the season, who’s going to get playing time. How much playing time is Jonathan Taylor going to have? A lot of people expect him to have a lot. Well, not so fast. If you read this article in The Athletic, or just this excerpt from this article, this beat writer is actually saying that he feels like Jonathan Taylor is not going to play as much as people think, that he’s just going to spell Marlon Mack mid-game and keep him fresh in the fourth quarter. That’s a lot different than we perceived early in the season.
Max Steinberg:
If that’s true, the Colts have a just amazing matchup versus Jacksonville. And that means that maybe, okay, Marlon Mack, I can be higher on him. Maybe you can even be a lot higher on him. And so processing this information and just keeping up on the news and finding more things like this, can be a really, really easy way to improve your projections and improve your lineups.
Andy Baldacci:
And then the next secret kind of builds on this and that is taking advantage of all the data that’s out there. And while we have your screen up, can you just kind of walk through some of the sites that you go to get more kind of quantifiable data, things that are still useful to look at that maybe the model isn’t processing or whatever it may be, but it’s still a useful to get in your own way.
Max Steinberg:
Yeah. I mean, I think there’s a lot of free sites out here. One of the ones that I go to every week is airyards.com, which is created by this guy, Josh Hermsmeyer. And there’s some really great data that has to do with how far someone’s targeted down the field that uses the NFL play-by-play data and gives us some good stats. One is average steps to target, and that gives us something called air yards and Josh actually found that air yards are a really predictive metric for predicting how well a wide receiver is going to do. So if you find something here that has some really good stat that stands out to you, that differs from maybe the projections you’re seeing at SaberSim, or the projections you’re using, you’d say, “Huh, okay. Maybe this player is worthy of a boost. Maybe this player’s worthy of negativing a little bit.”
Max Steinberg:
And I like just going through all the teams and seeing who stands out. I think you can get a lot of really good information here. And it’s really, really helpful. Obviously we’re just looking at 2019 data, but as 2020 goes on, there’s going to be more and more data. And this is a great resource to look at.
Max Steinberg:
Other things, there’s a lot of good people on Twitter. I really liked this guy, Kevin Cole. He’s a data scientist at Pro Football Focus. A lot of his articles are behind a pay wall, but a lot of times you’ll also tweet really good stuff or sort of tweet graphs and things like that that’ll help you with different contests and are going to be really useful information. [Derek Hardy 00:00:10:42] is another person. He has his own projection system. He also tweets out really, really good content about why he thinks a certain way about a certain person, why maybe this person is getting too much buzz. There’s a lot of good resources out there. Obviously there’s a lot of more sites than just airyards.com. And if you look, you’re going to find them and it can also be just a really easy way to improve your projections.
Max Steinberg:
Okay. So secret five, get your diversity through the quarterback. So the quarterback is kind of like the hub of correlations. He has a lot of correlations to a lot of different players, not just players on his own team. He has correlation to the wide receiver, tight end, running back on his team most of the time. But he also has correlations with players on the opposite team. Sometimes you will have pretty strong correlations with wide receiver, the quarterback who you’re playing showdown. Changing quarterbacks in the lineup will sort of just change the optimal construction of the lineup you’re using. And so when you’re trying to create 20 or more lineups, maybe 100 or 150, honing in on your exposure to the quarterback and changing how many lineups you have at X quarterback is going to be the best way to get diversity of your lineups, but not sacrifice that much expected value because if you change that quarterback, the optimal lineup is just going to be very, very different.
Andy Baldacci:
Then the last secret is just use the right tools. And when it comes down to it, if you can build lineups using those last five secrets that we just talked about, you’re going to be a winning player, plain and simple. But if you’re not using the right tools, it is going to take you all weekend to do this. And even then, you’re still going to feel like you’re guessing a lot of the time, because between your job, your family and just everything else going on in your life, you probably can’t set aside your entire weekend for DFS. I know I can’t, and I actually run a DFS company. So for people with more normal jobs, I can only imagine how difficult it is. But if you want to beat daily fantasy football and can’t spend hours and hours building lineups, then you need to use the right tool.
Andy Baldacci:
And SaberSim is the only lineup optimizer out there that understands the upside, which means that you can build winning lineups fast. And a lot of times when I’m talking to players, they feel that tweaking the projections and really trying to perfect those things is kind of a waste of their time because they only have so much time to dedicate to DFS. But they already are spending hours getting the rules and getting the groups and getting everything else dialed in because they have to fight with their optimizer to get good lineups. But if you’re using a tool that can kind of automate away a lot of that busy work, you’re then able to add even more value and get even more of an edge on the field by dialing in those projections that may already be strong, but can only get stronger by adding some research to it.
Andy Baldacci:
And so let me just kind of quickly show you how SaberSim works and how it can help you build better lineups in much less time than before. Working with some of the top names in DFS like Max here, as well as others, like Giantsquid, we have built and refined a one of a kind simulator that takes dozens of performance predictors and simulates every single game, play-by-play thousands of times. And it’s this unique process that gives us the data, that automates away all that busy work that traditional DFS tools bog you down with. And this lets you spend your limited time where it matters most. And so we talked about perfecting projections, but when you’re going in and just looking at this big list of players, it’s hard to know where to even start. And so one thing that we’ve done to make this easier to narrow it down is by promoting what we call a test build.
Andy Baldacci:
And so what you’re trying to do here is you’re just seeing who we’re naturally high or low on so you don’t have to research an entire slate. And we’re not going to make any adjustments. All we’re going to do is set this so that we are building for the play action on DraftKings, it’s a 20 max contest with 200,000 entries. And so I’ve put this in here, haven’t made any other adjustments, but what SaberSim is doing behind the scenes is adjusting all of our defaults to be best suited to a contest like this. And all we’re going to do is hit build.
Andy Baldacci:
And if you were to do a build like this with literally just clicking a couple buttons in a traditional optimizer without making rules, groups, whatever else and just spending a ton of time dialing things in, you would get horrible lineups because those tools simply don’t understand upside. And so without programming all of that in there, you’re not going to get lineups that have any upside in it. But because we’re simulating all those games thousands of times, we have the data that lets us understand upside and give you strong lineups right out of the box. And the more time you have available, the more players and teams you should be looking at, but regardless of the amount of time that you have, this is going to make sure that you’re able to spend it where it has the biggest impact.
Andy Baldacci:
And so what we can see here, this overall on the left is a list of all our players. We can sort through them by position, by team and everything else. And we can see what our exposure is to each of those. And so, Max, if you had a limited amount of time, say you just had 30 minutes to build your lineups, who would you be looking at? Where would you focus your research?
Max Steinberg:
Yeah, so I think where I’d focus is I’d look through the players that were getting and look at some of the value plays because missed projections or accurate projections with these types of value players are going to add the most value because little differences can make a big difference in how many players you got. So one person that stood out to me at tight end was Irv Smith. He’s 3,100, we’re getting him a lot. I might just look through the different positions and see if you see 55% exposure. I might look at running back, see the value players there and just sort of look through and keep an eye on that.
Max Steinberg:
And also maybe play the comparison game a little bit. I might sort by projection and then see, okay, let’s see, is Alvin Kamara and Austin Ekeler, do I think they’re similar or do I not think they’re similar? And look through the players and be like, “Should there be a bigger difference between these players?” And then maybe also look at defense because if you look at defense you’ll notice, and I think this is actually proper strategy, that SaberSim’s lineup builder actually seems to favor some of these low price defense, the Dolphins, Washington, the Cardinals, the Jets. And that’s because that probably is going to be the optimal thing is saving money at that defense. And so getting those projections honed in and finding those low price defenses that are really good are going to be key to making good lineups as well. So those are a couple of things that I’d look at right off the bat, I think.
Andy Baldacci:
Yeah. And this is, again, where when you use the right tool, when you use SaberSim, you can kind of get an advisor here. You can almost get your kind of personalized podcast where you can tell us what type of contest you’re building for. And then we’ll say, “All right, based on everything we know, here’s who we are kind of focused on.” And it’s, again, not to take that as the 100% truth, but it’s to say, okay, this just cut down everything I have to look at significantly by just showing, all right, who are the ones that stand out to me? Who should I look into? And again, the more time you have available, the more you should kind of widen the net of who you’re digging into. And if you want to see kind of how Max would do this in much more detail and how we adjust projections based on the research, check out that video I mentioned before, How to Beat Daily Fantasy Football in 2020.
Andy Baldacci:
But for now we just wanted to give you a quick overview so that you can get started. And these secrets, they work with or without SaberSim, but I really hope that at this point, it’s obvious just how much easier it is when you use the only tool that’s built for the job and that’s SaberSim. And if you want to beat daily fantasy sports and can’t spend all day… I just got tossed [inaudible 00:18:18].
Max Steinberg:
Okay.
Andy Baldacci:
I’ll just do the last part.
Max Steinberg:
Okay. I can do the last part if you want.
Andy Baldacci:
No, no. I mean, I’ll just… two sentences.
Max Steinberg:
Okay. Got you. Okay.
Andy Baldacci:
These secrets work with or without SaberSim, but I hope it’s obvious just how much easier it is when you use the only tool built for the job. If you want to beat daily fantasy football, but you can’t spend all day building lineups, then you need SaberSim.
Max Steinberg:
Yeah. And it’s going to… SaberSim saves you a ton of time and it allows you to work on the fun stuff. Do your research, hone in your projections. It makes the process of making lineups a lot more enjoyable. And the best part is you can actually try it out completely free for three days to see if it’s right for you. And all you have to do is head over to SaberSim.com, start your free trial, and you’ll be building winning lineups in no time. So if you have any questions along the way, you can reach out to our support team at [email protected] or email Andy directly at [email protected]. Or you can reach out to me on Twitter. I’m @MaxJSteinberg on Twitter. And thank you so much for tuning in and good luck this NFL season. Thanks.
Andy Baldacci:
Hey, how’s it going, everyone. Thank you so much for tuning in. Today, we are going to share the six secrets to beating daily fantasy basketball. NBA DFS is one of our favorite sports at SaberSim, and there is a lot of money to be made if you use the right strategy. And this is a video where we’re going to do our best to share what those strategies and tips are. My name is Andy Baldacci, I’m the CEO of SaberSim and I’m joined by DFS professional and SaberSim partner, Max Steinberg. How’s it going, Max?
Max Steinberg:
It’s going well. Excited to just jump right into this and talk about some DFS NBA, one of my favorite sports.
Andy Baldacci:
For sure. So, yeah, let’s just jump right into this and start with the first secret, which is staying on top of the news. And it might not be the best kept secret, but it is super important and critical to get this right.
Max Steinberg:
Yeah. So, NBA, there’s going to be a lot of news every day. There’s always going to be some sort of injury, some sort of late news. And I think this is going to be no different in the times of COVID. It’s probably going to be even more so. Players just being rolled out in the last minutes or hours coming up to lock. So, following the people and accounts on Twitter that are going to help you keep you up to date and up to the minute really, it’s going to be really important. I put alerts on my phone for specific accounts that have updates like FantasyLabs and following beat writers, or just following people who just have their eyes on things, it’s going to be really, really important.
Max Steinberg:
So, make sure that you do your research and find some of those people or follow SaberSimor be on our Slack channel as well. That’s going to help you get updates. And many of the sites, including DraftKings and FanDuel have a late swap function. So, if something comes up after lock, you want to make sure to look at projection updates as well as making your own judgements and to see if maybe your lineups are actually worth changing after lock, because sometimes a lot of good value can open up later.
Andy Baldacci:
For sure. The second secret is examine injuries. What did you mean by this one, Max?
Max Steinberg:
Yeah. So, I said this and I’ll repeat this again. Every day in DFS NBA, there’s going to be some sort of injury news that’s going to have a huge impact on what players minutes and their production. And this year with COVID, that’s not going to be an exception. It’s going to be even crazier. So, the places you were going to add the most value in constructing your lineups and to our projection model is going to be evaluating situations where there are big injury impacts with the teams. So, while we really recommend focusing on those areas where, okay, this player is out, let’s look at the projections. Oh, I like this player because X, Y, or Z. That’s going to be really important to actually adding value to the SaberSim process.
Andy Baldacci:
And that ties into this next secret, which is breaking down the components. Can you talk about that one, Max?
Max Steinberg:
Yeah. I think with NBA, more importantly than other sports, looking at the components of the player’s projection, this means minutes, points, rebounds, et cetera, and I think specifically minutes. That’s going to be the easiest place to add value, because if you’re going to adjust a player’s projections, and you want to know, “Why am I doing this?” And minutes, a lot of the times are the place where it’s sort of hardest to quantify. You have to say, “Okay, this guy’s…” There’s only so many minutes that a team can play in the game and you have to say, “Okay, which guy is going to get one more minute and which guy’s going to get one less minute.” And that’s really hard to do. And it’s something where projections, if they’re going to lack in anything are going to lack a little bit in that. So, looking at the depth of the team, seeing, okay, how many active players are there? That’s going to be really important.
Max Steinberg:
You’ll see some nights where a team might not actually have 10 active players. They could have nine or eight. These are the situations where there are lots of minutes available. And that’s maybe a place where you’re going to be able to project out some player’s minutes, because they can have those spike minute games.
Andy Baldacci:
Mm-hmm (affirmative).
Max Steinberg:
Another place is foul trouble. A lot of times due to entry, there’s going to be players who are going to have high projected minutes that they’re not used to having. And if these players are high fouls per 36 minute players, which is something you can see on a site like Basketball-Reference, these are players that could have their minutes cut short from foul trouble. And those are players that you want to keep an eye on and make sure that we’re not projecting their minutes too high, because they could get cut short real easily.
Andy Baldacci:
And for that one, Max, actually, I just wanted to stop on that quickly.
Max Steinberg:
Sure.
Andy Baldacci:
But it’s not even so much saying that the average is wrong for how many minutes they’re going to get, it’s that it’s just going to be pretty high variance and there’s going to be a lot of outcomes where they get pulled early if they do get into foul trouble and there’s just going to be a lot of variance in the number of minutes that they’re playing.
Max Steinberg:
Right. That’s a really good point. It’s more the probability that this player might have their minutes cut short becomes a lot higher, and that’s going to be really key for GPPs, because if you have a player that’s going to be really popular, but 30% of the time he gets into foul trouble, then that’s going to really kill your lineups and going to be an important factor in deciding, “Okay, maybe I should fade this player.”
Andy Baldacci:
Mm-hmm (affirmative). And there was one other point on this that you had, which was around injury impact on a specific position.
Max Steinberg:
Yeah. I mean, this is a key factor that I’ve been looking at for years. And I think it’s really valuable is when there’s injury impact specifically to the point guard position, because a lot of time with teams, if there’s injuries, they’ll sort of find ways to fill the position. A power forward gets injured, they’ll play a small forward in a power forward. A center gets injured, they’ll play a power forward at center. When a point guard gets injured, a team likes to just use a point guard. They don’t try to fill it in with random players. If they have a point guard, they want to use this point guard. And they’re usually smaller players who can take a lot of minutes. And so, if you see a team that only has one point guard or just little point guard depth, and I’m going to show you sort of how to look at this a little bit later, that player is going to have to play a lot and that’s someone you can identify where you’re like, “Oh, maybe this player could have a really high spike minute game or really high minutes.”
Andy Baldacci:
And the next one is know when to fade and eat the chalk. And if you’re new to DFS, you might not be familiar with those terms. So chalk, it just means these are the most popular players on the slate. They’re going to have high ownership. Fading that means you’re going to be contrarian and not play those high on players. And then eating the chalk just means playing some of them as well.
Max Steinberg:
Yeah. And I think with NBA, there’s some times where you just know a player is going to be chalky. There’s going to be a slate where James Harden is projected for 55, 58 fantasy points. And there’s no one even close to him. And in those cases, he’s going to be over 50% owned. But in all likelihood, because NBA is such a predictable sport, he’s going to get probably 50 fantasy points. It’s very unlikely that he doesn’t get that. And so, getting tricky and trying to fade a player like him, if he’s the highest projected player on the slate, is usually going to be wrong. And you’ll see with the SaberSim builder, you’re just going to get that player a lot. And I think that’s correct. This is not like baseball where Mike Trout is going to go 0-4.
Max Steinberg:
This is not like football where a running back doesn’t score a touchdown and he just kills your entire lineup. James Harden is going to shoot his threes, he’s going to go to fifth. He’s going to get his points. However, I think there are good places to fade. And I think those are usually going to the players who are projected on the low side, but because of the lack of value options, are going to be popular. You’re going to have players who project at 24, 22 points, or close to mid salary that people are going to play a lot. And these are the guys that I think you want to consider fading, especially if you can find reasons that they might fail or their ownership might be out of control. Again, going back to the minutes thing with foul trouble, maybe they’re just not a very high production player and you think, “Oh, this person could just not get any fantasy points,” even though his average fantasy points are pretty good, there’s some low downside games there. So, those are some things you want to consider for sure.
Andy Baldacci:
And so, for that one, just to be clear, you’re kind of looking for situations where there may be some narrative that is driving what you think is going to be a high ownership for a player. But when you do some of the digging into the things that we’ve talked about already, you’re not so sure that there’s actually merit to the narrative that the public’s really all jumping on. Is that fair?
Max Steinberg:
Yeah. Or there could be merit. I mean, like I said, it is pretty predictable, so it’s not like you’re not going to see us project someone at 24 fantasy points and they really should be at 16 or something. That’s not going to happen. But I think, in general, if someone’s projected at 24, 16 isn’t the range of outcomes. And if you choose a player that just doesn’t get a lot of fantasy points, that can just kill your lineup, because there are going to be some value players that get 25, 30 fantasy points on a given slate. So, I think it’s more just taking into consideration, “Okay. This person would actually kill my lineup if they don’t get a lot of fantasy points. And are there avenues for that to happen?”
Andy Baldacci:
All right. And number five is optimizing for upside. And this is really about how to take all of this research that you’ve done, and the things you’ve looked into, and these other secrets to put it together and build proper lineups, build lineups that have a chance of winning. And Max, can you expand on that a bit?
Max Steinberg:
Yeah. So, just like any daily fantasy sport, they all have the same fundamentals and basketball is no exception. We want to have lineups that are correlated that take into account ownership and take into account those high range of outcomes that you want to win a big tournament. And so, with basketball, even though the correlations are not as strong, they definitely exist and they’re subtle. Sometimes they involve game stacking. Sometimes it involves point guards with big men or high assist players with players that get assisted field goals and things like that. And then the upside, players who can have those spike games. And luckily for you, SaberSim, using our simulation data and our simulator, we’re going to be able to help you quite a bit with that. And so, it’s something that you don’t need to think about right now or as much if you’re using your product, but it’s something that obviously is very important to winning a GPP in any daily fantasy sport and NBA is no exception.
Andy Baldacci:
Yeah. And that leads into this last one, which is using the right tools. And honestly, there’s just so much out there, even if you just focus on the secrets that Max has covered, there’s still a lot that you have to take in for every single slate. And ultimately, where I find players get led astray is they almost try to just like reinvent the wheel. They feel like a lot of this has to be harder than it actually is. And if you’re using the right tools, you still have to do work for sure, but there are a ton of great tools out there that make this process a lot easier. And the most basic one is just projections. What we’ve talked about is what really goes into building projections yourself, but that’s a pretty big undertaking. We literally have a team at SaberSim, that’s all we’re doing is just building out the models and working on those.
Andy Baldacci:
And it’s a full-time job. If you want to do that on your own, great, but there are ways that you can leverage other people’s work like ours or any of the other ones out there to give yourself a starting point, so that you can still go through all these other secrets and make adjustments rather than starting from scratch. But even for actually doing the research, there’s a lot of great resources out there. And Max, do you want to share some of those?
Max Steinberg:
There are a lot of great free tools for NBA and a lot of great websites. And there’s things that I use daily that I think really can help you adjust projections, see players that might be under the radar plays, and look at different factors that I think are really important. And so, one of my favorite websites is this website called popcornmachine.net. And this actually shows you a really good visual of how teams are using their rotations. You can see when players have entered the game, when they’ve exited the game, who started, who played in crunch time, and things like that. And you can glean just a lot of good stuff just by looking at these visualizations. You can see, okay, who played backup point guard? Yu can see here with the Lakers, it’s a little weird, because they actually played LeBron at point guard, but maybe in a team like this, you see Kendrick Nunn coming for Tyler Herro, and then Tyler Herro coming back in for Nunn, and then Nunn coming back in for Herro.
Max Steinberg:
So, you sort of are saying, “Okay, these are the two players are playing point guard. Is there a reason that one of these players might play more minutes or less minutes the next game?” You can see, “Okay, are there players that are just playing a lot that the coach just does not like to bring out?” LeBron James is one of these players. I mean, it’s not rocket science, but it’s an important factor. And sometimes you can see things where you can say, “Oh, okay, this player got into foul trouble at a specific time,” or some other factor and you can say, “Oh, interesting. This player played a lot in the first half, then got into foul trouble. Didn’t play a lot in the second half.” This player might be underrated for minutes, because maybe we’re just not taking into account how affected he was by foul trouble in the last game. So, it’s something to keep in mind. And I love these visuals, looking at the last couple of games for teams to just see if there’s anything I can find.
Andy Baldacci:
Mm-hmm (affirmative).
Max Steinberg:
Another site, Add More Funds. This used to be called NBA WOWY. And it’s really cool, because it has sort of what are called on/off splits for players. So, you can sort of say, “Okay, if Trae Young, let’s say in Atlanta is with John Collins, what are his stats? What is his usage rate? What is his total shot percentage?” Things like that. But then you can say, “Okay, John Collins is actually out this game, so how does he do without John Collins? How does his usage rate change? How does his total shop percentage change?” I usually like to look at usage rate, because it sort of shows, “Okay, does this player take more shots if this player is out or in?” And that can be at a really important factor at looking at who might be affected by what injury impact?
Max Steinberg:
These are how we can evaluate an injury impact on a game to game basis. So, I think this is a really good website as well. FantasyLabs, this is not up actually at the moment, I think, but they have a match-ups dashboard. And I look at this every morning, because it just shows me okay, who are the projected starters, given maybe the injuries or whatever. And this can be really important just to look at every day and see, “Okay, who’s starting that I didn’t expect to be starting?” Because then you can go say, “Oh, interesting, this player is injured or there’s a new starter, how might the lineup be affected? How did this play out last game?” So, you can sort of use all these tools together.
Andy Baldacci:
Mm-hmm (affirmative).
Max Steinberg:
As for advanced stats, NBA actually offers these advanced stats on this website called stats.nba.com. And they have a lot of stuff actually. They have touch rate, they have points per touch. They have a lot of stuff, secondary assists. They have so many stats. And I think you can use a lot of them. I think the touch stats are really good, secondary assists are really good. And you could spend hours on this website. I think it’s really good, un-assisted field goals. It’s just a great website to see a lot of advanced stats that might be able to add value to your process. And then lastly, I think Basketball-Reference is a great site, just because I like looking at the per 36 minute numbers, specifically as we were talking about fouls per 36, because this can help you identify high fouls for 36 players and find those players that might have a lower downside than you might think. And all of these tools are free. They’re great I use them every day. I recommend you do too.
Max Steinberg:
And maybe we can put the links to these in the comments section, so people can find them themselves. And then obviously SaberSim. This is our product that I use every day, obviously. And SaberSim is great, because we’re going to optimize for upside for you. So, after you’ve done all this research, looking at all the stuff, you don’t have to spend more time jury-rigging settings and rules just to get the lineups that you want that are going to account for this research. That’s a lot of time that you might not have, and you might not want to spend. When there’s 30 minutes to go and a slate and there’s a really significant injury, you don’t have time to make an entirely new set of rules and an entirely new set of settings.
Max Steinberg:
We’re going to do all of that for you. So, that means you’re going to spend a lot of time, like I get to do, adding value to the process, doing fun stuff like researching, looking at advanced stats, and adjusting these projections, and the ownership projections. And it’s going to make the process a lot of fun. And you’re actually going to be able to add value in the ways that you like.
Andy Baldacci:
Yeah. And it’s really where… When it comes to the tools, there is a lot out there and you’re really just trying to find ways to leverage your time as best as possible and focus on those areas where you can add value. With SaberSim, you’re getting a strong projection system where we’re going to do our best to be on top of everything. But there are just always things that you, through your own process can do to add value that Max and I have talked about today, but we also then take all of that and look at correlations, look at ownership, look at the range of outcomes, and make it very easy for you to build those high upside lineups, so that you don’t have to spend hours and hours fighting with outdated tools or trying to do all this by hand. And one of the big things that’s important to us at SaberSim is we know that not every tool is going to be perfect for everyone.
Andy Baldacci:
We have spent a ton of time on building the best tools that we can, but we want you to try it out for yourself and see if this is right for you, see how it fits into your process. And that’s why we offer a free three day trial. So, you can sign up, play around with it, see how this fits in your DFS process, see if it helps you get better results, and make up your own mind before shelling out a single cent. So, to sign up, all you have to do is just head over to SaberSim.com and you can get started in seconds. And you’ll get access to all of the sports that any of the major sites offer as well as our projections.
Andy Baldacci:
And you will also get access to our Slack channel where you can talk to our partners, like Max and his brother, Danny, and all the other people in there and bounce off ideas from them, get even faster access to injuries and everything else, and really have that collaborative community, so that you can add the most value to your own DFS process. If you have any questions at all, we are always happy to help. And you can contact us at [email protected]. And Max, I know you’re also very active in Slack, but also on Twitter, how can people reach you there?
Max Steinberg:
Yeah, you can find me on our Slack channel or @MaxJSteinberg on Twitter. I’m happy to answer any questions.
Andy Baldacci:
Awesome. Well, thank you guys for tuning in, and as always, if you have any questions, just let us know and we’re happy to help. Thanks again.
Hey guys, this is Max Steinberg. I am a partner at SaberSim, and a daily fantasy professional. I’m here to make a video all about NFL showdowns. So if you don’t know, FanDuel and DraftKings have NFL showdowns, which are these one-game daily fantasy contests that have gotten incredibly popular over the last three years. They’re fun and simple to play, and they’re actually pretty beatable but they require a different approach than full DFS slate with let’s say 10, 11 or 12 games. So in this video, I’m going to talk about how the contests work and how you can use SaberSim to beat them.
So, let’s just look at these DraftKings offerings in FanDuel, and explain showdowns for people who haven’t seen them before. I’m sure most of you have, but let me just talk about it for a second. So unlike other daily fantasy contests, they are based on just a single game. On Thursday, this next Thursday, it’s going to be a Houston/Kansas City week one game, right? And instead of having a lineup with different positions, it is positionless. But, there’s one caveat which is you can designate a captain or MVP, and that player is going to score one and a half times his normal fantasy points.
So let’s say Tyreek Hill, you choose him as the captain and he has a 20 fantasy point game. He’s going to give you 30 fantasy points, and that gives a lot more lineup permutations and combinations, and changes strategy quite a bit. So, what is the actual strategy for these tournaments? Well, I’m going to get into that for a second, but I just want to talk about why I feel SaberSim is really, really well-suited for showdown strategy. Because what a lot of pros like myself try to do when they’re trying to make showdown lineups is, they try to think about what might actually happen in a game. Because in showdowns, there’s a huge concept of how someone does relative to someone else, in making your lineups good.
For example, let’s say Deshaun Watson you think is going to have a really good game, and he has a good one. He scores 30 fantasy points, when we project him at 22. However, in that certain game outcome, Patrick Mahomes then is forced to throw a lot, and he gets 40 fantasy points. Well, the fact that we like Deshaun Watson actually does not help us here. Having him in the captain’s spot over Patrick Mahomes would be a detriment to your lineup, and you’re not going to win that million dollars that you really want to win in these tournaments, right?
And so everything you are doing is trying to optimize your lineup, not for the highest score, but the highest score given a certain game outcome. You could try to sort of optimize for an 80 DraftKings points lineup, if that 80 DraftKings points lineup is going to be an optimal lineup in a certain game outcome. And the thing with SaberSim is, as you know, we are built on a one-of-a-kind simulator. So we take our data and we use all our models, and we run it all through our simulator, and we simulate these games thousands of times. And what that gives us is a range of outcomes for players, their correlations to different positions, but it also just gives us actual simulations that you can build your lineups off of, which I’m going to get to in a bit.
But that’s going to allow us to make some really, really great lineups for you for these showdowns. So, let’s just get into the keys of making showdown lineups. So I have three keys that I’m going to illustrate here, and I think the first thing is, focus on the little guys as much as the big guys, right? A lot of people are going to look at this and be like, “Okay, I want to make my Patrick Mahomes projection perfect. I want my Tyreek Hill projection perfect.” But what you actually should be focusing on is making your Harrison Butker projection perfect, your Sammy Watkins projection perfect, your Kansas City backup running back projection perfect. Because you do that, you’re going to make those lineups where your sixth best guy is going to carry your team, right?
And so we’re going to want to do some research, and I think the best way to start is to just look at news, right? So Kansas City, for example, let’s just look at their team for a second. This is a week before the season, so we don’t have 53-man rosters yet. But if you look at Kansas City, we have four running backs projected. We have their first-round pick, Clyde Edwards-Helaire. He’s projected a lot, but then we have these three guys at the bottom, DeAndre Washington, Darwin Thompson and Darrel Williams. If we can make a decision of who of these guys should be projected better than the other, that might help us a lot.
And so I think a best practice is just, go to Rotoworld, type in let’s say Darrel Williams, like I did here, and you’re going to find this player page, right? And this player page already has some really good, actionable information, right? So we see here, one, DeAndre Washington is in danger of being cut, right? So even if he doesn’t get cut that says to me, “Okay, well, is he really going to be someone who plays a lot?” Probably not. However, this beat writer for The Athletic, Nate Taylor, thinks that Darrel Williams will be the number two back and might even split rushing attempts with Clyde Edwards-Helaire. So that already gives us some great information, which you probably would not have looked at, that is going to be really helpful. You wouldn’t even think of it.
So I might give Darrel Williams five. I might give Darwin Thompson, keep it the same, and I might put DeAndre Washington at one, right? So now we’re focusing a little bit more on the little guys. Another person to focus on is some of these wide receivers. I think Randall Cobb really stands out, right? We have him projected at 5.6. Now, Randall Cobb did sign a 27 million dollar contract in the off-season, and if you look at Airyards.com, another free site I love to use, and you just look at some of the stats on Dallas you can see he actually had 83 targets last year, with a decent target share, a decent aDOT. He has some good stats, so he’s some guy that I’ve heard some people I respect on Twitter talk up. He’s someone, instead of 5.6, I’m probably going to have him about around eight, right?
And so you want to make sure that you’re focusing on these little guys, and making sure their projections make sense. I think also as well, I’m probably going to lower Kenny Stills a bit, because I think Cobb is going to get more playing time than him. Sammy Watkins is another person I’ve done research on. I think I’m going to raise him. I think Mecole Hardman, the buzz might be a little overrated here. I think he’s still working behind Watkins. And now we have what I would feel like is just better projections. I might lower the Chiefs’ defense, but I’m not going to focus on it that much. I’m going to do actually a lot more in the post-build process.
So, speaking of the post-build process, let’s talk about a build. So one thing that I think is the second key is using actual simulations, which we’re going to take care of for you. So have our build settings here, and I’ve shown the advanced ones, but these are the default settings. Smart Diversity is set at very high, and what that means is you can watch a video. That’s Smart Diversity on my YouTube, if you go on SaberSim’s YouTube page, if you want to hear a lot more about it. But essentially when it’s set to very high, it means we’re actually going to build your lineups based on just the results of two or even one simulation, right?
And that’s going to help, because our lineups are going to be optimized for what we simulate might happen in the game, which is what a lot of people try to do. So we’re going to use these default settings. We’re going to start build, and what’s going to happen is, Saber is going to build us some lineups. I just want to illustrate this point, because the difference between optimizing a lineup for just projection and optimizing it for what might happen in the simulation, right?
Because if we optimize for a projection, the lineups are going to look wildly different, right? It’s going to be doing just completely different things. We’re going to have a wide array of captains. But if we actually optimize for Saber Score, then it’s going to look wildly different. It seems to favor Darrel Williams. It seems to like Sammy Watkins a little bit, and it seems to be game stacking QBs a lot, which if you look at the correlations, opposing QBs are very correlated to each other and that’s really helpful.
So interestingly enough, I think that the fact that Darrel Williams is someone we get a lot, may have to do with that I boosted his projection a little too much and I might want to fix this later. But I think it’s an interesting thing, because if Clyde Edwards-Helaire does not get as much playing time as we think he does, Darrel Williams becomes a really good play. And so in the sort of game states, he becomes really, really optimal. So I actually think seeing this exposure is really important and says to me, “Okay, I want to get him in our lineups.”
But so the third key to this is, let’s actually use our post-build tools to take some angles, right? So I want to see what we sort of get in the captain’s spot, so I’m going to lower some exposures and see what’s here. I honestly am sort of loving what we’re getting here. I think wide receivers in general have high variance, and I think this is a unique slate because a lot of people traditionally would just want to put the high-scoring player in the captain’s spot. But the thing is on DraftKings is, there’s some high salaries here.
So let’s say we’re like, “Okay, well, I want Tyreek Hill in the captain’s spot. Okay, I want to pair him with Patrick Mahomes. I want DeShaun Watson with that correlation. Oh, crap. I only have like 3,000 left.” So different showdown slates are going to look differently, but in this one it seems like there’s actually some value at saving the captain’s spot. And so I think what’s interesting here is, it seems like Darrel Williams in the captain can actually have some good value, right? And so I think I’m definitely going to keep some lineups with a captain. I think Randall Cobb in the captain make some sense as well. If we have Randall Cobb, he’s someone who could be the highest-scoring wide receiver on Houston. I certainly believe that. I’m going to want to research playing time a little bit more and feel certain about this, but I think he could be someone.
I think Sammy Watkins as well. Obviously, if we look at Sammy Watkins from last year, he’s in a high-powered offense. He actually had a 20% target share, like he’s a good wide receiver. Obviously, he’s capable of some big games. So these are three players that [inaudible 00:11:15] I’m going to want in my lineups. I’m really curious also to see the Harrison Butker lineups, if they make sense to me. Because if intuitively they don’t, I might just zero him out of my lineups. We give you that ability in the post-build process. I can just say, max exposure five and we’d only get him once, or I can zero him out and we won’t get him at all, but I’m not going to do that yet.
But let’s see these lineups that we get with him. So Harrison Butker, Patrick Mahomes, Mecole Hardman, Travis Kelce, Clyde Edwards-Helaire and Will Fuller; what is the type of game that this could happen? I think this does make a little sense. This is going to be a low-scoring slate where Houston really fails to move the ball, does not do very well, and Kansas City does do well, well enough where they’re scoring quite a bit, but maybe not scoring touchdowns. They’re getting some field goals, and getting Harrison Butker in here allows us to get Kelce, and Edwards-Helaire, and Mecole Hardman, and allows us to get that stack with getting some pretty good points in the captain.
Will Fuller coming back I think makes a lot of sense, and obviously he would probably get a lot of receptions if they’re drilling a lot. So it makes sense that we could have a Harrison Butker lineup, so I don’t know if I’m going to remove him at all. I do think I’m going to cap my Darrel Williams exposure until it would come closer to game time, but I think you can use the same angle. If you’re doing 20 lineups, you can mess with this a lot. But if you’re not you can also … If you’re saying, “Okay. I want to make one or three shared as a single entry,” I would look through this and just see what game states make sense and seem the most likely to you, right?
So okay, let’s look at this one. Darrel Williams, Tyreek Hill, David Johnson, Patrick Mahomes, DeShaun Watson, Harrison Butker; this is an interesting one. It doesn’t stand out to me as a game state where I’m saying, “God, I really think this is going to happen,” or especially likely and stands out, so I’m going to go down a little bit more. Interesting; Darren Fells, DeShaun Watson, Will Fuller, so this is an interesting one. This could be sort of a lower-scoring game state. It could be Tyreek Hill has a big touchdown with Patrick Mahomes, Darrel Williams gets some runs, and then DeShaun Watson has a few passing touchdowns, maybe one to Fells, maybe one to Fuller, but there’s not a lot of yards in this game. I think this makes sense, so I might choose this one, right? Or just look down and see which ones stand out, right?
But what you can also do is just sort of use this post-build process to take an angle. Be like, “Is there anyone else I want to take an angle on?” Well, as we said before, I don’t really like Clyde Edwards-Helaire. I think Darrel Williams is going to get more run than people think. So I might just take the angle that I’m just not going to use him a lot, and give him 10% exposure, and let some other guys take up that ownership, right? David Johnson; Houston traded for him. I think it makes sense that they would run with him a lot. They traded DeAndre Hopkins for him … That they would use him a lot, that they might run the ball more. So he’s someone I like. I might do this to get more exposure to him. I might raise the amount of lineups that are there so I can get him into captain, and then give him 10%, and then change this again.
And so again, I think in general the important part of showdown more than ever is going to be this post-build process. Because basically what we’re trying to do is, we’re trying to just stuff the angles and the lineup types that we want into our lineups, and not worry about projection as much. Because you know, I don’t really care that this is projected at 93, and this line is projected at 95. I just care about, “Okay, if the game is simulated, would this be the best lineup? Yes? Okay, well, let me get lineups with these players, because I want these types of lineups.”
So, it’s kind of like a unique strategy and it allows you to take advantage of this post-build process, which I think is really, really fun. And I haven’t done much research yet. I probably, if I was just making 20 lineups, I probably would take some more definitive angles. Maybe I’d use a lot of Darrel Williams. Maybe I’d use a lot of David Johnson. Maybe I’d use a lot of Sammy Watkins. Do some more research, think about the angles you want, and then instead of adjusting projections, adjust your lineups in the post-build process to get those players.
So, those are a few layers that I think SaberSim can give you a nice edge in showdowns. I think they are contests that you can get an edge on, and they’re really fun, and it reminds you to think a lot more. It’s a lot less about projecting and a lot more about, who do I want to take an angle on? What players do I want to focus my lineups around? And then let SaberSim pick those lineups for you, because we’re simulating the games and they’re going to be the most optimal lineups for a certain game state. So, hope you enjoyed this video. NFL season is just around the corner, and SaberSim does have a great promotion, which is you get a three-day free trial by signing up, no strings attached. Sign up for us on Wednesday. You can use us for a Thursday showdown. Hopefully you’ll like us enough and pay for a subscription, so you can use us for NFL week one. Try us out. See what you think. So thanks again for watching, and good luck with your lineups.
Andy Baldacci:
There is about a 15 second delay, 10 second delay between when you guys see us on the YouTube stream and when we’re saying this. We’re going to do our best to stay on top of everything. There might just be a little lag there. Welcome to the second Ask The Sharks. We’ll just give everyone a minute or so to get in here.
Andy Baldacci:
Yeah, excited to do this and to talk about sliders and about this big new update that we’ve pushed out. Where’s everyone joining us from today? Max is in a random hotel room.
Daniel Steinberg:
Are you?
Max Steinberg:
Yeah.
Andy Baldacci:
Yeah. Danny is in California. Matt? You’re coming from Oregon?
Matt Hunter:
Portland, Oregon. Yeah.
Andy Baldacci:
Nice. I’m in Texas. What about you guys on the stream? Okay. Let me just share this to Slack.
Max Steinberg:
Someone’s from North Carolina. East Coast.
Daniel Steinberg:
Wait. Where’s the YouTube link?
Andy Baldacci:
I’m posting it in Slack right now.
Daniel Steinberg:
Okay.
Max Steinberg:
Ones from the 305, that’s [crosstalk 00:01:02].
Daniel Steinberg:
I’ve never lived in the 305. From Florida, Miami.
Max Steinberg:
Yeah. Maryland.
Daniel Steinberg:
Yeah, where in Maryland? Me and Max grew up in Silver Spring.
Max Steinberg:
Yeah.
Daniel Steinberg:
Spring or springs?
Max Steinberg:
I think it’s Spring.
Andy Baldacci:
You forgot?
Daniel Steinberg:
It’s hard to remember if there’s an S at the end or not.
Max Steinberg:
608.
Daniel Steinberg:
Parkway Drive. What’s 608? Indiana?
Andy Baldacci:
I tried. I tried to get Squid on this one. He’s on his NASCAR binge. I think he’s building like a NASCAR simulator right now in all honesty. I don’t know what started this. He’s like the least but most NASCAR guy you’d ever meet. He has no idea about anything to do with it, but loves drinking beers in America even though he’s Canadian.
Andy Baldacci:
Hopefully, at the end of this weekend, we might have some data NASCAR sims to go with. Yeah, he won’t be joining us today. Anyway, so I think we’ve given people enough time to get in here. We’ll kick it off.
Andy Baldacci:
My name is Andy Baldacci. I’m the CEO of SaberSim. With me today, I’ve got DFS pros Max and Danny Steinberg. We also have the founder of SaberSim, Matt Hunter. How’s everyone doing?
Matt Hunter:
Good. Hey, guys.
Daniel Steinberg:
Hi.
Max Steinberg:
Doing great.
Andy Baldacci:
Very lively. I know you’re on [crosstalk 00:02:39]. It’s 10:00 a.m. Anyways, today what we’re going to do is just really dig into SaberSim’s sliders and that’s correlation, ownership fade and smart diversity settings. Those are what we really use to build that pool of high upside lineups. This, frankly, will be most valuable for people already using SaberSim. If you’re not familiar at all with our tools, that’s fine. Still stick around because I think you’re going to get a lot out of this.
Andy Baldacci:
We also do have a free three-day trial. If you want to see it in action, you can sign up for that over at sabersim.com. When I was kind of just thinking about this and we were all figuring out the roadmap of SaberSim and what we want to build, we just went back to what originally got us into this and why we built SaberSim to begin with.
Andy Baldacci:
Really, we were just kind of frustrated with having to spend hours researching every slate and then setting all kinds of rules and groups and everything else in optimizers, in order to actually use that research to build strong lineups. The reason that software exists, the reason that we use software in our lives is to make things easier. It seems to me like the traditional DFS tools did not get that memo at all or maybe, I don’t know, they’re just too lazy to innovate.
Andy Baldacci:
Whatever it is, just the tools that were out there made things harder. What we try to do at SaberSim is just cut out all of that busy work so you can invest your limited time where it has the biggest impact. That’s on things like staying on top of the news and dialing in your projections, not adjusting 1000 different settings. Our mission is to make it easy as possible for you to build winning lineups. It’s always going to take work, there’s no easy button to beating DFS. We do the grunt work for you so you can build on a strong foundation, rather than just wasting your time starting from scratch every single day.
Andy Baldacci:
We have made some huge improvements on that traditional DFS process. There’s a lot more we can do. The update that we just pushed out maybe an hour ago, I don’t even know if it was that long ago, is a big step in that direction. To summarize, the way SaberSim works is we follow a three-step lineup building process. First, you review and adjust the projections. Then, you choose your bill settings and finally you dial in your exposures.
Andy Baldacci:
The update today entirely reworks that second step so that it’s not just more intuitive, but it also gets you better results. Just being honest, I think, the biggest mistake we made when planning out the app is how we presented those sliders, how we presented correlation, ownership fade and smart diversity. We kind of just threw it at you, gave you a few defaults for how to set them and just sent you on your way.
Andy Baldacci:
For us, we’ve been working on this for so long that these ideas were kind of second nature to us. It really is an entirely new way of building lineups. We weren’t doing enough to help you set them. The defaults that we had in there, which we didn’t originally, but we added in, the defaults helped, but they weren’t specific enough. You have to make too many adjustments and that led to a ton of rightfully asked questions. Like, how should I set these for single entry contests? Are they the same for NFL and NBA? What’s considered a small slate or a large slate? Are multipliers more like cash or GPPs?
Andy Baldacci:
There were just so many different variables in there that it made it really confusing. It ultimately had people feeling like they were looking for some sort of Holy Grail combinations of sliders that always worked. That’s really the issue that we’re trying to resolve with this update is that there are no best sliders. If there were, we would just give them to you. We’re not holding anything back. We’re not trying to hide anything from you. If there was this kind of Holy Grail combination of sliders, you’d have them.
Andy Baldacci:
The reality is that you can build winning lineups with many different combinations of sliders. They are an incredibly important part of the process, because they control how we build upside into your lineups. There’s no perfect combination. Really, you shouldn’t spend much time at all getting them dialed in, at least with this new update. Before learning how to make those adjustments was not simple because when you looked at all those variables, there were hundreds of adjustments you’d have to make and it was different for each sport, if you want to get it really dialed in.
Andy Baldacci:
We realize that we weren’t kind of living up to that ideal of making things as easy as possible for you and doing what software is meant to do. We were adding unnecessary busywork to the process. Rather than expect you to watch all the videos and figure out all the adjustments to make, in today’s update, we just built that into the app. Matt is going to show you this in just a couple minutes.
Andy Baldacci:
What you can do in the build settings now is you just tell us the contest type, the entry limit, and the size of the contest you’re building for. We’re going to set the sliders based on that and automatically adjust them for the number of games on the slate as well. Matt has spent the last few weeks building smart defaults for all of the hundreds, if not, thousands of different combinations. Again, he’s going to walk through those in just a minute.
Andy Baldacci:
We still do let you make any adjustments that you want based on your risk tolerance or personal preference. Because we’re taking into account all those important variables, you really can just leave them alone for the most part and just stop worrying if you’ve set them right. You’re going to have a much stronger foundation going forward. Again, make the tweaks, make the adjustments, but if you’re confused about them, if you’re not sure exactly what to do, you can leave them alone and know that we’ve got your back there.
Andy Baldacci:
Matt, can you walk through those new defaults and just kind of explain how these variables work? We’ll see how smoothly we can get this screen sharing.
Matt Hunter:
Yeah. Yeah. I’ll just share my screen here.
Andy Baldacci:
As we’re doing this, I’m going to be keeping an eye on the chat to see what questions are being asked. At the end of this, we’ll do a live Q&A. I think the demo itself though will get into a lot of the questions that are asked. Again, if you have anything, just put it in the chat, and we’ll circle back at the end.
Matt Hunter:
Yeah, does it look good? You can see my screen?
Andy Baldacci:
Yeah.
Matt Hunter:
Okay, awesome.
Andy Baldacci:
No one really cares anyway, so we’re good.
Matt Hunter:
Yeah. Like Andy said, basically, the change that we made is to have much more granular options for the type of contest that you’re entering. Then, being able to see how those different choices will affect the sliders. One thing that we want to show here is that you don’t even need to … for newer users that don’t even want to mess with it, or don’t really care what the slider is at, you don’t even need to see those settings. You can just close that and just choose based on your contest and that’s okay.
Matt Hunter:
The work that I’ve been doing is to make these slider settings as close as possible to what you’re looking for for these types of contests. I’m just going to go through how these sliders change and how the settings adjust based on the different type of contest. We’ll just start. The default view that you get is a GPP, which for those that are kind of new to the terminology is just a tournament in DFS.
Matt Hunter:
Generally, that’ll be top heavy so that a lot of the payout is going to the top 110 or 0.1% of the contest. We also have the ability to change based on the size of the contest and the entry limit. We’ll start off with just showing the 150 max. This would be like the big contest on DraftKings or FanDuel, where there might be 10 to 50,000 entrance. Often, these would be like a $15 entry limit or sorry, entry fee. They could really be anything. You might be playing the mini max that’s $1, but it’s still the 150 max 50,000 entrance.
Matt Hunter:
You can see that we start out with a very high smart diversity and a few takes up on ownership and a little bit on the correlation. Basically, the reasoning for this is smart diversity is a way to give you more upside in your lineups. You want this to be high, especially for a fairly large field, because you’re really looking to get in that top 0.1% of the entrance. What smart diversity does is it basically narrows the range of simulated games that we’re looking at in each lineup.
Matt Hunter:
You’re getting lineups that have a higher ceiling and we might be sacrificing a little bit in terms of the average performance for the lineups. What we’re getting is lineups that have more of a ceiling. Even if you have correlation lower, those lineups are still going to incorporate the correlation that’s part of the simulations. Because when we’re choosing these lineups, we’re actually adjusting the points so that they account for the way that the players relate.
Matt Hunter:
We can see that actually, if we look at the entrance, we can change it even higher. If you are entering a contest that’s maybe it’s 100,000 entrance or in NFL, you might have the Million Maker which is 250,000 or something like that. If we choose that, you actually see that the smart diversity goes to very high because, really, we care about that absolute ceiling for the lineups because it’s so difficult to get first place out of that huge field that you really want to look for those lineups that have absolute ceiling.
Matt Hunter:
Then, you also saw that the ownership fade and correlation increased as well. That’s just because those factors become even more important, the more we care about that upside and getting to the top 1%. With 150 max contests especially, those are generally going to be those contests that are more difficult, they’re more top heavy, and so there’s less value in getting a min cash where we’re just trying to get them the top 20% of the contest. We really care about getting at the very top to have like a plus CV outcome.
Matt Hunter:
When we change that, we see that the ownership increases, the correlation increases and smart diversity increases. As we go down this list, we’ll see that all three of these settings are going down with it. We still care about correlation and ownership and a 1000 entry contest, but not as much and we care more about just having high average outcome rather than that ceiling outcome because we just don’t need to have a really, really ceiling performance out of your lineup when you’re only competing against 100 to 1000 entries.
Andy Baldacci:
One of the cool things that I really like about this is one it obviously just makes it a lot easier. It also just makes studying this process very straightforward because rather than kind of having to use rules of thumb on, okay, if I were to change, like … what is the impact on a single entry limit versus a 20 max entry limit? I can just see by switching from those two and just see what we think changes. It just makes it a lot easier for me to just build like a mental framework for, okay, here is what the baseline is for these type of contest. Here’s how they differ.
Andy Baldacci:
As I make changes, these are the types of things we’ll be looking at. Matt, these are different for every sport as well, right?
Matt Hunter:
Yeah. I can go over to NHL right now and …
Max Steinberg:
Can I cut in for a moment?
Matt Hunter:
Yeah, sure.
Max Steinberg:
Thanks. One thing that I think is important is smart diversity and correlation are linked together in an interesting way, because as you raise smart diversity, as Matt said, we’re building your lineups on a smaller and smaller group of simulations. What correlation is in practice, right, is how players actually score together in the real outcomes of games, right? That’s what that correlation number is.
Max Steinberg:
If you raise smart diversity really high or the preset smart diversity is very high. What that means is, we’re just building your lineups based on the projections of just a few games. Matt, tell me if I’m wrong, the highest setting is just actually one game.
Matt Hunter:
Yeah.
Max Steinberg:
When you have smart diversity very high, we’re actually already incorporating correlation in a way because players are going to do well together in those small packets of simulation. That’s why a lot of times in certain sports, especially NBA, you might see smart diversity as a preset setting very high and that correlation low. That doesn’t mean that we’re not taking into account correlation. It just means that we’re not being as repetitive with it because smart diversity already has taken that into account.
Matt Hunter:
Yeah. Yeah. Just to show you more about the smart diversity when you have it very high, you’re looking at individual outcomes. That’s why when you get into showdown, we really prioritize the smart diversity more than in a classic slate with more games. When you have smart diversity really high for showdown, literally, every lineup is essentially the optimal lineup for a given simulation of the game. For especially these large field GPPs in showdown, that’s what you want. You need to get the top scoring possible lineup depending on …
Andy Baldacci:
Yeah, you’re basically saying like, given this game script, actual game outcome, what is the best lineup from that? That’s what showdowns are all about is saying, okay, here’s one game, what has to happen for these players to do well together? That’s something that a traditional optimizer just cannot do because they’re not going to get the correlations. They’re not going to get how players’ performance impacts the others. By just looking at the outcome of a single simulation, that’s basically the most straightforward way of doing that.
Matt Hunter:
Right. Yeah. With other optimizers, you can create rules that will kind of get you closer to that kind of game script and incorporating that into lineups. Without having a simulator, really, it’s not possible to get those really fine-tuned lineups that have those specific game scripts. You can create all the rules you want, but you’re not going to be able to get every single possibility there. Yeah, that’s why for showdown, we don’t even have correlation set because all that really matters is those simulations which are incorporated into the smart diversity.
Matt Hunter:
If you’re not in correlation, you’re almost double counting the correlation because that’s already incorporated there.
Andy Baldacci:
One of the things, I just wanted to stop you there, Matt, for a second. Just to get into a couple of the questions in the chat. Don asked if the settings are different for different DFS sites. Right now, they’re not. This is an area where you could potentially make some adjustments. If say, Yahoo or FantasyDraft, their payout structures are flatter than DraftKings. You might want to make some adjustments but it’s going to be pretty minor.
Andy Baldacci:
In that case, what I would do is probably like, if assuming that is actually accurate, which I’m not sure, if it’s a flatter payout structure, meaning the prize pool is spread out amongst more of the participants. Then, I would just tone down the sliders slightly from whatever the defaults are. That’s getting pretty in the weeds. I frankly wouldn’t worry too much about that.
Matt Hunter:
Yeah. For sites like Yahoo, they’re all ready, I think, all the contests … well, they might have 150 bets now but a lot of the contests are three max or 20 max. That’s being incorporated already is when you have the entry limit that’s going to impact how those sliders are treated and how we’re kind of building those lineups. Other than maybe NFL, most Yahoo and FantasyDraft contests aren’t going to be 50,000 entrance and 150 max. If they are, I think you probably do want to go with a similar sort of settings.
Matt Hunter:
Yeah. I think you’ll see if you play around with these that you can notice the patterns and how the sliders change depending on these options. If you want to get advanced based on the payout structure, I think, that’s definitely reasonable. You want to pay attention to what the defaults are giving you to begin with and sort of be careful with how you’re adjusting those and you don’t want to overcorrect.
Andy Baldacci:
Right. Because it already is sort of implied in that.
Max Steinberg:
Although, with the style as well that also has to do with the payout structure, right. Because if you do satellite or wire, it sort of is something that speaks to how top that’d be. If you’re doing like the DraftKings Millionaire Maker, you could set that like a satellite just because it’s if it was like one of those ones where the price falls to a million that’s like one million …
Andy Baldacci:
Yeah.
Max Steinberg:
Yeah.
Matt Hunter:
The assumption with satellite is essentially that it’s a winner take all. If you are in a winner take all, I think it’s safe to treat that as a satellite. You can see even with the same entrance, same entry limit, we’re increasing all of the ownership fade, the smart diversity because you need to be at the top of the contest. That might get impacted here.
Andy Baldacci:
Mike just asked in the chat, what’s the rule of thumb for determining if it’s small slate or large slate? With this update, basically, you don’t need to worry about that anymore. That was one of the mistakes we made before is we just weren’t thinking it all the way through when we were forcing you to do too much work. With this, what’s happening, we’re not asking you to determine if it’s a small state or a large slate because we know how many games on the slate.
Andy Baldacci:
We’re not just considering the style entry limit in entrance when putting the sliders together. We’re also behind the scenes saying, Okay, this is a six game slate or this is a three game slate or this is a 10 game slate, whatever it is, we’re also adjusting based on that. That’s kind of just built in automatically. That’s something you don’t need to worry as much about.
Andy Baldacci:
Before we go on, I want to just jump back to a question Brian asked about why only one sim is better than hundreds for identifying upside. It’s not necessarily that it’s better as a rule of thumb, you always want to use one sim in every situation. What we were talking about specifically is with showdowns. The reason for that is because what upside ultimately means is who has the potential to significantly outperform their average projection. Because if someone gets to an average of whatever, we’ll say 30 points, but their range of outcome is always between 20 and 40.
Andy Baldacci:
Or there’s another one who is between 10 and 20, some other time, but then they’ll have the game where they get 50 points, whatever it may be. That’s not going to show up in the average. The more granular, a better way to say it is like the fewer simulations you’re looking at, the easier it is to pick out those spots where someone really has a potential for those bigger games.
Andy Baldacci:
That’s something that, again, does get pretty technical. Matt, those are for showdowns. Matt, can you add to that a little bit?
Matt Hunter:
Yeah, yeah, I can add to that. I think even outside of showdowns, you’re right that individual simulations don’t necessarily give you a better idea of upside. Part of it does depend on your lineup pool as a whole. When you’re building 150 lineups using really high smart diversity, what you’re getting is you’re going to have more of players that have those higher ceilings in terms of the exposure to those players in your entire lineups than you would if you’re just building on average projection.
Matt Hunter:
I think one way that I like to think about it is if I’m building my lineup pool and remember when we build our lineups, we’re actually building 1000 or 1500 lineups in the background that we kind of then can filter through. If I have smart diversity at the max, my lineup pool is essentially consisted of approximating the probability that each player is going to be in that top lineup in the big lineup basically. Then, you can sort through that pool and figure out, okay, but I want to incorporate … if you sort by projection, instead of Saber score, you might be having a little bit less risk there.
Matt Hunter:
Or if you want to edit your min or max exposure to certain players, you can adjust that. When you are building your pool with this high smart diversity, your pool is consisting of those high upside players and better incorporating the probability that a player is going to hit that ceiling. I think you don’t want to just think about it as the individual lineups with the individual sims but more about probabilities and having your exposures match the probability that players will hit that ceiling and be in that top scoring lineup.
Andy Baldacci:
Were you going to add something Max?
Max Steinberg:
Yeah. I was just saying and I remember simple just something that can remind is, if you have higher smart diversity, what you’re essentially doing is you’re sort of simulating the slate. In a real world slate, what ends up happening is every game is played once on the slate, right? When you have smart diversity set to very high or at the highest setting, what you’re basically saying is … we’re saying, okay, we’ve work in one simulation of every game on the slate and seeing how it plays out and then picking the best players from that way that that slate played out.
Max Steinberg:
It actually is almost simulating not just the game. It’s simulating actually what happens in the slate and then picking the best players from there.
Matt Hunter:
Yeah.
Andy Baldacci:
Yeah. One of the things I was just going to touch on quickly is, yeah, to kind of build on what Max is saying is that ultimately, for a showdown specifically, there’s one game literally being played. It’s just that one game. You’re not looking to see how players perform on average at all. You want to see, in a single game, what are the possible outcomes and by pulling out individual simulations, that’s basically what you’re doing. That’s how we simulate all the games is we play by play, go through and simulate every game 10,000 times.
Andy Baldacci:
That’s basically 10,000 separate games of this game. What you’re looking for is if you pull out one and say, okay, if this game plays out this way, what is the best line from that? That’s what we’ll give you. There is definitely some risk with that. Because you’re hoping that the game does play out that way. That is also probably the highest ROI approach. If you’re going to use more traditional approaches of looking at averages, you might cash more frequently.
Andy Baldacci:
There’s almost never a game that plays out exactly like average. Hopefully, that helped explain it a little bit. I have been keeping track of a lot of other questions. I’ll try to work those in as we go. If I don’t get to them right now, though, at the end, we’ll circle back. Matt, I was going to ask actually, we’ve been focused really on GPPs. Do you mind just showing cash?
Matt Hunter:
Yeah. Cash is a lot simpler because, really, you’re going for more of that average outcome. We don’t really care about correlation. We don’t care about ownership much and the smart diversity is low, because we really are caring more about the average outcomes here. You can even see, so we’ve been focusing on the sliders but with cash, we increase the minimum projection and have the high min salary as well. That’s just to ensure that the lineup you’re getting is more geared towards just maximizing that average outcome.
Matt Hunter:
With cash, I would say the sliders are not as important as they would be for GPPs. For the tournaments, we’re really caring about those distributions and the correlations, especially for sports like MLB and NHL where correlation is so important. For cash, it’s really the focus, I think, we kind of have these defaults, but the focus is really on projections and just maximizing the average score of your lineup. I think our approach is still good here of having the sliders here. If you really want to focus on more getting those chalkier players, you can even have a negative ownership fade where you’re trying to get those chalky players in.
Matt Hunter:
In general, I think we would recommend you don’t really incorporate correlation or ownership and you basically just care about building lineups based on close to much bigger groups of sim, so it’s more closer to the average outcome.
Andy Baldacci:
Yeah. What we should do is also we’ve been focused on MBA and we obviously aren’t going to be able to kind of walk through everything for every single sport. I think it’d be interesting to show how they do differ for NHL, which is a very different sport in terms of variants and everything else involved than MBA.
Matt Hunter:
Yeah. The cash kind of stays the same because …
Andy Baldacci:
The principles are the same.
Matt Hunter:
Right. The principles are the same. When we get into GPP, so just one thing to keep in mind is we don’t currently have ownership projections. We have this as neutral now. If you’re uploading your own ownership, for hockey, sorry, for hockey we don’t have ownership projections. If you’re uploading your own ownership, I think, you would want to incorporate that, similarly to other sports. Just looking at correlation and smart diversity, you can see it is a lot different than in NBA, because correlation is just such a huge factor in hockey.
Matt Hunter:
If you haven’t really played hockey DFS or you don’t know the sport too well, it’s just really important because you’ll have players that are skating on the same line together. What that means is that players will basically be spending most of their time in the game with a few other of the same players. They’ll kind of shift in and out of the game in groups or in lines and for power plays as well. It’s a lot specific power play lines. When a player gets a goal, it’s very likely that someone on their same line will get an assist or vice versa.
Matt Hunter:
You really want to care about those correlations and try to get that upside because most of the points do come from goals and assists in hockey. That’s why we care a lot more about correlation in hockey. It’s similar to NBA where as we get lower in our contest, in our entrance, our smart diversity is going down a bit. Correlation is still staying pretty high because we still want to have those stacks and we still want to keep our alignments together because even getting in the top 1% of the of 1000 entries, you still have to have the stacks and you’re still looking for that correlation to hit that upside.
Andy Baldacci:
Did you want to jump in, Max?
Max Steinberg:
I just want to say, Matt, would you mind showing people the player pages for NHL versus NBA just so they can visualize the players and …
Andy Baldacci:
Yeah.
Max Steinberg:
… its difference. I think that hammer on the point like the correlations are much different and much stronger in NHL than there in NBA.
Matt Hunter:
Yup. This is just like a frontline skater and you can see that it’s very high correlations with the players that he’s likely to be on the …
Andy Baldacci:
I’m not a math guy and Giant Squid took most of my math tests growing up, so I’m going to need some help on this one though. What do these numbers mean for correlation? How do I know that that’s high?
Matt Hunter:
Yeah. Basically, the scale goes from negative one to one. Negative one would basically mean a perfect inverse relationship where it’s linearly the more that one goes up the other goes down. A correlation of one would be basically just, yeah, a perfect positive relationship. Every single time, Max his, I don’t know how to pronounce his last name, Pacioretty every time his points go up, Mark’s points would go up by the exact same percentage.
Matt Hunter:
It’s essentially just a scale on that, so 0.29 might not seem like a lot when you think about it from zero to one. In terms of comparing to other sports, you’ll never see an NBA correlation that is that high. Generally, they’ll max out at maybe 0.1 or 0.15 MLB, which people consider like a very high correlation sport. Generally, you’re not going to see two batters that are higher than 0.15 or 0.2. For MLB, they stack on top of each other so you’ll sort of get a higher correlation as with the lineup as a whole, rather than the two players. This is still a very high number. It might not be as much as, say, a quarterback and a line receiver, but it’s still very high.
Matt Hunter:
Especially, when you consider the correlation between multiple players at once where if one of these players gets a goal, it’s very likely that one of the other ones got an assist or two players got an assist. You might get two assists on with one goal and so all three of them go up at once. You really see that correlation there. Whereas for NBA, it’s just not going to be as high and you’re certainly not going to get the amount of stacking with NBA where you might have two players that have high correlation, but you’re not likely to have three players that have a high correlation.
Matt Hunter:
It just kind of goes down as you increase the group whereas hockey and baseball, you’re going to keep increasing that correlation the more players you add.
Andy Baldacci:
Danny, what were you thinking?
Daniel Steinberg:
I was just going to say like a really simple way to think of like why correlation matters is if you think of a toy game where you’re making two lineups and Max Pacioretty is in one lineup. Like let’s say Max Pacioretty has a really good game, would you rather have, is this [Malik Stone 00:34:57], I can’t actually read the last run.
Daniel Steinberg:
Would you rather have Matt Stone with Max Pacioretty has a really good game is going to do really well? Or maybe Noel whatever, who was a zero percent correlation who may be better projected than Matt Stone. I mean the thing is, even if someone’s like better projected than another player, if they’re correlated, that means it’s more likely they’re going to have a good game when the other person has a good game. That’s exactly what you want when you are trying to make a lineup for this high upside.
Daniel Steinberg:
Correlation is a really good way to measure the tradeoff between, okay, do I want this guy who may be a little higher projected? Or should I put this guy in a lineup who has a little more correlation who may be worst projected?
Matt Hunter:
Right.
Andy Baldacci:
Napoleon, actually, he asked a question with that, I think it’s relevant to talk about now. For hockey, why not set correlation to the highest settings for the sliders?
Matt Hunter:
I think it’s still all just a balance between upside and risk and average performance. Again, it does depend on the type of contest. We do increase the correlation depending on what you’re selecting here. With the highest correlation, you’re going to get a ton of stacks and you might even get four or five player stacks. While they might have kind of an absolute ceiling, they’re not all necessarily … if you have five players from the same team in hockey, maybe one out of 1000 or one out of 10,000 times, they all do really well.
Matt Hunter:
When you have increase in the correlation that way, you’re kind of isolating those top performances, but you might be sacrificing more of the average performance. You don’t always necessarily want to have it at the complete max because you’re then ignoring a lot of those simulations where they don’t necessarily all do well together. It’s all just that balance between risk and upside. That’s how I would describe it. I don’t know about you guys.
Andy Baldacci:
Yeah. I think just as kind of a word of not even really caution, but at the extremes of the scale for correlation and ownership fade, you can just get some kind of unexpected outcomes. I think unless you’re playing super top heavy contests, you really just try … I just always try to avoid the very highest end of the scale. Because I don’t think there’s enough benefit there.
Andy Baldacci:
I think even at high for hockey in this example, you’re going to get basically all of the benefit of correlation that really is there. Above that is going to be a lot of noise and I just don’t think it’s worth the extra risk that it adds. It’s just trying to be cautious there just because correlation is important doesn’t mean you just want to hammer it all the way to the right.
Matt Hunter:
It does depend on the sport. We don’t have MLB up yet, but I think MLB is going to be a little bit different than NHL, even though they’re both very correlation dependent because MLB the correlation just keeps going up and up, the more players you have in your stack. Whereas hockey, it’s not necessarily entirely linear and they’re sort of … you’ll get a ton of correlation with three. If you could, in theory, put eight players on a team in a stack, you would be lowering your ceiling a little bit because not all eight of those players are going to do well.
Matt Hunter:
Whereas baseball, they continuously correlate with each other more and more. You still do have a little bit of that effect where if you could put the entire nine men lineup, baseball lineup, in a DFS lineup, you might have a little bit of absolute ceiling if for some reason if the team scores 30 run or something. You’re limiting your upside because it’s unlikely that all of them will have that outcome.
Matt Hunter:
There is nuance there between the sports and that is partly why we are trying to make this change is because before that, it’s harder to show that nuance can get that into your lineups because there wasn’t as much granularity in these options. Whereas, now that we have this with MLB, we can adjust these settings and adjust the slider so that you’re better balancing that ceiling upside and your average performance and not increasing risk too much.
Andy Baldacci:
I’m going to jump in and go over a couple of the questions. Just a few quick ones.
Matt Hunter:
Do you want me to turn off the screen share or should we keep this up?
Andy Baldacci:
We can shut it off. Yeah, so there are a few quick ones. Jamie asked when will we be getting projected ownership for hockey? That’s something we have in the works. We didn’t really say anything about it. Matt and Giant Squid actually pushed out a pretty big update to the NBA ownership model that we’re really excited about and through our testing has made a significant impact. We’re going to be working on applying that to other sports, probably starting with hockey. It does take some time and there’s a lot we have in the works to get ready for baseball, which starts up in almost exactly a month.
Andy Baldacci:
I can’t guarantee when we’ll be getting that out but it is something we are working on and just improving all of the ownership models in general.
Matt Hunter:
Yeah.
Andy Baldacci:
Don had asked earlier, will we include FanDuel single game? That is something we do plan to add probably shortly after baseball begins or right around that time. Then, more on the topic of the sliders, a good question is, do the sliders work when using custom projections? The short answer is yes. Matt, do you want to give a little bit or a longer explanation?
Matt Hunter:
Yeah. Obviously, this is a very tricky situation to solve because the sliders are … especially smart diversity is really based on the simulations. If you’re uploading your custom projections, those aren’t based on simulations, but what we do is we basically take…
Andy Baldacci:
Not based on our simulations.
Matt Hunter:
Yeah. Yeah, you might have a different stimulator. If you upload your custom projections, essentially what we’re doing is kind of shifting the distributions for the players that we created and shifting them to match your new custom projections. If you have a projection that’s say 10% higher than ours for a player, we’ll essentially take our player distribution that just kind of shift it over 10% so that you’re still getting the benefit of the standard deviation and those distributions that we create but then with your own projections.
Matt Hunter:
It’s not necessarily like the 100% perfect solution but I think it does a pretty good job of still allowing you to take advantage of those simulations while still using your own projections and be able to input your own research and work into it.
Max Steinberg:
Yeah, and I’ll just add to that, Matt has worked really hard on this aspect of it. I think, in general, intuitively how you feel, think it should, when you change someone’s projection, how much exposure, how it changes the upside. It does fit what you would intuitively think. Yeah. When you raise someone’s projection, you are going to get them more. If you lower some projection, you’re going to get them less. The players will interact in the same way.
Max Steinberg:
You’re still getting the best of our product. We try as best as possible to make it sell you adjusting or making your own custom projections work as best as possible and I think we do a really good job with that.
Andy Baldacci:
Yeah. It also goes back to what we were saying before about how, at the extremes at putting smart diversity all the way up and using custom projections. It’s not going to be a perfect match, but it is more than good enough and it is basically about as good as we can get it. There are just some technical limitations there. I like my simple answer of yes, we do the things with your custom projections. Matt added a lot more nuance there if you do have questions about exactly how we do that.
Andy Baldacci:
One of the other questions was, I’d like it just because of how kind of broad it was but I think it can tie in a lot of things. It’s just, how important are projections and are SaberSim projections good enough? Should you be doing anything beyond that? I’ll just start things off there. I think what we really want to emphasize with this update is that, the focus of your efforts shouldn’t be on the sliders. You should give them some thought, but I really don’t see this step needing to take more than a minute of your time, after you’ve played around with it and seen the impact. Maybe you’ll start to build one set of settings and go back and do a different one.
Andy Baldacci:
You shouldn’t be focused on this. Where I think the bulk of your efforts should be is on the projections and on really dialing those in. It’s not that we don’t believe SaberSim’s projections are good enough. We will put them up out there against anything and we are always improving our models. The point is, though, that an algorithmic model, a program is never going to be able to be perfect. It will never be able to capture all of the nuance that exists in sports. The projections matter a lot. While, again, we think ours are good and a very strong baseline, you do want to spend your time seeing where we may be off. Maybe we didn’t account for some late breaking news in the exact way that you think we should have or maybe we just missed it or whatever it may be.
Andy Baldacci:
Those are the areas where I would focus your limited time. This update is really to give you more time to do that kind of things because we understand that a lot of us have jobs. We’re not pros like Max and Danny and we’re trying to balance this with everything else out there. While it will always take work to win and we do always want to stress that even though we’re improving our models constantly, we do expect you to keep tweaking them to fit the way you see things. This is really just our effort to help give you more time to do those things.
Andy Baldacci:
I guess in my rambling way, the projections are very important. We do think ours are good enough but we think it’s important to always do what you can to try to make them better. Is there anything you guys would want to add to that?
Matt Hunter:
Yeah. I mean, I think, one other thing that you can consider, I agree that the projections themselves are really important. Just kind of adding on that, we’re a model and any projection system is going to be basically a model and any model is going to have biases or flaws and you want to always keep in mind and keep an eye on what the projections look like. See if you’re seeing consistent bias towards or against certain players or teams. It’s okay to account for that.
Matt Hunter:
Then, the other thing is that, there is also work that you can do after you run the build because one reason that we designed it so that we build a pool of 1000 plus lineups is that those are all, in theory, viable lineups that you can play in a tournament. Just hitting build and then downloading the lineups that you get at the end maybe sorted by Saber’s score sorted by projection. Those are going to be good.
Matt Hunter:
There are ways that you can adjust your exposures if you want to limit the amount of risk that you have or lower the exposure you have to some players. You still want to do it in, not scientific, but in a somewhat objective way. You don’t want to just lower. If you’re just lowering the exposure to some player because you think that they’re not going to do well. You want to adjust the projection first rather than messing with it at the end. I think there are also ways that you can use the tool to adjust your exposures in a way that that might lower your risk without sacrificing too much value and other things that you can do there.
Matt Hunter:
I think that’s just one other way. I know Max and Danny, I think, have focused a lot on projections and adjusting projections. Generally, I focus a lot more on the post build exposure adjustment than they do, but that’s just kind of a lesson that there’s different ways to use the product. That’s kind of like there’s the step one of adjusting projections. The step two of adjusting sliders which is the step that this update has really taken a lot of work out of that. The step two, you don’t have to do too much because you just choose. We have set the sliders up better. Then, the step three of that post build adjustments is a place that you can spend more time on now in terms of just dialing in those exposures.
Andy Baldacci:
Yeah, exactly.
Max Steinberg:
Yeah. I mean, I have something to add to it, if that’s okay.
Andy Baldacci:
Okay.
Max Steinberg:
I mean, I think if you’re playing Daily Fantasy to be able to just like click a button and spend out money then, that’s not like the way that you’re going to be a successful DFS player. It’s like, you want to do stuff that’s going to add value to whatever you’re doing, whether you have 5 minutes to do that or hours to do that. It differs. Sometimes, I’m making Daily Fantasy lineups and I do it in 15 minutes but I always try to do things that are going to add value to the lineups I’m building.
Max Steinberg:
SaberSim handles really well making great lineup construction for you. We are going to use our simulation data to construct really quality lineups for you and give you a lot of control over adjusting projections and how your actual lineups look in the post build process. What you can do, if you have even 5 minutes or 10 minutes or hours is, you can add value by changing an ownership projection, changing a player projection, changing a team projection. There are many ways to do it. You can do advanced stats, you can look at sports books, you can do a lot of different things.
Max Steinberg:
Adding that value is going to make your lineups more profitable. I think your goal as a Daily Fantasy player is going to be, how can I add the most value to the lineups and building before they get built? I would certainly focus on that. I would not focus on just 5 minutes before lock, clicking the build button with the default projections and just thinking, oh, I’m going to win every time. Obviously, that’s not going to happen. You don’t want to just have profitable lineups. You want to have the best lineups. To do that, you’re going to have to do something to add some value.
Andy Baldacci:
Yeah. That adding value is really what we think a lot about as we’ve been building SaberSim. It’s what kind of started this entire thing is that, when you look at traditional optimizers, there’s zero value in me telling it to build a stack, like that’s obvious. There are so many things you have to set that it’s just dumb. It wasn’t built to understand what goes into a high upside lineup. I have to go in and set all these rules and build all these groups and do all these things that it should be able to do for me, but it doesn’t.
Andy Baldacci:
We have done our best to automate the things that we can and to put into the program tools that take away that truly busy work that is not adding value so that you can focus your time in those areas where you do have leverage. It’s going to be different for every person what that leverage is. Max and Danny, I think, each really love digging into the stats and building up their own models and work on the projections. Whereas for Matt, he likes to get into exposure management, all of that. That might be something entirely different for you. With this update, we’re basically just saying, you’re not really adding that much value in the majority of cases by messing around with the sliders.
Andy Baldacci:
We have put out an update that takes in consideration basically all the variables that exist for this kind of problem, and put in place very smart defaults for each of those so that this is one less thing that you have to worry about. It’s not to, again, give you that money printing machine where you just click a button and win all the contests. It’s to give you time to put it into the places where you can add value. Max and Danny have a lot of videos for the major sports on what they personally do to improve the projections, to work on other things like that. We’re going to be keep coming out with more of those. We’ll have Matt in there and talking more about his exposure management and this and that.
Andy Baldacci:
Today, what we’ve really been trying to do with this update is just take that bullshit busy work that none of us need to do. There’s no value there. Take that off your plate and do a better job than we could on our own. There’s still more work to do and it’s always going to take work to win, but we think this was definitely a big step in the right direction. We’re coming up on that hour mark. I’m going to try to go through some of these other questions a little bit quicker and see what else I can get to in the chat.
Andy Baldacci:
One, we had a while ago from Nick was, does Giant Squid still differ a little bit on the slider settings? I know he like lower standard deviation some cases than the norm? The answer is, yeah. I hope at the end of this, it kind of is clear that you can differ from these. I wouldn’t flip them and go completely opposite from what we have as defaults. For your personal preference, whether it’s risk tolerance, whether it’s just your style of play, whatever it may be, going a couple ticks up or down from the defaults is fine. It’s not going to make or break your lineups. Play around with it, see what fits for you. If you’re unsure what to do, leave the defaults as is. B. Henson asked, go for it, Max.
Max Steinberg:
I was just saying, I think that’s especially true with ownership fade because I’ve talked about this on other videos, but I think ownership fade can be … there are some pitfalls with it, especially if either your ownership or if you sort of like to adjust projections a lot. If you like to adjust projections a lot, what ends up happening is ownership fade can accentuate the ways that you might have a personal bias or maybe the projections or bias. Because let’s say someone’s projected at 1% ownership and you have them at like 10X value. Chances are that’s wrong. I mean, maybe you’re a savant and that’s right.
Max Steinberg:
What it’s going to do with you if set that ownership fade to really high, it’s going to make it so you just get that player at probably at 100% or whatnot …
Matt Hunter:
Yeah. You’re like double counting it.
Max Steinberg:
Right. You end up, I think, accentuating some mistakes that you might have made. That’s something that, I think, in general, it depends on how much you’re adjusting projections, how contrary in person that you are. That’s something that we default it to pretty low and that’s sort of the reason in general.
Andy Baldacci:
Yeah. Next question is, we’ve got it couple times, it’s basically about limiting player pools, is it counterproductive? Is it something I should do? Should I just trust SaberSim and let it do its thing? You guys can jump in but I think ultimately, this goes back to what Max was talking about with adding value. I think if you are stuck in the more traditional mindset of an optimizer of I need to tell it exactly what to do or it’s going to give me just bad lineups, you don’t need to get a very small player pool for SaberSim to build you good lineup.
Andy Baldacci:
Just off the bat, don’t worry about that. If you find spots where you think we’re off on somebody and rather than just trying to get the projection exactly right, it’s easier for you to just remove them. That’s fine. That is a spot where you can add value because there are going to be some times where maybe we don’t accurately predict how someone’s going to play coming back from an injury, maybe whatever it is that we’re not accounting for you, you disagree with, remove them from the player pool. Don’t feel like you have to cultivate this perfect player pools because if there’s someone that doesn’t have any upside, we’re not going to put them in your lineups. That’s all I would say there. Max, did you want to add anything?
Max Steinberg:
Yeah. I’d just say I think actually, if you have the time to do a build before you do your final build, I think, you definitely should not exclude anyone. Because seeing who you get on a build is a great way to find out, okay, who do I need to look at? Because you say, okay, I’m getting this person 40%. I was not even considering this player. Okay, let’s look at the stats that were projecting them out and see and if you think this might be off or look maybe at stats on NBA or some other data source and see, okay, is this off? Or right? Or do I need to adjust this projection?
Max Steinberg:
Then, if you’re doing your final build and you’re like, I do not want this player. Sure, exclude them, that’s fine. I think, for the process before you make that final build, I would not exclude anyone because I think that’s a good source of information to see who is popping up.
Matt Hunter:
Yeah. Adding on top of that, I think it’s probably more valuable to exclude players that maybe you’re just getting 1% of or just kind of a one off, not necessarily blindly but when you’re doing your final build, I think it’s okay to look through your player pool and see, oh, maybe this player is coming up because I have a high smart diversity and maybe he did really well in one sim. He’s too risky of a player to put into my exposures, and you want to remove them.
Matt Hunter:
I think that’s totally valid, but you should have reasons for doing that. I agree that just excluding a player because you don’t like them or they burned you in the past or something. That’s a really bad approach and a way that you’re not winning the long-term by doing that kind of thing.
Daniel Steinberg:
Yeah. I think maybe just a side note, a legitimate reason to exclude players if they’re questionable or you’re not sure they’re going to play in baseball, it’s like that the lineup hasn’t come out and you think maybe this guy is not that likely to play. I mean, exclude them. Or if they’re questionable, exclude them. That’s a really actually good reason to exclude.
Andy Baldacci:
Yeah. Especially if this is almost like, in my mind, like a requirement. If you are playing NBA and you’re unsure or you know you’re not going to be available closer to the lock, exclude players that are questionable. It’s not a guarantee, crazy things still happen and there are still late scratches. By excluding all the questionable people, unless it’s a small slate and there’s a lot of questionable people like you’re probably not giving up too much. It is much better to do that than to have someone in your lineup that gets zero.
Andy Baldacci:
That’s an edge case sort of but for NBA, it’s really not. It is something to keep in mind is that’s when to do that. The question we had earlier from B. Henson is, “If in multiple 150 max tournaments, should we use the top 150 in both or build 300 lineups?” I think the guys will have opinions on this. I feel like it differs by sport, by the size of the slate, contests and so on. I think, if you’re doing like a pre-max contest, a single entry contest or even a 20 max contest, using separate selections of lineups in each of your contest is less bad because you can have a large number of profitable lineups.
Andy Baldacci:
By diversifying it, you can lower your variance and it’s, I guess, probably a fine trade off depending on your risk tolerance knowing that if I pick lines 1 through 20 and put that in Contest A and lineups 21 through 40 in Contest B, those second set of lines are probably slightly lower ROI than the first. It’s not that big of a deal. I think when you get to 150 lineups, that’s where you’re giving up more especially for something like basketball. In hockey or baseball and bigger slates, it’s probably less of a big deal. That’s just generally how I think about that. Do you guys have different opinions or what do you guys think?
Matt Hunter:
Yeah, man. I think it does probably depend on your settings as well. Obviously with this update, there’s not as much focus on the slider settings but if you have more tighter exposures where you have really high exposure to certain players, you have a lower smart diversity. You have smaller player pool. Your 151st and 300th lineup are not going to be as different in terms of value from the first one as it would be if you had a much more spread player pool and higher smart diversity, where as you get farther and farther down the list, you’re giving up more and more value.
Matt Hunter:
I think it does depend on that as well. I would have generally agree that for single entry and three max and that kind of thing, it’s probably okay to spread them out. If you’re putting in 150 max to a bunch of different contests, we would probably want to just stick with 150. Yeah, I think, it kind of depends.
Andy Baldacci:
Do you hear that, Danny?
Daniel Steinberg:
I mean, I think it’s just a matter of risk tolerance, I guess. I think there’s probably not that much of a difference and that I just normally just go with 150.
Andy Baldacci:
Yeah. Let me see if there’s … I’m sure there’s some …
Daniel Steinberg:
Hey, there’s a question by someone, [Don Poletti 01:02:08] said, “You offer stacks but I’ve noticed that Max and Danny are against builds.” I think [crosstalk 01:02:14].
Andy Baldacci:
Yeah. I saw a few of his other questions. Don, let us know if we don’t get this right but I think it’s basically just saying, should you force stacks in? Is that something you should do? Generally speaking, Max and Danny, from watching all of your videos myself, I don’t think you guys are saying that stacks are bad. You’re just saying that traditionally, the correlation will take care of it for you.
Daniel Steinberg:
Right. I mean, I think the beauty of SaberSim is you don’t have to pick all the stacks and jerry-rig this whole part of the builder, just right to get the perfect stacks. We have a simulator. The simulator is taking into consideration the correlations between different players. When you build lineups, the simulator is going to do a much better job making stacks than you could by just picking it manually. That’s why, I think, we don’t recommend prime to actually manually set different stacks you want to use. Just trust the builder to make the best stacks, given whatever projections you’re using.
Andy Baldacci:
Matt, did you want to add something?
Matt Hunter:
Yeah. I think I agree with that in a sense. I think it’s okay to add stacks. I wouldn’t say that you shouldn’t. If we didn’t want people, if we thought it was really bad, we wouldn’t include that in the site. I think it’s okay but I do think that you should keep in mind that if you add a four stack to your build, you probably want to decrease your correlation a little bit because, if you don’t have a stack and the correlation is kind of building those naturally. If you add a four stack then, you’re adding the stack and then we’re doing correlation on top of that stack. You’re getting more correlation than you would normally.
Matt Hunter:
For baseball, if you don’t have any stacks, you might get some five man stacks, you might get 5-3, 4-3 or whatever. You’ll get kind of a wide variety. Then, if you force them in and then at correlation, you’re just going to have like only maximum, 5-3 stacks or whatever. If that’s what you want, that’s okay. I think you just keep in mind how the stacks are interacting with the rest of your settings.
Andy Baldacci:
What were you thinking, Max?
Max Steinberg:
I’m just saying, I also think another reason is, we actually allow you to do the same thing in the post-build process. In our sports, we have the stack types. We have filters by stack types and you can set a min and max exposure to the stack type. What we’re saying is, don’t limit that. I mean, you can do this and I think there are some reasons to do this. If you don’t limit it before we say a percent build to lineups, you can limit it after SaberSim builds you lineups.
Max Steinberg:
That allows the builder to do all the things that it’s supposed to do. It doesn’t force it to do something that might not be the optimal way to do things. I think we just have that option later. Whereas, I think, a lot of optimizers just don’t have that option later. Either you have to decide right then and now, do I want to build lineups that are four stacks? Or do I not want to build lineups that aren’t four stacks with us? You can just say, okay, well, I see this like a 3-3-2. Let’s say in baseball. I kind of like that.
Max Steinberg:
There’s like a 2-1 and then no stack. I don’t like that lineup. You’ll zero that and say, I don’t want this lineup in the post build process.
Matt Hunter:
Yeah. It’s kind of similar to the one I was saying before about moving players from your pool is, you can do that build and then in the post build step, you can look at your exposures, not just players, but also stack types and make those little adjustments. If you have one or two lineups that are stacked. If you don’t like, just remove them and that’s okay and there’s nothing wrong with dialing in your exposures afterwards.
Andy Baldacci:
Yeah. To wrap up, we’ll do one more question after this but I just want to add that a lot of what I’m getting out of this, and I hope viewers are as well is that there’s no right way to use this. There are multiple ways to approach it. Oftentimes, Matt, you have a different approach than Max and Danny each do. I’m sure Max and Danny each have several approaches on different things as well.
Andy Baldacci:
That all comes down to the fact that through our simulation process, through our lineup building process, we’re able to build smarter lineups. We’re able to build high upside lineups. We’re able to factor in all of those variables that matter so that we give you that strong foundation and there’s a lot you can do from that. You can set things up front. If you’re more comfortable with that, you can adjust them down the road, if you’d prefer. There’s not one right or wrong way of doing it and it’s trying to just move past that.
Andy Baldacci:
Then, the last one, the last question we’ll answer is and I’m really curious to hear Max and Danny, you guys have thoughts on this, is there a way to build based on the cost of the tournament? The one thing I will first say is that, generally speaking, the higher buy in tournaments have a smaller number of entrance and a smaller entry limit. That’s already going to be accounted for in the sliders. With that in mind, are there other changes you guys would make for some of the higher buy in contest? How would you approach that?
Daniel Steinberg:
Yeah. The first thing that comes to mind or maybe the only thing that comes to mind is just in the higher buy in contests, you have sharper players and normally see ownerships are a lot more concentrated. I would probably maybe move the ownership setting to the right a little bit for a higher buy in contest than for a lower buy in.
Andy Baldacci:
Great.
Matt Hunter:
You could also adjust your ownership projections themselves, if you think, oh, I’m going to enter this high dollar tournament and I think this guy is going to actually be twice as own just because it’s sharper, then change the ownership. Because ours are kind of based on sort of the average tournament. If you have a specific tournament in mind, and you know ownership is going to be different, adjust the ownership projections.
Max Steinberg:
Yeah. I will also add on to the point. I mean, I think a great example is if you played NBA last night, Nasseri, if you looked at the lower buy in contest he’s about 20% owned. The higher buy in contest, he was about 40% on DraftKings. The reason for that is that with people who are like late news type of players, usually the savvier players will get on that a lot better than in lowering buy in contests. If you have sort of something that’s a late news thing that the players somehow has become the chalk, you’d have to have … I hope if you’re playing high stakes NBA, you have a lot of awareness of how players choose players and awareness about what’s going on in the slide.
Max Steinberg:
Because if you are, then when there is that late news, you could maybe jack up a player’s ownership a lot and use that ownership slider to a higher setting. I think that’s really the only thing is, the ownerships are going to be a little sharper and they’re going to be a little more concentrated. I think you also can just trust it. There’s no difference between a higher stakes contest other than that. You also can just trust what you’re doing, your process and SaberSim and you’re going to do well.
Andy Baldacci:
Yeah. To close things out, it goes back to what we were saying about adding value. With this update to the sliders, we’ve made it so not that it’s impossible to add value there and not that there’s nothing left for you to do and to adjust, but that it’s probably not the highest leverage spot for you to focus on and the efforts that you’ll have there to get more value, to add more value to what we’re doing are going to be pretty quick.
Andy Baldacci:
I think some of the things we’ve talked about have shown how and when to make those adjustments. Really, I hope this update makes everything a lot easier for you so you can focus your time on the projections and the exposures. That’s really where we think the most value is to be had. If you guys have questions at all on any of this, or anything else, you can always shoot me an email at [email protected]. You can hit us up in Slack. Reach out to us on Twitter, wherever it is, you want to get ahold of us, we’ll do our best to get back to you.
Andy Baldacci:
This is a fun one. Hopefully, you guys got a lot out of it. We really appreciate it. We’ll talk to you soon. See you guys.
Andy Baldacci:
How should a new player, someone who’s new to DFS, just getting started with things, how should they try to develop their DFS process? Because that’s one of the things that everyone talks about is build your process, focus on the process, and this and that, but when you’re just getting started out, it’s not really even clear what that actually means. So if someone is committed to the game and does want to get better and really build out that process, how would you recommend that they do that?
Max Steinberg:
Yeah, I can take this one. I would say there’s a couple of things that just come to mind immediately when you’re talking about this. I would say one is just reading up, watching videos. We have a lot of videos with SaberSim on our YouTube channel, reading strategy, just try to find all the content you can and just eat up that content, because there’s good information out there. There’s obviously a lot of smart people who play Daily Fantasy, and the info is out there and you can always learn more and more. I even … I’ve been playing Daily Fantasy for five or six years and I still will read articles or just when someone tweets something out who I follow, read and try to learn something.
Max Steinberg:
I think also in that same vein is looking at … especially if you’re entering contests that good players are in, maybe even single entry contests or cash games, look at those players’ lineups. Look at the players who are entering thousand dollar heads-up, thousand dollars GVPs, look at their lineups in these double ups or single etry GVPs and just see how they’re different from yours, what players they’re using that are unexpected, and think about why, because I think there’s a lot and you can just learn from those players. They’re good. They’re playing these high stakes for a reason, and I think obviously they’re going to make some plays that you might not understand initially, but you possibly can learn something from.
Danny Steinberg:
Yeah, I could follow up on that too. I think one thing that can be really a red herring is just paying attention to the results of each slate or each week. There’s a lot of variants in DFS, so a lot of times when someone does good or bad, you’re not necessarily going to know whether that was luck based or that was something that was predictable. So I think … I mean one thing that’s really helped me and Max both get a lot better is … our learning computer programming and data science is something that’s an integral part of our process. So I think really that’s a way that I’d really recommend if you really want to get good at DFS, that you have to have a data science aspect to your research.
Andy Baldacci:
And or that, Danny, do you think that is a necessity for all players? Or is it something where … I think there’s got to be a line you’ve got to draw somewhere, where if you want to compete at the absolute highest stakes, you do have to –
Danny Steinberg:
Yeah, totally.
Andy Baldacci:
– treat it like a full time job and probably look at some advanced analytics, but if you’re trying to make some money on the side playing the under $3 contest, do you think it’s … how far do you have to go?
Danny Steinberg:
Yeah, I mean I think there’s a lot of really good research out there and a lot of good people who have done a lot of good research, and I think for someone who maybe doesn’t want to do all the research themselves or doesn’t have the ability to do that research themselves, there’s a lot of people you can follow on Twitter. There’s people who have written articles everywhere that have really interesting thoughts on Daily Fantasy strategy. There’s a lot to learn from just everyone on the internet, and I think you can learn a lot and get better just by learning from other people.
Max Steinberg:
Yeah, I’ll piggyback on this as well. You don’t really have to … in order to win at Daily Fantasy, especially for using a sophisticated line of [inaudible 00:03:59] we offer, if you’re taking advantage of correlation, you’re taking advantage of with upside, all you really need to focus on is getting better projecting or getting better at doing ownership projections, and I think if you can focus on making sure that when you make an adjustment to say a person’s baseline projection or ownership projection, you’re adding value in some way that’s going to be a win, and if you have enough wins every slate when you’re adjusting projections or ownership projection, that’s going to make you a profitable player, and so if you can focus on where am I sure that I can add wins to these adjustments, I think you’re going to do a lot better, and you’re going to be profitable, especially on these low stakes contests or these single entry contests or so on and so forth.
Andy Baldacci:
And I think that actually is a good way to kind of transition into the next question, and it’s something that did come up and chat and also someone had asked this over email. It’s just come up a lot, and it’s kind of the idea of if SaberSim or whoever you’re using for projections is putting out solid numbers, when should you make adjustments here and why? What are the things that are going to add value to that? You talked about adding wins and finding ways to really kind of improve upon the projections, but what types of things specifically are you looking for there, Max?
Max Steinberg:
Yeah, so I would say anything that uses very advanced statistics or new statistics … SaberSim mostly works off of historical data. So if there’s something new that is something that’s like stats.nba, which uses cameras to get all these advanced statistics that we don’t use, something that has to do with expected points added or complete … there’s someone who is named Ben Baldwin on Twitter who tweets out this completion percentage above expectation stat that is really helpful. Things that are essentially not just plain historical data, you can use these sources and they can add something. I mean I think there’s a fine line, because I think sometimes people can get into “advanced statistics”, and you end up just using something that isn’t predictive and not helpful, but especially in NBA, I think this is just coming to mind for me a lot.
Max Steinberg:
It’s just stats.nba. There’s a lot of stats out there. People don’t really use it that much, but there’s a lot of really detailed, really good stats because they’re based off of actual cameras, video taping the games, and actually getting movement and player position, so on and so forth, and so if you use something like that and you find something that you think is helpful, something like [inaudible 00:06:36] field goals, drives to the basket, something, have a theory about it, try to look up and see if this theory may or may not have some weight, and use something like that. I think that is certainly going to add something to SaberSim’s model.
Danny Steinberg:
Yeah, I think that’s a good point. I think that sort of very nuanced information can be really helpful if you’re interpreting it for some something that’s a real deep dive into what’s going on with a team from a very advanced statistical level, that you can make sense out of. I mean some of the easier stuff to add is just providing the projections with context that’s useful. So for example, basically every projection system is relying on historical data to project out the future. Now has a player been playing through an injury? Did they have a game where they left halfway through the game? That’s something that a model’s not really going to know. So adjusting for that sort of context. Sometimes teams will decide to drastically change the role of someone on a team. Sometimes a team will be like, we’re committed to this guy at running back and we’re really going to try to establish the run that happened this year with David Montgomery. So I guess reading through the news and understanding where a team has decided to drastically change direction in some way, that can be really useful in understanding the context of the historical data can be really useful.
Max Steinberg:
Yeah, I think a great example of the injury example is Adam Thielen last week.
Danny Steinberg:
Yeah, a perfect one.
Max Steinberg:
Obviously, been playing through injury for four weeks, but we know historically he’s been a really, really productive player, and it ended up working out last week when he was back to his old self. This week he might be injured again, so who knows? But when you can catch those things where there’s a recent injury that’s sort of marring someone’s performance, you can end up just finding an easy way that someone’s under projected and most people are under projecting them and then they end up just being back to their old self, which is not hard to predict.
Andy Baldacci:
And Max, you touched on this a little bit, but when we get into advanced statistics, when we get into research, when we get into just kind of going over historical data, where do you think people can kind of fall into a trap and over value of the data out there? Max or Dan, either of you guys can jump in on this one.
Danny Steinberg:
I’ll handle this one. I think stuff with matchup can be really overrated. There’s just a lot of statistics where you can think that they would be predictive of the future when they’re really not, like recent performances, as an example. I remember one time a few years ago, Brett Gardner, the lead off hitter for the Yankees went on a streak where he had seven straight home runs in seven straight games, and he was the [inaudible 00:09:27] player every night, and everyone just thought, “Oh, this is never going to end. He’s going to always hit a home run. He’s just really keyed in,” but in reality, very extreme things are going to happen because of luck alone. So just because someone’s done really well in a certain situation or in a certain context doesn’t mean that that’s necessarily going to be predictive of anything in the future.
Danny Steinberg:
I mean another example is how a team performs against the spread in different contexts. That’s something that a lot of touts use in the sports betting industry. It’s just extreme things happen and those extreme things aren’t necessarily going to be predictive of something extreme happening in the future, and another obvious example would be steals in basketball. Someone has a 10 steal game, are they likely to have a lot of seals next game? No, of course not. Good steals are the most unpredictable stat in all of basketball. They’re very, very random, so don’t pay attention to stats that are really, really noisy, basically.
Max Steinberg:
Yeah, and I’ll just add one more thing. I think especially in football, I see this, is people making game script assumptions based on Vegas lines. I think when you’re … especially if you’re playing cash games, I think that can be valuable, but in tournaments, how the game script actually goes, what I found is random enough to the point that unless the spread is just giant where it’s an almost certainty that a team is just going to be leading the whole game, i.e., like a team is a 14 point favorite, but even we see that, the Dolphins versus Patriots, I think the Patriots were a 17 point favorite in week 17 and they lost.
Max Steinberg:
So anything can happen, and I think people just make these assumptions with a five point spread that this team’s going to be leading a whole game, and the game script can really go anywhere, and luckily for us, SaberSim accounts for that, so we don’t need to worry about that with our smart diversity, taking into account small [inaudible 00:11:30] of simulation, which I talked about in another video. I won’t get into it, but that’s something that a lot of people, they try to think, “Oh, how’s this game going to go? How’s this game …” they’re five-point favorites. This is just going to be a run heavy game, and it’s like, no, that is not what’s going to happen. You’re going to see it today, because there’s a lot of teams that are seven point favorites, and one of those things is probably going to lose. So making over assumptions on that I think is a huge mistake.
Andy Baldacci:
And this I think ties in pretty well to kind of the last question we have about just kind of building that new process, and it’s saying when there is so much noise out there, when there are so many statistics that aren’t truly predictive or aren’t predictive without a much larger sample size, how can you objectively judge whether or not your processes actually improving or not?
Max Steinberg:
Yeah, I think it depends on the sport, because you have to think about how much variance is in a sport, and I can just tell you, baseball is one of the higher variance sports. Football is a pretty high variance sport. Basketball is a low variance sport. So if it’s something like basketball, which is obviously going on, we’re in the middle of season right now, so that’s something probably people want to hear about, but with basketball, you can actually look at the box scores and say, okay, I adjusted this per person’s projection, A, because I thought it was going to do really well, and where obviously there’s some variance statistics there, if you adjusted it up, you should’ve got higher minutes than SaberSim expected and you should have had higher production in terms of fantasy points per 36 minutes or just the [inaudible 00:13:12], whatever you thought, it should basically happen.
Max Steinberg:
There could be variance in outcomes as the person gets into foul trouble, they shoot poorly, but you have an idea that okay, this actually worked out, and sometimes that’s not the case, but you can actually get a lot from what the actual results are. In baseball, you can’t do that at all. The results are so random, and we can use that to our advantage, because we can stack some of the low on teams that have high upside, and that’s going to benefit us, and I think with a sport like baseball, you have to think, okay, what is the … you have to actually theorize more. You have to say, okay, why do I think this is going to be the best stat? Think about what are sort of the under the radar stats. Maybe try to think of what pitcher might struggle, look at weather staffing. We take this all into account, but sort of think about it that way and maybe make some adjustments to our ownership projections if you’re using SaberSim and want to go that route, and then also just look at, again, what good players are doing and try to assess things that way, because if you’re doing something right, probably really good tournament players will be doing it with you. So I think those are the things you have to keep in mind.
Danny Steinberg:
All right. That’s well said. I just wanted to add just a few things. One is, I think you’re definitely right with what you’re saying about basketball, but I just want to emphasize if you’re judging your process based on the results of a slate in baseball, that’s very, very wrong. It’s so random that if you’re saying, “Oh, this team did well and I had this thought about them, so that must be … or they had this factor in their favor, that must mean that this factor really matters.” Terrible. Exactly. Whether someone performed well or badly in a baseball game is very dominated by luck. So it’s kind of like poker. You kind of have to have some theories about the game and sort of believe in yourself and your understanding of the game and whether something makes sense at a theoretical level, or you have to do some really high level research.
How do you handle downswings?
Andy Baldacci:
Assuming you do have faith in your process and long term you have results to back it up. You understand the theoretical underpinnings of it. How do you still deal with those down swings? How do you deal with getting through the burnout of yet another slate where you’re getting crushed without second-guessing yourself when you do feel like you have a solid process? How do you navigate that kind of mental side of it when things aren’t going your way?
Danny Steinberg:
I mean, it’s very difficult is really the answer. I’ve been gambling for a living for like 12 or 13 years and it’s never easy when you go on a downswing, but you have to keep things in perspective. If you’re doing a multi-entry GVB tournament’s in Daily Fantasy, you’re going to have times where you go on a downswing. It’s just going to happen every year. So you have to not get to have your emotions get too wildly swingy on winning a whole lot and getting on winner’s [inaudible 00:01:21] and thinking you can predict everything perfectly to losing a lot and thinking everything I do is wrong and I have no idea what I’m doing. I think there’s a middle ground there I guess that you have to keep perspective on.
Max Steinberg:
I mean we used to play poker professionally, it’s a lot easier than poker. Poker, you go on a downswing and you have to really be in the game and playing and make decisions on the fly. And that’s really hard. I think in the beauty of Daily Fantasy is there’s always another day. You have time to prepare, you have time to get yourself in a good mindset. And I think just make sure you’re getting yourself into a good mindset every day and you can always go on [inaudible 00:02:06] and keep being really mad then night of. But just make sure that you’re not mad two hours before a lock in and being emotional. When you need to make good lineups, right.
Andy Baldacci:
Yeah. No, that’s a really good point because in poker, if you lose a huge hand, you’re getting cards dealt immediately after that, and you have to make good decisions immediately afterwards. And if you’re playing against the same player that just stacked you on a bad beat you have to keep your emotions controlled. Whereas in DFS, there’s a lag in between results from one day and starting up the next, which doesn’t necessarily make it easy, but it does give you more opportunities to control your emotions and keep them in check there.
Max Steinberg:
Yeah.
What mistakes do people make when switching to SaberSim from a traditional optimizer?
Andy Baldacci:
I thought it’d be valuable to talk about just some of the mistakes people make when using SaberSim and specifically, when using SaberSim and coming from a traditional optimizer, what some of those common mistakes are. Because fundamentally, traditional optimizers and SaberSim function very differently, but it’s not always easy to grasp. So I wanted to ask Max, I guess, if you could kick it off, just what common mistakes do you see people make when switching to SaberSim from a more traditional optimizer.
Max Steinberg:
Hey, yeah. I get a lot of Twitter DMs about this, so I’m also acutely aware of this. I mean, everything that we do with SaberSim is based on our simulation data and it has to do with using correct slider settings. We have our Correlation Slider. We have our Ownership Fade Slider and we have our Smart Diversity Slider. That’s basically the complete key to our entire process. I see a lot of people who will DM me on Twitter and they’ll show me their sliders. And they’ll be like, “How does this look?” It’ll be something that’s outrageous, like it has nothing to do with the preset settings and it shows a lack of understanding the sliders.
Max Steinberg:
So I think one thing that is really important is educate yourself about what SaberSim actually does. There’s a lot of videos that Danny and I have made that explain in detail. You can ask us on the Slack channel. We’re happy to answer any questions, especially now if you’ve created this whole thing … knows it pretty well. So I think if you want to get better at using SaberSim, you have to understand what each slider does. Because once you have that understanding, I think you’ll find actually that the build process begins way easier and you’re not actually making as many adjustments post-build. Which I think is another big mistake that I think a lot of people do, is they just try to control too much as they try to set too many settings before they build. After they build, they’re adjusting exposures too much.
Max Steinberg:
The thing is is all you should be worrying about is making correct projections, adjusting projections so they’re as correct as possible and adjusting the ownership projections so they’re as correct as possible. If you do those two things well and you understand our sliders and set them correctly, you’re going to make really good Lineups, especially in sports work correlations, very importantly NFL, hockey, baseball, et cetera.
Max Steinberg:
Things where that’s really important, you’re going to make Lineups that are certainly better than Lineup optimizers out there, right? So focus on ownership projection, understand the sliders, make sure they’re set right, to the right tournament that you’re playing and you should do pretty well.
Danny Steinberg:
Yeah, I think, just adding to that or I mean, I think one mistake you can make is just not trusting the solver to do what’s optimal. You can try to hack it to do what you want. You should trust it to do the right thing based on whatever projections you have. If the solver’s not getting someone that you really want to get and you agree with his projection, but you feel like you should be in Lineups, just no, he shouldn’t be in Lineups. The solver knows better than you. You got to trust it a little bit.
Andy Baldacci:
I think that touches on a good point where it’s not as though we are saying SaberSim is perfect. It’s obviously not. No tool is going to be. Even though we are spending a ton of time and always will be improving it, you do have to ask yourself, “Okay, what is the tool telling me because we’re not functioning as a glorified spreadsheet, which optimizers basically are. We’re actually considering things upside.” So there is insight in what SaberSim will give you. Like you said, Danny, if you do agree with the projections, if you agree with the Ownership, the point projection, but you disagree with the exposure that you’re getting in your build, there’s some disconnect there.
Andy Baldacci:
Because we’re saying, based on everything you’ve given us, this is how much you should get that person and if you don’t disagree with their variables going in, then you really check that assumption.
Danny Steinberg:
That will vary on all of them.
Andy Baldacci:
Again, that’s why all of it goes back to those projections because that’s where you do want to make adjustments and you do want to improve and build on what we’ve given you. But once everything is dialed in, it’s not as though you just set it and forget it and don’t make any changes from there. But you do want to at least listen to what we might be trying to say.
Max Steinberg:
Right. I think using the post-build adjustment features that we offer should be making small changes in, we call it quality control, right? You’re making small adjustments to make sure you get a little more, maybe get a little more of this flair, a little less of this flair, but you shouldn’t be changing the entire build with just post-build. You should be adjusting projection and you should be adjusting Ownership projection or you should be adjusting the sliders because your sliders are not correct for the GPP you’re playing or the cash game you’re playing. But you don’t really need to be adjusting things and you shouldn’t be adjusting things too much, trying to basically rig your Lineups to get them right, because you don’t need to do that with us.
How do you build lineups for cash games?
Andy Baldacci:
When going from GPPs to cash games, what adjustments are you making to your approach? How are you changing the way you’re thinking about the entire process when you are approaching cash games?
Max Steinberg:
I can take this one.
Danny Steinberg:
Okay.
Max Steinberg:
Yeah, I think, again, no matter what tournament you’re playing, your focus is always on making the correct adjustments to projections and the correct adjustments to ownership projections. And, The difference between the cash games view is just what you’re valuing. Like how you’re valuing correlation, how you’re valuing upside, how you’re valuing ownership fit, right. So with cash games you actually want your players to be reverse correlated. If you don’t want to go that route because that is actually going to sacrifice some projection and that way in order to get that…
Andy Baldacci:
Can you just ask a little bit Max on what that actually means when you say you want reverse correlated players?
Max Steinberg:
What does it mean to you basically want your, I mean, so the goal of a cash game is to, so cash game is head to head,50/50 or double off, right? So, and head to head, you’re trying to just beat one person’s line up and the double up or a 50/50 you’re trying to beat the [inaudible 00:01:19]. It doesn’t matter if you’d be good at if you’ve got 1000 out of 2200 person field or are you at first, right? So maybe a lineup that you’re building to get first is really, really stupid, right? Want to line up that as much of the time as possible. Your going to be that top half or beats someone [inaudible 00:00:01:40]. Right? And so with correlation, when you want a really high positive correlation means you’re trying to build line ups that all your players are running, going to do well together. They’re correlated, their performances are correlated. That’s what it means is they’re going to do well together.
Max Steinberg:
But when you’re playing a cash game you want it so your lineups basically, no matter what happens are going to barely be above that 50% threshold. And so you want players that don’t do well together. You want them to sort of one player does well but does another player does badly but they sort of like even each other out and you sort of just want to line up. That’s going to get a point enough points that’s just going to be at that field and so when I say negative correlation, I’m just mean that you want players that are just not going to perform well when they are performance while in football is going to be sometimes two opposing running backs.
Max Steinberg:
This might mean using a defense and a wide receiver on opposing teams, it’s you’re going to be doing things that just obviously would not be good if you’re trying to get a 200 point NFL score but are going to be really good if you’re trying to beat 50% of the field. But I think in the process of saying right is now all I do from week to week, if I’m adjusting projections and try and get good ownership projections, right when I’m building for a cash game, it just has to do with said again thing. This slide is settings correctly and that they’re going to be negative correlation. That’s not going to value ownership and it’s not going to value upside. Right? It’s just going to be let’s get good average rejection, negative correlation and be with the field basically. It’s fine with me.
Andy Baldacci:
Danny, do you have anything to add to that?
Danny Steinberg:
Yeah, I think your approach really should just be the same as to make the optimal projections and then change the sliders and settings based on what kind of game you’re playing in. So I think, I think the process is overall mostly the same. You’re just really, as Max is saying, you’re trying to get your projections right and trying to get the ownership projections right and then adjusting the sliders based on the type of game you’re playing.
Andy Baldacci:
And then does this apply? Like do you guys treat all cash the same? Is it? If you’re playing head-to-head, if you’re playing a 10 person 50/50, or planning, one of the big 50/50 says multi-entry, like are you playing all of those roughly the same way, like with the literal same lineup? Are you building lineups differently? Are you playing different lineups and just like how do you think about the different cash contest types?
Danny Steinberg:
I just use one lineup. I’m curious if Max would.
Max Steinberg:
I know I, I’d say the same thing. I think you know you can get some more new op of you’re doing it triple op or quintuple op or something like that, but if we’re talking about 50/50, double ups, and head to heads and you can do different lineups, but that just means you’re going to sacrifice some ROI from going off the best line. All right, fine. I mean people do that, they hedge their risk, but if you’re new to [inaudible], I probably wouldn’t recommend it because you don’t want to risk possibly putting in a lineup in that is minus CV. In order to headdress grad, you want to make sure that the labs you’re putting in are going to make you money. So I hedging in that way by paying for different lineup, six different cap lineups. You run that risk of putting in a lineup that’s just not a good one and so.
Andy Baldacci:
One thing that just came up in the chat that I want to touch on and it’s just kind of common was so change sliders and when all the monies, and I know we’re not heavily saying I think they were joking, but it is something good to touch on is that we, I guess from my perspective is I think almost too much focus is given to getting the sliders exactly right. I think as long as you’re within general ranges for the game type, that’s all you need and you can kind of dial in and beyond that to your own personal preference. But there’s not like one exact set of size that this is perfect for this contest every time it’s going to be based on risk tolerance in business. A lot of factors involved, but that is kind of the aside of it. The big thing really is dialing in the projections and we behind the scenes are always going to work on improving them.
Andy Baldacci:
But that is kind of what we are saying is the where the most value can be added is kind of spending your time improving that. And none of this is easy, but I think just because we are doing things a little bit differently by having those sliders, it can throw people for a loop where they kind of overthink that and that is the phase that it’s not that it matters the least cause it does matter, but it doesn’t require nearly as much time as the other phases do. So I just kind of wanted to just clear that up a little bit. Do you guys agree with that general sentiment?
Max Steinberg:
Yeah, of course.
Danny Steinberg:
Yeah, for sure.
Andy Baldacci:
Then the last thing on this topic I wanted to ask is do you think cash games are worth playing for the average player, and I guess that’s a bit of a blanket statement, but like do you think cash is profitable for all players not all players, but like a serious player for every single sport or how do you guys personally approach that?
Danny Steinberg:
I mean, I think one thing is one thing I really believe with SaberSim is our solver does a extremely good job taking into considerations all the factors that really matter in GPPS. So I think I’d really recommend if you’re using SabreSim to do GPP is you can definitely win at cash, but if you’re good at projecting players, you’re probably going to have a bigger edge of GPP than you are at cash. And SaberSim’s going to do a really good job building the best lineups with the projections that you have.
Max Steinberg:
Yeah. I also will piggy back on this. I think people get attracted to cash games because they think there’s less variance in them because they think oh well. Like a lot of the weeks, I’m just going to double up and some weeks I won’t. But if you’re making, you know 20 or 50 lineups and these low stakes do GPPs from getting a good spread, the variants are really, is not as high as one would think.
Max Steinberg:
And so I think the things that are attracting people to cash games actually can’t exist when you’re being GPP player and your ROIs are going to be higher in GPP is because of everything. Danny just said. SaberSim is built to make really good GPP labs. I think now people are attracted to cash games. They think it’s low variance. GPP is really aren’t that much higher variance and they’re going to be more profitable, which I think is the most important thing. Cause again, the worst thing you can do is put in a negative CV lineup into one of these tournaments. Cause variance doesn’t matter, then you’re just losing money right.
How do you pick your 3-max lineups?
Andy Baldacci:
You did have that video on single entries that covered a lot of the stuff there. We talked about more of it here. But, for those three-max contests, Max or Danny, do you think it’s… It’s still going to be a somewhat manual process at the end of it of picking the lineups because you don’t need to just pick the exact top three, you can kind of tinker a little bit from there without giving up much. But, in a three-max contest, are you going to… Would you do that? Would you just pick the top three or are you going to find three lineups that aren’t very correlated to each other?
Danny Steinberg:
Do you want to take it, Max?
Max Steinberg:
Yeah. I think it depends on how much time. Again, it’s variance time profitability, right? As is, how much time do you want to devote to this of making a bunch of lineups and sort of hand choosing the ones that you like. Which is totally fine strategy, but it’s just going to take a lot more work to make sure that you have the specific three entries you want or you can just set smart diversity to very high and just trust that the lineup builder is going to make a diverse array of lineups. Make sure that it’s high enough that you have three different QBs, things like that. I think in general, I usually don’t worry too much about really managing all my contest entries so they’re spread out a lot, and that I have the perfect three entries for three-max contests, and I have the perfect five for five-max, and the perfect 20 for 20. I trust that the SaberSim builder is building me three good lineups, five good lineups, 20 good lineups, right?
Max Steinberg:
I think a lot of people get really micromanagy in a way that doesn’t really add value. So, I think personally for me, the SaberSim builder builds the lineups. Again, I’m saying the same things over and over again, but I’ll just say it again. I’m working on projections. I’m working on ownership projections. SaberSim builder’s doing the rest. I set the sliders to the settings I want, depending on the contest, and that’s all I do. Right? I think micromanaging can feel rewarding when it works. It can feel terrible when it doesn’t work. But in general, you can’t micromanage your way to the perfect lineup. You just have to have some trust, and I think you can trust the lineup builder to make you quite good lineups, especially for these three- and five-max and 20-entry max contests.
Andy Baldacci:
Yeah. I think elaborating on that a little bit, I guess, or kind of going in different direction. It seems like we could also say that the more lineups you’re entering in a contest, the more costs there is to tinker in. Like, two. If you start with a base from a lineup optimizer or from SaberSim, whatever, your top few lineups are all going to be very solid. If you choose the number two instead of number one, number three instead of number one, whatever, you’re probably not giving up too much. As you start entering more lineups in the contest and start kind of picking from all over the board, the risk that you’re swapping out a plus CV lineup for a minus CV or just giving up a lot of EV between the lineups gets substantially higher. So, it’s just as you put in more and more lineups, you just want to be careful about the micromanaging because the costs get a lot higher and you can realize those a lot faster. Is that fair to say?
Danny Steinberg:
Yeah, that makes sense. Yeah. I think another thing is just with like, basketball for example. You should be paying attention to the last second news and if there is any, and to be prepared for that. A lot of times I make my lineups with like, five minutes before lock. So, there’s not time really to micromanaged if you’re paying attention to all the injury news.
Andy Baldacci:
That’s true.
How do you decide when to fade or play the chalk?
Andy Baldacci:
When do you know if it’s good chalk or bad chalk? When is it time to eat the chalk or fade the chalk? How do you guys evaluate when to, I guess, pay attention to ownership and when to ignore it?
Max Steinberg:
Yeah. I think we can talk about the underlying concept between why to fade the chalk or play a chalk, right? The underlying concept is, is the chalk worth it? Is the chalk is getting a certain amount of high ownership and is this player projected high enough, then will not fail enough where this is worth it. And so there’s reasons why this might happen maybe this player is higher variance than people think. They might be high projected player, but they actually fail quite a bit of the time and they have high upside and they have downside. And when players are really chalky you can kind of think about it in reverse. If the player has high enough downside, then they’re going to fail enough where it’s not worth it or if they’re not projected high enough where there’s not a big enough difference between this player and a player that’s just a couple points projected lower but is going to garner away lower ownership you’re going to want to fade the chalk.
Max Steinberg:
But if that player is just way higher projection than everyone else here and I want to play it. But with regard to say if you’re using SaberSim, all you have to do is project the players right, project the ownership right. Use the Ownership Fade slider to decide how much you want to consider fading in ownership and then let the builder to do it and it’s going to make the decision for you. But the underlying factors with that is does this player have a lot of downside? Is he not actually as good as this other low-on possibility? And that’s all basically SaberSim is considering.
Danny Steinberg:
Yeah, I just want to be clear for cash games, you shouldn’t want to avoid high ownership. In fact, you may want to lean into I ownership a little bit. So this is just specifically for GPPs because they’re top heavy. So you want to take a higher variance approach. So fading the chalk can sometimes be good and for exactly the reasons Max said. And I think SaberSim really takes us into consideration well, and if you’re playing like a high Stacks multi-entry GPP where ownerships are going to be really concentrated, you may want to fade ownership a lot. But if you’re doing a contest under $3 or something that is going to have a lot of beginners in it, then maybe don’t take Ownership Fade that much into consideration.
Max Steinberg:
Yep.
Andy Baldacci:
And one of the other fundamental things that you guys touched on a little bit, but I want to emphasize is, is the underlying variance of the player. Because in baseball, if you had a batter who was projected to be on like 60% it probably almost never makes sense to take that bat. Whereas for a pitcher, it’s a little different because just inherently the variance of those positions, of those players is going to be different. Because if you had someone who was very highly owned but knew with true certainty that this is the score that they’re going to get, that’s much different then saying this guy’s the ton of upside, but they only get it 10% of the time, 20% of time, whatever it may be. And I think it’s kind of understanding those trade offs are very different for different sports.
Andy Baldacci:
It’s like in basketball some of the starters are going to get a certain amount of minutes with a very high certainty and you can expect a certain amount of performance. There’s a range in there, but that kind of the band of uncertainty is much smaller for them. Whereas someone coming off the bench or someone who’s could be potentially injured, someone where there’s more uncertainty around the minutes, around performance, all those things. The more uncertainty there is, the more willing I guess you should be to fade and that seem like a good summary as well?
Max Steinberg:
Yeah, I would say that. I also think, and this is something SaberSim is going to take into account, but just as a general thought is, if there’s any leverage off this player doing badly so if it’s a pitcher in baseball, you can stack against that pitcher. So that’s going to add value to fading that person. If there’s a running back that people are going to really like, and you actually play the defense on the other team. Okay, that’s also another way where you can have some leverage off of that player. So I think if there is that leverage, it’s going to encourage. That’s another reason that you might want to fade a player.
When do you take a stand on lower owned players?
Andy Baldacci:
For those lower-owned players, for those cheaper guys, what do you look for when you’re taking a stand on them? And again, with SaberSim, it’s going to point you in the right direction, but fundamentally what types of qualities are you looking for in those people who might not be as popular, for taking a stand on them? Do either you guys want to touch this one?
Max Steinberg:
I thought I was going to let you take this, Danny. I was just thinking-
Danny Steinberg:
Yeah, well-
Max Steinberg:
… it’s upside. Right? You want…
Andy Baldacci:
I had no idea what he was doing. I thought [crosstalk 00:00:41].
Danny Steinberg:
Why do you want me to look higher or something?
Max Steinberg:
No. I was saying that when you’re taking a stand a low-end player it’s like, why do you like any player in any sport? If you’re talking about [inaudible 00:00:53] is the high-end side, right? You want [inaudible 00:00:57] they can have a big game. So if you’re playing a lower end player, these are going to be the guys whose average projection is not max they’re upside. People think, this person on average is going to do this, but they have a big game at them. Right? Or, I mean maybe they’re going to be underrated for other factors, but that’s what you’re looking for as you’re saying. Okay. For some specific reason I think this player has higher upsides than people think and maybe SaberSim is saying that too.
Max Steinberg:
So that’s those are going to be the players where I’m saying, okay I think this player can have a big game and that’s why I’m taking a stand by them. It’s not because I think his average projection is high, I think is his upside is really high.
Danny Steinberg:
Yeah. I think, I mean sometimes I’m just surprised. I kind of let say SaberSim build my lineups or let the builder build my lineups based on whatever objection it’s using and sometimes I’m just surprised that too I’m using. You know a lot of times the best low end plays are just going to be guys who are really practical. Like you don’t necessarily have to have the most genius idea in the world to have a good low on play. Like you just have to sometimes something someone just really practical is getting very overlooked for some reason and it may not be why. I mean someone in like a really bad matchup, for example, they could still do really well or you know there’s many reasons why someone could be low on but still be worth playing. But often I don’t know who I’m taking a stand on until I use the SaberSim lineup builder and see who’s actually showing up in my lineups.
Andy Baldacci:
And I guess on that note, what are you looking at when you do see someone surprising in there and like you’re getting a ton of a specific player. Cause that definitely does happen. Probably… Not probably it does happen more often in SaberSim than a traditional optimizer where we are just going to be out of the box super heavy on a couple of guys that does seem out of nowhere. How, what like questions are you asking yourself? What sort of things are you looking into? What are you guys doing in those situations to evaluate whether something’s wrong with the model or maybe we’re onto something.
Max Steinberg:
Right. I’m looking at their projection and their ownership projection because you’re saying, okay is this person’s… I mean it’s, it definitely you don’t want to take it at face value cause maybe their projection is too high or maybe their ownership projection is too low. Right? So you want to question whether it’s the correct play. But you don’t want to question whether it’s the correct play, if you check the projection, you check the ownership, it looks very-
Andy Baldacci:
It all checks out.
Max Steinberg:
It all checks out. You’re using the proper sliders. Like if that’s all the case, that just means that player’s a good play. It doesn’t mean that you should try to get less of them, like lowers max exposure, it just means there’s a… he’s projected right, his ownership projections right, and there’s a reason you’re getting him and he’s probably going to be a very profitable play if he is in fact projected be really lot.
Danny Steinberg:
Yeah, I think a lot of times you’re able to understand it. Like I’ll look through my projections and be like, Oh, this guy is going to be high owned, but I’m getting a lot of this guy and he’s actually just a little lower projected and is supposed to be way, way low round. So that, makes a lot of sense. Or maybe, Oh, you know what this is a leverage off of some very popular on play and that’s why it’s doing it. Sometimes when I’m using SaberSim’s builder, you get that sort of outcome where, you know it’s leveraging out of some high owned player having a bad game, in some way.
Andy Baldacci:
Yeah, and I think this, you guys both touched on this a lot more in kind of the videos you did on the quality control, the third step of the ladder building process for basketball and football, where you just talk about what to look out for, but just to emphasize some of that here. It’s just looking at those outliers where it’s looking at the people who you’re getting way more of than you expected. It’s checking to make sure their ownership and their projections are aligned, but going further, you can also click on their name in SaberSim, bring up the distribution of their outcomes, and you can see the correlations, and see if maybe we have them correlated to someone else who is in a lot of your lives positively or negatively correlated as someone who is going to be highly owned or whatever it may be.
Andy Baldacci:
It’s… There’s a lot of things you can look into, but those are types of players to check out when you’re in that kind of quality control phase and you can do the same thing for, Oh, I really thought I would have whoever it may be in my lineups, so I’m not getting him at all. You can hit the show players box and then see him and just kind of dig into why that may be. Maybe there’s someone else who’s somewhat projection cheaper, whatever it might be. There’s a lot in there, and it can be confusing because it can seem like a bit more of a black box in traditional tools. And so that’s why we always try to recommend is: digging into it, looking at all the data as best you can, but when you have those specific questions, you can always shoot us an email, reach out to us in Slack, and we’re happy to kind of help diagnose those things. But hopefully this gives you a good idea of what to look for in most situations.
When do you build your lineups?
Andy Baldacci:
Is there such a thing of setting your lineups too early? And the answer is yes. It depends on the sport, but for basketball specifically, you want to wait until it was close to lock as possible and even then you’re still going to miss some things. The later news comes out in the sport, the later you want to wait to build to your lineups. I think there’s nothing wrong with, if you don’t have a ton of time of doing a build early, just to get an idea of where things are going and who top plays look like, that sort of thing. So that when you’re in there for your final build, you’re not just coming in blind. But for sports like basketball, you want to be as close to lock as possible.
Andy Baldacci:
Football, they put out inactives pretty consistently about an hour and a half before kickoff, so you want to wait until that, but even there, that’s less dangerous. There’s usually not many surprises. You can exclude questionable players from your builds and probably not give up a ton, but that’s one where it’s on the line a little bit.
Andy Baldacci:
For baseball, lineups, it’s not as regimented when they come up, but those usually come out far enough in advance where you don’t need to wait till the last minute. But basically all of this comes down to wait until all the information is confirmed to finalize everything, but there’s no harm in getting in there early just to see where things are shaking out. Does that seem reasonable enough?
Danny Steinberg:
Yeah, no, that’s really well said. I think another advantage to doing… I think, if you’re going to prioritize time to do lineups, I’d definitely try an hour before lock or in that window. If you play DFS a lot, like I do, there’s been three or four times where I’ve had complete mistakes where I forgot to put in lineups or something or forgot to do a swap and just had totally dead lineups, so there’s no harm in doing an early swap or getting some good lineups in early just to play it safe, just in case something terrible happens and you’re not able to swap in time. Getting something in early is always a good idea, but you should also try to get a final run in within an hour of lock, at least, or as basketball 15 minutes before lock. But yeah, baseball and football, an hour out you should be fine.
Andy Baldacci:
And I’ll try not to go on too much of a rant here, but the short of it is just do not do dummy lineups. Like when you’re reserving your entries, if you want to do a single entry for that and just put it in 150 times, whatever, just get your lineups in there, that’s fine. But immediately after you do that, do a quick build with an optimizer, received some whatever, just something so that if the unit goes out, if something happens with our site, with another site, whatever it may be, you’re not completely screwed and running a train of whoever it is and leaving $20,000 in salary on the table, like get in some build early. If you are reserving lineup. That is probably the most costly thing you can do. And I think that might’ve just made Max remember some bad times.
Max: Hey guys, once again, this is Max Steinberg here with my brother Danny Steinberg. We’re partners here at SaberSim, and we want to make a video on how to get started with mass multi entries and tournaments. Meaning how to make 20 or more lineups for a single contest. Now I’m sure many of you watching this video may have some negative views on mass multi entry. You might think, oh, it’s just way too complicated to do it, it’s too complicated for me. Or you think, well I like having control over my lineups. I want to just make a few so I can really make sure who I get and if I make 20 more lineups I just have no control, or maybe you think it takes too much time, you might think it takes a huge bankroll. There’s a lot of reasons that people have sort of a negative view of mass multi entry.
Max: And while this might be the case for trying mass multi entry or what a lot of people refer to as MME, which is what we’re going to refer to it as for the rest of this video. So this might be the case trying MME with a lot of line up builders out there, but we’re here to show you how using SaberSim can make MME easy, quick, and it’s actually very, very, very profitable.
Danny: Yeah, very well said. So let’s address bankroll concerns. So we touched on this in our building a bankroll video, which you can find on our YouTube channel. We actually think that mass multi entry GPPs requires less of a bankroll than playing cash games. And this was because when you enter 20 plus lineups into a tournament, you actually mitigate risks by not placing all your eggs in one basket. So by spreading out the exposure of your players and having a wide variety of diversity in your lineups, you actually lower your risk every week while raising your chances of having a really big score.
Danny: You can also experience much higher ROIs if you are disciplined and use proper game selection. So as tempting as they are, if you can avoid tournaments like the Million America and the Sunday Million which are super high variance and select tournaments that either limit the amount of entries people can have, like such as the 20 entry max play action, or exclude or limit experienced players entirely. Any contest over $3 or one that exclusively says no experienced players allowed, you’re going to be much more profitable. So this means your bankroll is likely to grow and it’s much more unlikely that you are busting your bankroll by trying to grind a thinner edge in cash games. So again, if you want more information to build a bankroll, head over to our YouTube channel and even watch our video on the subject.
Max: Yeah, well said as well. So I think MME in general, one benefit is it’s a lot more scalable than cash games. So when playing double ups or other cash games like that, the only way to get down more money is to move up stakes because there’s only so much action you can get on each level. But with MME, you can actually play the same games you’ve always played, but increase the amount of lineups you create. If you get to the point where you can confidently build 150 lineups like Danny or I, suddenly you’ll find that you can invest quite a bit of money playing DFS every week and generate some real cash on the side while exclusively playing the least risky and most profitable stakes.
Max: So let’s talk about the other concern. Mass multi entry takes too much time. To win at DFS today, you need to build high upside lineups that take advantage of correlations, ownership, and variance. Traditional line up optimizers, literally by definition, optimized exclusively on average projected scores with no consideration at all for any of these components of upside.
Danny: Yeah, and if you’re using a traditional optimizer, you really just have two options. One, you can use simple rolls to save time, or you have to spend hours dialing in every single setting, which can be extremely time consuming and frustrating. Simple rules may be easy to understand, but they’re going to lead to poor results and suboptimal plays, and let me give you an example of that. So let’s say we want to stack a QB with a pass catching running back in order to take advantage of the correlation there. If we create a rule for this, just forcing a QB to be paired with a running back, we’ll end up doing this with every situation.
Danny: So while this may make a lot of sense for running back like Alvin Kamara who catches a lot of passes and is going to be correlated to the QB, you can see Max pulling up that Bridgewater is 24% correlated with Kamara. It’s going to make zero sense for a lot of other running backs like say, Mark Ingram, who scores almost all his fantasy points on the ground and you can see has only a .05 correlation with Lamar Jackson. So when you’re creating lineups this way with these roles, you’re going to make a lot of suboptimal lineups.
Max: Yeah, and then maybe a lot of people, they attempt to account for ownership in GPP. It’s a very, very important trend to project how much a certain play will be owned and the GPP are playing in and possibly trying to fade them or avoid them given the situation. So what you could do to account for that is try to set a rule to avoid players in your lineups that are, for example, projected at 20% ownership or above. This is something that many people playing DFS will do in GPPs, but sometimes players like this are projected to be 20 plus percent owned for a reason. They can be the best plays by far, even in a big field GPP. And so if we set a rule like this, we end up sometimes completely excluding some of the best players on the slate. Something that we really don’t want to do.
Max: The only way to avoid this problem is to spend hours upon hours creating the perfect settings, like putting players into groups. I mean all of these optimizers have these different ways to do this, and it takes a long, long time. And then suddenly Sunday morning comes around, late breaking news comes out. Suddenly a running back is now the best play on the slate. You’re going to have to start this entire process over from scratch. And even for pros, it’s hard to find the time to do this. You have limited time sometimes, there’s a time crunch before the game starts, and this is going to be especially hard for someone who’s just doing this to make money on the side if you’re not a pro.
Danny: Yeah. So I mean with these outdated optimizers, mass entering contests does take too much time. But with SaberSim, it’s way different. Instead of making rules to take advantage of correlation and ownership and all the factors you want to consider in GPPs, we let you decide how much you would like to take these things into account. And then we leverage our simulation data to build these lineups in the way you want. So what you will end up finding is that you end up getting the stacks, the game stack, the ownership leverage plays you want in every lineup, without any of the tedious work.
Max: Yeah. SaberSim makes it easy to build better lineups in less time. And in a second, I’ll show you how SaberSim’s lineup builder works, and how you can actually start doing mass multi entry on your own.
Max: I’ll start by walking through how I built a set of lineups for a great contest to get started with. The 25 Cent Hail Mary on Fandel, which is this contest here. There’s a similar contest on DraftKings called the Quarter Juke Box, but I actually literally cannot see it because I’m an experienced player. But it’s a 20 entry max and it’s also 25 cents per entry, and it’s another great tournament gets started with. In these contests, you can test the waters of mass multi entry for only $5 without any of the top pros, and you still can win some real money.
Max: So let’s take a look at SaberSim’s lineup builder. For those of you who have used a traditional optimizer, this might look way, way too simple. But the reason we can simplify things is we can leverage our simulation data in a way that easily takes all of the important aspects of building great a GPP lineup in to account. And all you have to do is worry about customizing the settings of the three main sliders. So we have our sliders here, and what we feel is most important in GPPs as we talked about earlier is taking advantage of correlation, playing off the ownership, and diversifying your lineups in a way that is smart and takes in to account upside.
Max: So let me just talk about the sliders here. So correlation, obviously that’s something that we really, really want to take into consideration in our GPP lineups. And this means stacking our quarterback with receivers, this also can mean something like game stacking. And you can see if you look at an example of a player who is playing this week, let’s say Alvin Kamara. You can see that he’s correlated with his quarterback, he’s also correlated with some players on the opposite team. And this is the reason that, or we can look at [inaudible 00:09:03] too as he’s coordinated with [Travis Kelsey 00:00:09:07]. This is the reason why stacking is really important, is basically these correlations. And so if we take in to account correlation at a high level, what’s going to end up happening is the line up builder is going to naturally get us these stacks that we want and take that correlation in to account. So there’s no need for rules, but if you’re playing a GPP where correlation and getting the stacks are important, you’re going to want to set the setting to a very, very high setting. So usually I set this about two from the top,
Max: Ownership fade. This really depends on the contest you’re playing in, and given that we’re playing a low stakes contest and I haven’t made my own ownership adjustments which I might make off of maybe buzz or something I’m reading on Twitter. I’m probably going to set this to a pretty low thing, just considered just a bit. And the reason is because we’re playing a low stakes contest, we’re hopeful that the best plays are actually not going to really be high owned and people are going to not be playing in a very smart way. So I don’t think we have to consider this as much. When you move up to stakes like the Millionaire Maker and the Sunday Million, this is where you’re going to want to really focus on perfecting those ownership projections, perfecting your own projections, and you can set this to a lot higher.
Max: And then smart diversity. I actually touched on this in another video on our YouTube channel, which by the way has a lot of great content. But it’s essentially our way of diversifying your lineups in a way that also takes in to account upside. It’s really, really cool. If you want to learn more about it, you can consult our YouTube page and watch this video. But for now, just you can trust me that you want to just set the setting to a pretty high setting. Because we do want the diversification in our lineups, and we want to be able to lower our variants like we talked about earlier. And so now I’m going to just start building our lineups and I’ll pass it on to Danny just to talk about this a little more.
Danny: So SaberSim gives you a lot of control after the builder is over, while keeping the integrity of the lineups we’re creating for you. So this is because with traditional optimizers, when you request 20 lineups, they’ll make you 20 lineups. But with SaberSim, we actually build you much more lineups. If you asked for 20 lineups, we’ll build you a pool of a thousand lineups and give you the 20 best lineups out of that larger pole. Allowing you to control the exposures of individual players or stacks very quickly, without having to rerun your build.
Danny: So this is important because we want to have flexibility and control once our lineups are built. Otherwise, with the traditional optimizer we’ll have to continually adjust and readjust our protectionism and then you rerun the build and then if that builds not right you have to readjust your projections gamut, and it’s an incredibly annoying process. But, and I love this about SaberSim, because our builder creates a large pool of lineups, you can easily swap in and out the kind of lineups you want to have in your portfolio without having to rebuild your lineups again and again.
Danny: So now that we’ve built our pool of lineups, it’s time for what I call the quality control part of the process. So SaberSim gives you a great visual of not only the players we’re getting in our lineups, but also the team stacks, game stacks, and player stacks distributed through our lineups as Max is showing you on the screen.
Max: Yeah, and you’ll hear a lot of daily fantasy tournament pros talking about having quote unquote, “shares of players” in their lineups. And this is because many players look at their lineups as a portfolio that they’re investing in to GPPS. And like any good portfolio, we want to have a diversification of great investments to ensure that our lineups profit the most while mitigating a risk. And this is how you should be looking at these percentages, whether it’s a player, team, or game, we want to make sure we’re getting to the right exposure to the right players and stacks without risking too much in one area.
Max: So let’s say for example, I want a little more of Dalvin Cook in my lineup. All you have to do is adjust the min exposure, let’s say instead of we have him in about 10% of our lineup, so two out of our 20. We want to get him in four of our lineups. All you have to do is adjust the min exposure and suddenly he will appear in four of our lineups, and that happens instantaneously.
Danny: I know, this process almost seems like magic to me when I do it.
Max: Right, yeah.
Danny: It’s like, well what happened? How do we now have Dalvin Cook more?
Max: Right. So you can do the same thing with team stacks, and you can do the same thing with game stacks, and you can do the same thing with stack types. So for example, let’s say I kind of want to cap the exposure that I’m getting to a particular game stack because I just want more of a spread. I think game stacking is really important, and I’d like to spread it out. We can just cap the max exposure to these game stacks and suddenly we’re getting a wider variety of games stacks in our lineup and spreading this out more and more in a way that’s going to keep the profitability of your lineups. Because game stacking in and of itself is already going to give you an edge when you’re making GPP lineups, but you’re just going to get a wider variety. You’re not going to be sacrificing that much expected value from these lineups, but you’re going to get a wider variety of game stacks for stacking six out of 10 games a slate. And that’s going to ensure that’s going to really lower your variants and increase your odds of actually having a big score in a GPP.
Max: You can do this with team stacks as well. Maybe you want more exposure to Cleveland and you don’t want to worry about individual players. You can adjust this there. Maybe you want to look at the players more and maybe you see someone like Steph Diggs that you liked for a particular reason. You can basically just raise them in exposure. You have basically just so much control and such an easy and quick amount of control, that you’re just literally just do not have with other lineup builders. And if you look at this actually, what ends up happening is, remember when we were talking about when we’re doing this. A lot of lineup optimizers, what you’re going to do is go make a rule and you’re going to say, “Oh, I want a quarterback with two receivers,” or “I want a game stack,” or things like that.
Max: If you actually look at this, we are just, just simply by using these slider settings which was the simplest process in the world. You actually have, every lineup has a stack. I think if I’m calculating this correctly, every lineup has a game stack, almost every lineup. You have a wide variety of teams you’re stacking. You have 56 total players that you’re using. And by the way, even cap exposures here too. You can say, “Oh, I just want 50% [inaudible 00:15:56], I want 30% of the Redskins D.” You just have so much control here, and all you had to do was literally set some sliders. And by the way, we actually sort of do this for you anyway by giving you some default settings to go by, but you can switch them back however you like. And what ends up happening is you have some really great lineups and you can actually just look through them and say, “Okay, do I like this?”
Max: And it’s like, look, okay? This is actually just a beautiful line up. We have Houston, we have DeSean Watson stacking with Duke Johnson who’s a pass catching running back who’s going to be correlated to him. Deandre Hopkins, we bring it back with both Tyree Hill and Travis Kelsey. Oh, he has Will Fuller too, by the way. So this a massive game stack of a really high over under game, and you’ve looked down again and you have another Patrick Mahomes. We now have a Jacksonville stack with Michael Thomas, so here’s another game stack. You’re getting just a wide variety of lineups and you check it out and see, okay, do these lineups look good? And I think you’ll find wow, they do.
Danny: Yeah, I just want to reiterate the point you’re making. Max didn’t make any stacking rules. We’re just creating the optimal lineups and they happen to be stacks. We’re doing all that for you. He didn’t have to make all these tedious rules of who to stack on what team. We’re doing all that work for you, and we’re doing it so much better than if you set stacking rules on your own.
Max: Yeah, and I think there’s some nuance here that makes things a lot different. Like for example, we were just showing you Patrick Mahomes’ lineups, and we were talking earlier and in the first part of this video, of stacking a quarterback with a running back as you saw with this or different stacking rules you can make. And you can see that this is just a much nuanced approach. Like if we set a stacking rule for Cincinnati, or we just set a stacking rule and we said, “We want a quarterback with three receivers always, because I really want that correlation.”
Max: And by the way, we’re making quite a bit of lineups like that. That would do the same thing with Andy Dalton, which would just be really foolish. You do not want to stack three players with Andy Dalton, but you do want to stock Tyler Boyd with Andy Dalton. And you do want a game stack maybe a bit here by bringing it back with someone like Mark Andrews in the other team. So there’s just a lot more nuance here that you’re going to get, and if you used a stacking rule it just wouldn’t work.
Max: Yeah. So this whole process with explaining this to you pretty completely took about what, 15 minutes? And it’s actually that easy. Mass multi entry with SaberSim is not that hard, and there’s a lot of great contests that you can make really good money on the side playing mass multi entry with them. And once you start to figure out how to use this lineup builder, suddenly you can raise the amount of lineups you’re creating pretty easily. You can start making 50 lineups or even 150, and it’s a great way to expand the money that you’re putting down and raise the money that you’re making from week to week.
Max: So hope you enjoyed this video, I hope you learned something from this. MME is a great way to make money playing daily fantasy, and SaberSim makes it easy. So we’re offering a three day free trial. So try us out, start on Friday, mess around with the lineup builder for a while, see what you’re creating, see the lineups you create. And then try entering them in to one of these small stakes contests and see how you do. And you can literally do that for free. So do you really have nothing to lose trying us out? I really recommend it, we are the best lineup builder in the business. So again, hope you enjoyed this video and good luck this weekend.
Max: Hey guys, this is Max Steinberg here with my brother, Danny Steinberg. We’re partners here at SaberSim and we wanted to make a video covering the most effective ways to build a bankroll in DFS. There are a lot of different views out there on how to best do this. Some people suggest grinding cash games for a while, some suggest focusing on a few lineups every week. But we think that far and away the best way to build a bankroll is to mass multienter. And this means entering 10, 20, 50, or even 150 lineups, into small stakes GPPs with a lineup builder like SaberSim. And while that might seem difficult and complicated, we’re going to show you why it’s really not.
Danny: I think the first thing to talk about is game selection. The most important thing by far, when you’re building a bankroll is to ensure that you have a positive ROI in the contest you’re playing in. It may seem like if you’re a good player, you should be able to win in any type of contest, but some contests are much softer than others. Some contests are single entry or capped entry, which means the best players normally enter one or a few lineups in that contest. For example, the Play-action is a 20 entry max compared to 150 max of the Millionaire Maker, which means the millionaire maker ends up being a lot harder. Then you even have the Power Sweep, that’s three entry max. And you even have tournaments that have only one entry you can make.
Danny: FanDuel and DraftKings also offer contests that exclude experienced players entirely. These are players with over $1 million in entries in their account history. So me and Max are both experienced players that are restricted from entering contests on FanDuel and DraftKings that are under $3, or that’s say no experience players. So if Max tries to enter this no experience player contest, he gets an error where it says, “The contest is not open to players of your current experience level.” In DraftKings, we can’t even see the contests under $3, so you can’t even get that error message. But these are really great contests to take advantage of.
Max: And there are similar contests like this with cash games, like $2 double up I can’t enter either, or head to heads under $3. You have the similar types of contests with cash games and these experience players like us can’t play them. However, we think that GPPs in general are a much better way to take advantage of the mistakes of inexperienced players. In cash games, all that matters is projecting players and there’s a lot of information out there on how to do that well. You’ll notice if you play cash games, even players who play low stakes are going to have pretty reasonable lineups. The ownership play percentage of the best place in low stakes contests will often be 60%, 70%, or even 80%. Good players just don’t really fly under the radar anymore.
Danny: Right, but in GPPs, there’s so much more than just projection that goes into making a good lineup. You have to think about optimal lineup construction, the correlations of players among each other and on different teams, upside ownership. There are a large amount of factors to consider to make a lineup that’s better than your opponent in GPPs. And because of this, your potential ROI is going to be a lot higher in low stakes GPPs than cash games. With some lineup builders, there are a lot of settings to mess with. You have many lineup builders that work off something similar to what we call stacking rules.
Danny: These are rules that force the lineup builder through heuristics to create specific types of lineups that the builders think will be successful lineups and tournaments. This means maybe forcing a QB wide receiver stack or a QB stack with two wide receivers, trying to group together players in the same game to take advantage of correlation. But with SaberSim, we don’t use heuristics. We leverage the data from our simulation to mathematically take this all into consideration. You don’t have to worry about jerry-rigging our builder to get the optimal lineups you want. SaberSim is going to use the data we get from our sims to do this for you.
Max: So this is where these sliders come into consideration. We have sliders that allow you to consider how much correlation you’re taking into your lineups, how much you’re going to fade the potential ownership of players in a tournament. And then a really smart way that I talk about in other videos of diversifying your lineups and capturing the upside of players. So, I already made a build earlier before this video and if you can look at the example build I created, you can see that without any rules and just by setting these sliders, our lineup builder actually gets the type of lineups that you want. It’s stacking in the way you want.
Max: You’re using a QB with two receivers, you’re using a QB with three receivers. All of these lineups contains some sort of stack. And it’s also using game stacks, which you’ll see a lot of these high stakes players and GPPs do. It’s where you’re taking advantage of the correlation between players in the same game and stacking them up in your lineups. What ends up happening is just by using these slider settings, your GPP lineups are actually being constructed really, really well. And these well-constructed lineups are just going to raise the floor of your ROI in these low stakes GPPs, because you’re going to have that edge over other people.
Danny: One misconception people have about GPPs is that even for these small stakes tournaments, you need a big bankroll because GPPs are so high variance. Yeah, GPPs are high variance if you’re entering one lineup, but if you enter multiple lineups, like 20 to 50, GPPs can actually be a really low variance way to make money. This is because when we build lineups for GPPs, we’re going to use a wide array of players every week. So you can see in this build Max made, really the highest owned player we have is Keenan Allen, which is 62% owned. And you can even adjust that to say, well, maybe I don’t feel comfortable and I only want 40% or 50% as the max exposure. And you can change his lineup stat.
Danny: But let’s say we have that 60% exposure. Let’s say the worst case scenario happens and Keenan Allen gets injured. Even when that happens, you’re still not screwed. Let’s say you made 50 lineups, you still have 20 lineups left that don’t have Keenan Allen that still are going to have a good shot of winning. Even though you may risk some amount of money every week, me and Max, we’d only lose a third or half of our investment in GPPs when making lineups because of the diversity we have.
Max: Usually some lineups miss, some hit and we make some money back. And then occasionally you get lucky and the skill that you’re putting in and the lineups that you’re putting in pays off, and you grab that big score. That big score is when your bankroll is going to get a really nice boost, and probably it’s going to come when your lineups become better, your projections are becoming better. And then you can start moving up stakes and playing these contests like a Millionaire Maker or a Sunday Million, where you have a chance to win a million dollars, which is really the dream. So in general, we really feel like, especially with a lineup builder like SaberSim, where it’s just so simple to just build 20 to 50 lineups that are really well-constructed. Playing these low stakes GPPs and entering them quite a bit is going to allow you to really, really easily build your bankroll way, way more easily than these double up contests, or even these higher stakes contests where your ROIs just are not going to be as high.
Max: So luckily for you, SaberSim is offering a three day free trial for anyone who wants to try this out. What that means is you could try SaberSim out on a Friday, mess with the builder Friday and Saturday, use it to enter lineups on Sunday. And that basically means a free week of great lineups for Daily Fantasy NFL. We really hope you try us out. I think there’s really literally nothing to lose, it’s free. I think you’ll find that you’re going to start winning. You’re going to start actually making lineups that place highly in these small stakes GPPs and you’re going to start developing as a player and maybe develop to really win some big, big bucks. I hope you enjoyed this video. If you have any questions, feel free to tweet Danny or I @maxjsteinberg on Twitter. Danny is @danielsingers. And if you have any questions, please hit us up. I really hope, wish you luck this weekend with your Daily Fantasy contests and hope you win big. Thanks.