Transcript

Jordan:
All right. Hi, everyone. Welcome back to another SaberSim strategy session here. My name’s Jordan, I’m joined with Max. We’re talking a little bit of US Open PGA DFS strategy today. I think now this is our fourth one of these we’ve done.

Max:
I think it’s our fourth. Yeah.

Jordan:
Yeah. So cruising right along here. These have been a natural counterpart here to our office hour streams that I’ve been doing every day, 2:00 PM Eastern. Typically, we’ve been hosting these on Thursdays here, but with the US Open obviously starting tomorrow morning, we wanted to do a Wednesday special stream here talking a little strategy for the tournament and PGA DFS overall. We already have covered sports betting strategy, MBA short slates and showdowns, and a little bit of Esports last week, and all three of those videos are up on our YouTube channel, so check those out if you haven’t already. But we’re excited to talk a little PGA today. Max, how are you doing? Are you excited for the tournament?

Max:
I’m doing great. Yeah, I always love doing PGA DFS for the majors because the prize pools are really big, and it makes it a lot more fun. I qualified for the Fantasy Golf World Championship a couple of majors ago, so I’m hoping to build on that, and really excited.

Jordan:
Awesome. Yeah. Right on. I’ve been pumping this up for the past week. I’m pretty excited as well that the majors, I think get a little bit softer too with some of the [inaudible 00:01:33] players joining the pool, so it should be a lot of fun. Cool. I’m going to go ahead and get us right into this. If you’ve been watching a few of these, you’re familiar with how this works, but we host these live for a reason. We want questions, we want clarifying questions and follow-up questions from you guys watching if you have them. So, anything you’d like us to answer or talk about a little bit, I know we already have a few questions in our queue here, but just pop those into either the YouTube chat or the Office Hours Slack channel in our Slack channel if you have any questions. Otherwise, we’re going to get right into it here.

Jordan:
We wanted to start this one with just a couple of general takeaways, some broader themes that we want folks to leave with at the end of this year in terms of leveling up your PGA DFS strategy. The first one of those is to build lineup constructions that have upside. We’ll talk a little bit more about what that means here in a moment. The second one of those is to really focus on ownership and value the game theory components of PGA DFS in particular. And then the third one is to spend some time crafting a player pool that makes sense. We’ll talk a little bit more in detail about how you can accomplish those things in a minute. But Max, if you maybe want to go ahead and kick us off just with your general approach to PGA DFS.

Max:
Yeah. I think the main thing with PGA DFS, it doesn’t roll off the tongue very well, that I really like to focus on is mostly actually ownership, because if you’ve played PGA DFS for a while, you’ll notice that there’s a lot of variants. There’s no such thing as really someone who’s a [inaudible 00:03:20] or someone who’s a total fade. A lot of fade can be random from tournament to tournament, especially a tournament like the US Open, where the course is going to play very tough. And so, I think in general, I’m not really going for maxing out or doing anything crazy, I’m usually making 20 or so lineups. I’m definitely not playing cash games, just focusing on tournaments, and trying to find those guys who might have high ownership for one reason or another, and trying to depress their ownership in my lineup pool, and really just get a nice portfolio of lineups with a diverse array of players that I like.

Jordan:
Gotcha. Yeah. I think something important to note there too, is that the sites set these players salaries based on some of the opening lines to win the tournaments. And in a sport like NBA, where if news changes and a player all of a sudden gets scratched, it can make a lot of those player prices on the rest of the team really inefficient. There’s really good shock that opens up in a sport like that, because there’s really strong value. In PGA, you don’t really get that effect, and I think people end up finding very subtle salary inefficiencies or other reasons to get on a golfer and then overinflate the ownership. Exactly as you were saying.

Max:
Right. Right. I mean, I think the positive about team sports like basketball or even baseball is there’s really a lot of nuance to a Vegas line. Like if you just played basketball players based on the over-under the game, you would be in big trouble because if a player at teams plank six or seven person rotation, or someone gets blocks or steals or something like that, that over, under doesn’t help them. But in golf, since it’s an individual sport, if someone is X percent to win according to Vegas, there’s not nuance to that. It means he’s that good. So, it’s a little harder to get that much differentiation based on that, or not use that because it really is a good indicator skill.

Jordan:
Right. Well, cool. Let’s go ahead and start here then, since you mentioned ownership, do you want to maybe talk about a few of the signals that you typically use in addition to our projected ownership that we have in terms of assessing where you think the field is going to go, where the shocks going to end up?

Max:
Yeah. I think there’s a few factors that I’m really looking at. One is going to be driving distance, especially on these courses that are par 72 yards, and they have a lot of par fives. Even if they’re not, it’s just players gravitate towards these guys, and I think for good reason, actually, because players with high driving distance usually actually have more upside, in my opinion. I don’t have the data to back that up, but I think it’s true. So guys like Bryce DeChambeau, Tony Finau, people like that are just going to get high ownership. I had a friend who’s really into daily fantasy golfing, one tournament, Tony Finau got high ownership out of nowhere, this is a few years ago, and I’m like, why does he have high ownership, and he’s just like Tony Finau now always has high ownership. His drives are really far, he just always does. And there’s just guys like that who are just going to get played a lot.

Max:
So I think looking at driving distance where you find those numbers, a lot of places, even the PGA Tour website, that’s a good place to look. I think another thing is youth, if you’re a young player who drives it really far, even better. I think there is some … Scott, I forgot this guy’s name, but there’s this young guy, he’s an international player who has huge driving distance, possibly isn’t that good, but he’s probably … I think his name’s like [inaudible 00:07:06] Bauer or somebody like that. Probably will just get some ownership just by the fact that he’s a youthful guy who hits really far. Cameron Champ was another example of this when he got on the tour. People just started playing him because he’s a young guy, and he hit it really, really far. And then, I think the last thing is course history. I mean, you, on these streams, will sometimes share data golf, some of their free staff. And they have a course history tool that I definitely look at every week.

Max:
I think a lot of people will look at that because they’re like, “Oh, I want to find the guy who his projections are going to raise because his course history is really good.” I look at it to see, okay, who are the guys that people are going to overrate their course history and own them too much? Those are a few things I’m going to look at, and you just put this up right there. I’m basically looking for guys who have a solid enough sample and they’re true strokes gander is just really high. If I remember correctly this week, I don’t know what you’re seeing here, but I think Finau is actually one of them, Adam Scott was one of them, Leishman maybe, I’ve heard people talking about.

Max:
So basically, players like that who just have really good history. These guys can fail, they can miss the cut. This isn’t like that strong of a signal, so I take a look at that and try to put that all together to say, okay, which of these guys who are just going to get really high ownership, and either depress their ownership in the [inaudible 00:08:41] process or just straight up fade them?

Jordan:
This is the perfect example of what I was saying, where you have this, maybe some very small edge or something that is very slightly predictive that the field is going to latch onto and overstate. So, it’s almost, I guess in some ways, a little bit like BVP in baseball here where you’ve just got this signal where the sample size is too small to really trust, and you’re going to have people overrate it. Is that the right idea here?

Max:
I think it’s better. I mean BVP, which is irrelevant basically. No, it’s better than that, because it’s certainly irrelevant. I think what ends up happening with particular players, not all players, because you’re going to see some guys on this list, maybe someone like Bubba Watson or even Phil Mickelson, who it’s probably not actually going to move the needle for them just because they just are … I mean, Phil Mickelson obviously just won a major, but I think the consensus is he’s not very good, so I don’t think that’s going to really boost his ownership that much, or Bubba Watson has been so terrible. So, you have to be careful in thinking, oh, this guy’s going to get a lot of ownership. But there’s going to be particular guys where people are just going to wait it a lot because it just fits perfectly into every narrative possible.

Max:
We want these narrative to merge together in a way that’s going to really get that ownership really high. I think another thing I forgot to mention too, is what I’ve noticed over the years playing daily fantasy PGA is that usually you don’t get that high ownership on value players, and so you’re really looking for these high ownership fans, there are going to be guys who are mid tier players who are like 8,500 and 9,500 rand or something like that, who have all those narratives fitting in, and those players are going to be the ones who are going to get 20, 25%, 30% ownership where we might be looking to fit.

Jordan:
Yeah. Makes sense. The other thing I’ll add too on to that point is just listen to what you’re hearing around the industry. There’s a lot of free content out there, especially on YouTube, there’s plenty of golf DFS touts out there putting out free content, and a lot of times you end up seeing this positive feedback loop where nobody wants to get caught holding the bag when the chalky play has a bad tournament. So, everyone touts the same players, and there’s not always a lot of strong data-driven backing behind some of these players that end up getting chalky. You have this feedback loop that happens over the course of the week, and some guy ending up missing the cut, ends up highly owned because a lot of the touts were on him. And you can get that information for free just looking and seeing what people are saying online. So pay attention to social media, see what’s out there in terms of what names are popping up over and over again.

Jordan:
Because again, the core point here is that there’s just not really that level of strong value opening up based the way that these are set up over the course of the week. There’s no backup player suddenly thrust into the starting role in golf.

Max:
Right. Absolutely.

Jordan:
Cool. Let me go ahead and bring my screen down here. I want to talk a little bit more here too, about lineup construction. If you’re unfamiliar with golf, we have a cod halfway through the tournament, roughly half the players on the average tournament are going to get to play two extra days of golf, and the winning lineups are consistently constructed with players where all six players get across the cut, and from there, maximizing the upside of landing top five’s first place, second place, top 10s, top 20s in the tournament. Go ahead. Were you going to say something?

Max:
No, no, no. Keep going.

Jordan:
Okay. Yeah. I think one thing that’s important and one thing that can be really nice about optimizing with SaberSim compared to a traditional optimizer, is that we need to build lineup constructions that actually have a chance at real upside in the tournament. I think a traditional optimizer is often going to build you lineups where it captures some of the high projections on some of the highest salary golfers in the pool, and in order to fit some of those players into your lineups. throws in multiple lower, cheaper players that have a lower chance of making the cut. When you have multiple players that only … two or three players that maybe only have a 50% or less chance to make the cod at all in your lineups, that effect compounds over the course of three players with a 50% chance or less chance to make the cod creates a lineup that has a very low overall chance of getting all six of six golfers across the cut.

Jordan:
What this means is that I think it’s important to ultimately end up trying to construct lineups that focus more around that middle range of golfers or less studs and duds types lineups, because those lineups just possess low overall upside in terms of what they can win. You may end up with a golfer that wins the tournament or two golfers that end up in the top 10 because you played some of the safest plays, some of the plays that are the most likely to get there, but because you had to drop the salary of the remaining golfers in the lineup to get those kinds of players there, you ended up with only five of six golfers making it across. I think that’s an important thing to note when you’re building your lineups. Do you have any thoughts on that, Max?

Max:
Yeah. I mean, I would say to add to this, just talking about the model that we have at SaberSim is, our models based on a machine learning model that regresses quite a bit. I think one thing it does a bad job at is the players who are at the bottom, I think it’s a little too high on. And so in general, I think you’re going … I mean, we’re going to talk about this as really tightening up your player pool and making sure that you’re not going too deep on those value players, especially if you’re using our projections, some of those value plays are going to be a little overrated. So you really, I think when you’re playing daily fantasy golf, you want to tighten that player pool when you’re building lineups.

Jordan:
Yeah. Let’s talk about that a little bit here. I’m going to go ahead and pull this up.

Max:
Yeah. I mean, I’ll just talk about it a little bit. I think if you’re playing daily fantasy baseball, or you’re playing basketball, or you’re playing football, and some people do this anyways, where they’ll do something where they tighten their player pool. They’ll actually just eliminate a lot of players from the pool of players that they’ll build lineups from because they just want to have their core players. When I’m using SaberSim for my sports, I don’t do that. In golf, I do though, because I think it’s really important to basically use the players that you think are viable and make a curated pick of that, so you can get a nice balanced portfolio of those players that you really like, and not get some of these stragglers that maybe you just don’t want any piece of.

Max:
So, something like raising your men projection, Jordan and I were talking about this before the stream, or just taking top 20 percentage or something like that, because basically to win one of these tournaments, all your players are probably going to have to be in the top 20, having some cutoff like 10 or 15% or something, that’s going to serve you really well. So, I think one thing that we’re going to recommend is raise that min projection to something like 50 or even 55, or look at that top 20 percentage, and have some cutoff and make sure that you’re only using players that are going to have a really good shot at getting up there.

Jordan:
Right. Yeah. I completely agree. I think that that top 20 is something that I personally like to use. I’ll take a look at what the pool looks like overall. A common number I’ll look at is something around 10% of a chance to make the top 20, which coincidentally, we saw this right before we hopped on is right around that 55 number that you had mentioned as a min projection. What we’re trying to do is just get some of these players out of our pool that especially when taken together in a six golfer lineup just create a lineup that has a very low win equity, a very low chance to capture some of that upside here, because these players have a low chance of top 20, right?

Max:
Yeah. I think to stay on top of that too, is it can really compound, because if you have a player that’s 10% to make the top 20, then you add a player that’s 5% instead of 10%, that’s going to lower your odds of having six in the top 20 really, really significantly. So, you really want to make sure that you get a balanced pool of players that are actually going to have as much of a chance as possible. Otherwise, that’s going to really kill your lineup equity, especially this millionaire makers, which we’re playing to win a million. So, you don’t want to wimp out and be conservative.

Jordan:
Right. Cool. Another question, I think something interesting to talk to, what does your typical golfer exposure look like when you’re actually going through the process of building your lineup? Say you’re building 20 or 150 or whatever really makes sense for your contests, do you find that you’re attaching to a tighter core and rotating other golfers around them? Or do you get pretty spread out? What does your strategy look like with golfer exposure?

Max:
I usually get pretty spread out. I think there’s going to be occasionally a situation where there’s someone I really want to get leverage on for one reason or another. So, sometimes I’ll have 80, 90 or 100% exposure on a particular player if I really feel conviction about him, but usually, I’m maxing out at 50 or 60% exposure in a particular player.

Jordan:
Gotcha. Yeah. I think I’m around there most of the time too. I think I maybe play a little bit more conservatively in that regard. I think a lot of times my most own golfers, somewhere around 40%. That obviously depends a lot on the kind of contest you’re playing, and so on there too. We did get a question too, it looks like here came in through YouTube chat, Jeff Coley asks here in a three max, would you be building three unique lineups or do you use a three-year foreman core for your lineup? That’s what got me thinking about this. Does that change at all for you if we’re looking at something like a three max?

Max:
If I’m understanding him correctly, is he saying that you use 18 different players for the lineup, is that correct? Is that how you’re interpreting?

Jordan:
That’s what I’m understanding here. Yeah. Three completely uniques or building a core?

Max:
No. I mean, it depends on risk aversion. I think I’m going to have probably a core of at least two golfers, maybe three, but I’m not getting too attached to it. Again, this isn’t like basketball where there’s just a play that fit, maybe isn’t a sure thing, but it’s a really good play that’s going to hit a lot of the time. Golf is a pretty random sport, and so getting too attached to one player thinking, oh, this is going to be the week for this player, I think is wrong. It’s not going to work out for you, especially given it’s a weekly sport, like NFL, getting diversity of players I think is really important.

Jordan:
Gotcha. Cool. Let’s see, I think maybe one thing that might be … Oh, let’s see. Another question.

Max:
We also had a question from Jack in SaberSim.

Jordan:
Yeah. Let’s go ahead pull this up. Jack asked a really interesting question here. He said, I would love to hear opinions on whether or not the current form of PGA DFS is overall just a low edge sport, no correlation in a huge player pool where the ceiling for chalk is typically 30%-ish in large field stuff. You heard Nelson, [inaudible 00:21:18], mentioning he found a linear relationship between total lineup ownership and DK point scored, which would imply the ownership market is relatively efficient as well. We talked a little bit about this before the stream, Max, what are your thoughts on this overall?

Max:
Yeah. I think first of all, I think that really has a lot of truth to it. I think PGA is a low edge sport because of what we were talking about earlier, is the Vegas odds match out so well with how a player is actually going to do. You’re probably not going to get that much edge from a projection standpoint, so it’s more of something where you’re either going to find something, whether it’s a correlation edge because of weather or something like that, which we can talk about it a little bit later or some player you want a lot of leverage on, or you think is a good fade or something. If you don’t have an angle like that, your edge is probably going to be very little. So, I think it makes sense to approach PGA as something that is like, okay, I am doing this as a bit of a gamble, unless there’s some strategy reason where for that particular week, it’s not. You can’t just play every week without an angle and expect to just profit out every time, and there’s going to be a lot of variants to it.

Jordan:
Yeah. I think again, looking at ownership, waiting for other people here to make mistakes, I think a lot of times on this show here, I’ll talk about how larger slates allow other players to make mistakes in terms of the players that you’re using just based on a pure projection. Like players don’t necessarily get on the best plays on big NBA slates or things like that. I think in golf, you really can allow people to make their own mistakes in terms of that ownership. The pricing is pretty efficient, and I think that’s probably true here that in general, the ownership market is efficient as well, but I think it’s probably less efficient than the pricing market for golf DFS, at least that’s just what I’ve found. So, really paying attention to where those chalky plays end up, especially at the high end.

Jordan:
The difference between the odds of some of the top golfers on any given slate winning the tournament, a lot of times is a lot smaller than the difference in the ownership share that those players end up taking up. So, it’s a crucial point of golf DFS and building your lineups. All right. You had mentioned-

Max:
I was going to say, I would just also say if he was talking about having a linear relationship between total lineup ownership and points scored, I mean, that doesn’t necessarily … I just want to explain that a little bit, because all that means is that owners, total lineup ownership is correlated to point score, which of course, it is. You have basically a bunch of players who are trying their best to make the best lineups picking these plays, and that’s going to have some signal, unless we were all really big idiots, that of course, is going to have some signal. It’s going to have a signal and basketball, baseball, or football. So, I don’t think just the fact that there’s a linear relationship necessarily means it’s unbeatable or something, it just means what we already know, which is that players, if we all put our heads together, and I’ll try to pick the best players on the slate, it’s going to have some signal, some of those guys are going to be right. So, I don’t necessarily think that that means it’s unbeatable or anything.

Jordan:
Right. Cool. Where’s my mouse? We had another question come in here from John on our YouTube chat, he said, what are your thoughts on inter lineup correlation when building lineups specifically in salary tiers and the fact that only one player can win?

Max:
Yeah. I’ll say two things, one is our builder is … The golf projections are built on the simulator, and so when you’re building lineups, especially when you’re using high smart diversity, this is going to be taken into account. Every simulation we’re projecting, whether what plays to golfer is going to get given, that set of simulation. So, it’s going to take that into account, and I would think just as a general practice, I think this only is going to be important either in showdown, or we’re talking about two players who are really towards the top. So Jordan was … well, I don’t remember if you were recommending this or not, but not doing a stars and scrubs type of lineup build. I think that definitely can do … if you take two players like let’s say Jon Rahm and Dustin Johnson or something, they do take away from each other in that way a little bit, so I think that does make sense when you get to the middle tier, is that they …

Max:
I mean, someone’s going to have to get first, someone’s going to have to get second, you don’t have to worry about it that much.

Jordan:
Right. Yeah. And this has come up too a little bit in Office Hours, we’ve talked about here. When we’re looking at the projections, we’re looking at a mean average across all of the simulations. So, every player’s win equity, the simulations that they win the tournament in play into that projection. But the question here really is getting at in reality, in the one single tournament that’s going to take place over the weekend here, only one player is going to win. So, I think that that is captured really well in smart diversity here, especially with a smart diversity that will end up being at like nine or 10. You can see, we’ll actually start to capture the real ranges of outcomes of the players, and your lineups will be built in a way that makes sense. If you’re using two higher salary players that have a higher win equity for the tournament, the simulations of the tournament will be there to back that up and justify the usage of that player.

Jordan:
So, even if the simulation that you’re looking at is one where maybe you have both Rahm and Xander, and Rahm wins and Xander takes second, it’s still a lineup that has tournament winning upside given that potential outcome.

Max:
Yeah. I actually didn’t even know he had that visual because I think that the visual of how the simulation looks like is actually very important to look at, because it’s a nice visual of how golf is an interesting sport as opposed to something like basketball, because there’s this binary outcome, you can make the cut, or you cannot make the cut, and that affects how the player does quite a bit. So, it’s something that’s interesting to think about is how to cod dynamics of fact if a player is going to be a good player or not. One of the reasons I like to focus on ownership is because players, if they miss the cut, have epic downside as they literally are … you’re dead if you only get five out of six making the cut. And so, that’s why something like fading can be really important with these high-end players, because if that ownership gets too high and display or misses the cut, suddenly, you’re in really good shape against 25 or 30% of the lineups in the pool.

Jordan:
Yeah. Absolutely. I completely agree. There’s really no distribution curve across any of our sports that look just quite like this, so it is pretty unique in that way. Let’s see, a couple other questions trickling in here now in YouTube chat, Austin asks, does the model update during the week, does ownership change? I only ask so I know if I should wait until the night before to start building.

Max:
It’s not going to update through the week. We currently do not take into account weather in terms of course difficulty or anything like that, and obviously, the players are not going to become more or less skilled from Monday to Wednesday. I mean, unless we’re tracking their practice or something, so the skills are not going to change with the players, and since we don’t take into account weather in terms of course difficulty, nothing’s going to opt in that way. I think possibly the ownership model will, although I’m not positive because I only built the projections, I didn’t build the ownership model, so I don’t know if that might update. But I think if you built lineups on Tuesday, I don’t think much is going to change.

Jordan:
Yeah. I think the only thing worth making sure that you double check there is the night before, always keep an eye out, or even the morning before, if you’re an early riser, keep an eye out for those withdrawals that can definitely throw a wrench into things. I don’t think that ownership is going to update on its own. The ownership model here is pretty data-driven, so what it’s not going to do, and another reason why it’s so important to just keep an eye on the industry, it doesn’t track industry buzz very well at the moment. We’ve talked a little bit here about ways we could maybe incorporate that in the future, but just another way to really get a little bit of additional edge on that ownership side is just listen to what people are saying out there, what kind of golfers are getting touted, and so on? Because it’s indicative of the fact that the projections don’t change across the week, but the industry buzz does, that is a symbol of the market inefficiency there with ownership.

Jordan:
A golfer doesn’t get better from when the player pricing comes out to when the tournament starts, but ownership changes, and what the industry is going to do so much over the course of this four day period, that you can take advantage of that.

Max:
Yeah. I think also, that’s such a great point. I think you’re not going to get as much value adjusting projection as you’re going to get adjusting ownership projection. You can definitely add value to our ownership model and tweak it, and that’s going to help you if you set ownership fade to high in it, like pretty high, like you might want to do in a million or maker type tournament, that can have an effect on your lineups. And obviously, the thing that we want to do when you’re using SaberSim is you want to find a way to add value in some way. That’s the way that I usually focus on.

Jordan:
So, what does that look like in practice for you? I’m curious. Do you find yourself maybe taking a few of the highest own plays and bumping them up on the ownership, or taking things down, or what does that end up looking like?

Max:
I usually will bump up their ownership projection, and then in the post [inaudible 00:32:08] process, I’m going to maybe do something like cap their ownership or eliminate them from the pool entirely or somebody like that. I think our tools are really good for managing the ownership, so you don’t have to completely fade someone, you have that. I mean, like 5% or 10% of your lineups or something like that. But it’s going to be adjusting that, and then managing it in the post [inaudible 00:32:32] process.

Jordan:
Gotcha. Yeah. Something I’ve noticed too, is that these two ideas of limiting your player pool by either excluding players or setting a higher min projection, along with tweaking ownership numbers, and then using a higher ownership fade, work really well together. What’s going to happen if you put a really high ownership fade on right out of the gate and just start building lineups with a wide open player pool is you’ll see some of these depressed ownership numbers of some of these really low salary players in the pool. The builder at a certain point will start to prioritize these plays because that ownership is so low. When you have a higher ownership fade and a tightened player pool, the builder will instead prioritize some of the higher dollar players that have a legitimate chance of success in the tournament with just a seemingly illogical lower ownership number associated with them.

Jordan:
So, I definitely recommend checking out and trying those two things in combination with each other, where you’re limiting your player pool to players that you really actually want to use, and you want in your lineups and give lineups with upside along with a higher ownership fade, looking at some of those higher end plays. Cool. Let’s see. We’ve still got some more questions trickling in here. I’m going to just keep moving here. Samuel asked, does the two to three unique players be set somewhere within the builder? It’s not something right now that we can do, but we’re working on it as a part of one of our next feature releases here. I know we hear that a lot, especially on the golf side, I think folks that are familiar with some of the other golf optimizers out there are very used to that unique player number. So, that’s something we’re working on right now. I will say … well, go ahead. Were you going to say something?

Max:
Well, yeah. I was just going to say, I mean, in the post [inaudible 00:34:28] process, the reason you’re doing two or three unique players is you basically want diversity of players in your player pool, you want diversity in your lineups. If you use the post [inaudible 00:34:42] tools and you depress the maximum ownership of specific players, that’s just going to happen naturally. So for people who are looking for that, and we are going to add up, but we’re mostly adding it because people want it. It’s something that you can also just get by depressing the max ownership of a lot of players, and eventually, you get a wide array of lineups. So for the time being, before we have that feature, and maybe when we released the unique feature, I’m probably not going to use it, because I like doing it with the ownership players, you manage the ownership players, that’s how you get a good diversity of lineup. You don’t really need to use that unique players feature.

Jordan:
Right. Yeah. And you’re going to have a pretty diverse pool of lineups already just captured by having your very high, smart diversity, which is going to be the default setting on practically every tournament or contest type that there’s going to be out there, especially for a major. So, lineups should already be pretty diverse as they are. Here’s an interesting question, we’ve talked a little bit more about this already, does the model take into consideration the weather when it comes to AM, PM waves?

Max:
It does not. I think obviously that would be better if it did, it’s complex to do. But I think in general, one thing that I … I think weather is really, really important, and the reason it’s important, I think more than just how it might change the projection of players is it can make players correlated, which obviously we know from any other daily fantasy sport that correlation is a very valuable tool for making lineups. If there’s a certain weather thing, weather is not certain, especially wind. And so, you’re going to be setting this taking a chance on the wind behaving in the way that’s being predicted. But if the wind does do that where the AM on Thursday is five miles an hour, and then on Thursday afternoon, it’s super gusty, 20 plus miles an hour, and then Friday, it’s just calm, then that means those Thursday morning golfers are going to be at an advantage.

Max:
And so, something I like to do is look at T times whether it’s on pgatour.com or whatever website you can get that from. Well, sometimes well, actually, if the weather splits are strong enough, we’ll actually just exclude every player from the player pool that is in the worst way, and just only play players from that wave because you can get a good advantage with that correlation. I think a lot of people use this as a way to frame their projections, but I think a really valuable way to do it is just literally going ham on just playing a particular way of and taking advantage of that correlation.

Jordan:
Yeah. That’s a really good point. And that goes back to thinking about PGA as at times a little bit lower edge, right? That’s a very clear edge that you can identify and push if they’re on the right tournament.

Max:
Absolutely. I mean, if you see my lineups in PGA DFS, sometimes you’ll see me do this, or just all my entire player pool is just stacking T times, and I think that’s a valuable thing to do in some occasions. Sometimes I think the weather differences are so … people are thinking, oh, it’s like five, and then 15 miles an hour, and then five and 10, so there’s a slight advantage, and it ends up being not an advantage at all. So, I would certainly only do it in specific situations, but there are some situations come out where it’s really valuable.

Jordan:
I think for this particular tournament, whether it’s trending to be a huge factor, unfortunately, but something definitely to look for. And something to look forward to if you’re playing Showdown, right? Showdown is even harder to find some of those edges. I think there’s maybe a little bit more juice that can be squeezed out on the round four Showdowns where you actually take into account finished position, but if you’re playing showdowns and looking for an edge, honestly weather is maybe one of the only ones that you’re really going to be able to squeeze out on the average showdown slate for golf. Cool. Let’s see. I’m going to keep rolling with some of these questions coming in. We’ve got one from Matt here, how much would you generally adjust the projections if you dis, or like a player from a percentage standpoint or perspective?

Max:
That’s an interesting question. I think you could go as much as 10% in either direction, but I would say also that I’m probably more focused on the post build process with golf because I’m just focused on getting a nice, I call it a portfolio of players. It’s like you’re getting a diverse array of players and getting exposure to all the guys you want, and maybe adjusting a min exposure or [inaudible 00:40:02] max exposure when I think the projection is too high or low or something like that. But I would say 10%, so if someone’s at 60, giving them six more points or six less points, I think is pretty reasonable.

Jordan:
Yeah. I agree with that. I also will say, I think again, thinking about these price points already being rather sharp, I mean, at least the way I’m thinking about my approach, most of the time, I haven’t gotten to Wednesday afternoon or whenever I’m building my lineups and feel like I’ve found that one golfer that nobody else is on that I want to bump up. I think more often what’s happening, at least for me is I’ll be looking at golfers that stand out as maybe a potential outlier for their salary band, and bringing them slightly back into line. And in that case, I think it may even be just a few percentage points sometimes, but I was looking for one to see if I could find an example, and there’s not really anything that’s coming in. But you can also manage that in the post [inaudible 00:41:04] process exactly like Max said. A golfer that’s overly projected for their salary may end up showing up in 60% of the lineups after the build, you can tune some of that down and bring that down just by looking at your exposure.

Max:
Yeah. I would say that when you’re doing this, I’d look for players that really have something that isn’t just like, he’s playing well recently or something. Someone that sticks out to me maybe was Gary Woodland, because he was a very, very good golfer a couple of years ago, and had a hip injury that’s really just hobbled him for the last a year or more. He’s someone I have my eye on when he starts playing better, because I think as he heals from that injury, he’s going to be a top player again, if he ever does, and so that’s someone I’m looking at where I’m like, okay, if there’s a reason performance is good, maybe he’s “back”, things like that where you think someone’s injured or think someone has been injured and it’s affected his historical performance, I think that’s where you can get some edge adjusting projections.

Jordan:
Gotcha. Yeah, for sure. And on the subject of short term performance, I actually think that’s another signal that the field generally overreacts to, less so maybe on golfers that have necessarily won a recent tournament, but I think sometimes golfers that have maybe struggled or missed a few cuts, even just in the very last tournament, sometimes can just seemingly see some depressed ownership on their very next tournament they play. I think that’s another signal that you can look at for finding some guys, especially at the higher end, a guy … up until two weeks ago at the Memorial, Patrick Cantlay was someone that had missed, I think, three or so cuts in a row, or three out of the last four or something like that, and players got him at a relatively depressed ownership in that tournament, just because he’s still one of the top golfers, still top salary guy, a guy that Vegas isn’t likely to undervalue, but the field does, and you get that depressed ownership on a golfer like that.

Max:
Yeah. I would push back on it a little bit, because I feel like … like Phil Mickelson, for example, we were talking about, he just won a major, right?

Jordan:
Yeah.

Max:
If he got more than … I mean, I guess we’re projecting about 8% ownership, but I would bet anything, he gets less than 5% ownership. I don’t think people who play daily fantasy nowadays are trained to sort of, especially the most recent performance, are trained to just not value it so much that they feel like they almost completely ignore, especially when it’s the last tournament. I think when it’s really like the last four or something like that, and they’re missing every Cod, and there’s some sort of rhetoric around their play, like oh, they’ve lost it or whatever, or their driving accuracy is really bad or something, the trap is when there’s a staff to back it up, and then suddenly everyone’s off them, and then suddenly, their driving accuracies back because people are not robots, and they work out, they’re driving or something was going on, they’re fine again. I think it’s got to be more than, oh, this player has won, because in my experience, people just don’t value that, that much.

Jordan:
Yeah. I think it’s different with winning the tournament. I see it a lot with cuts is where I particularly just around the industry, and in some of the content that you see is players that have missed multiple cuts in a row. I think there’s safety in saying this golfer has made four cuts in a row, so we’re going to tout him, we’re going to tell people to play him. And if he doesn’t, well, he made four in a row, how could we have known? That kind of thing is something that I see at least on occasion, and something to look for in your research. Yeah, I know, I agree with you. I mean, at least, most … I don’t expect Phil Mickelson the winner of the most recent major to come out and be the highest own golfer of the week, but I would expect …

Jordan:
I mean, I will say, someone like Garrick Higgo, who won last week, who I think is down at 7,200 is a player that I think is probably likely to come in a little bit higher owned than we have projecting right now. We’ll see how it all shakes out, but I think there is some of that value in the player that won, and maybe surprised people the week prior. But I could be wrong, and it’s something that has always come up-

Max:
No. But I think you’re right, but I would actually see it’s got a … This is what I was thinking about with narratives coming together, is Garrick Higgo is young, he drives it far. It’s not just that he … Because I would agree with that more than if you gave me the argument that Phil Mickelson was going to be higher, because Phil Mickelson is old, he doesn’t drive it very far, everyone thinks probably his last major one was just pure luck. Garrick Higgo, it’s like, okay, this is a young up and comer, drives it pretty far, that type of narrative built on top of that, I think will increase his ownership. So, you’re right.

Jordan:
Yeah. I think that’s what you hear, is like, is this the start of something special with Garrick Higgo? I think golf fans in general are a romantic bunch of people, talk about the classic runs of golf history all the time, and I think everyone gets excited that, oh, this is the start. Higgo took his first win last week, this is his first major, and so on, and that starts to pump up some of the ownership. And maybe he’ll go on and win the whole thing.

Max:
I was going to say, our ownership model really … I think something that’s a real benefit to ownership model, if I understand it correctly, it’s based on how the lineups are actually going to be built based on different factors. So some people try to assign ownership, but they’re not thinking about, okay, when someone actually tries to build a lineup, what happens? And that’s something that’s special about our ownership model that I think is really cool. And so, when you see someone like Xander Schauffele, we projected him at 27% ownership, that’s just probably right. Not only does he have that narrative, he also fits into lineups really well. So, I think you don’t have to just make up ownership all on your own, trust the model and you make adjustments to it, but also just something to keep in mind is someone like Xander Schauffele is probably going to be the highest ranked player, I would guess.

Jordan:
Yeah. I think so. Yeah. And no, you’re exactly right, it’s using real lineup constructions that would actually make sense for the contest to determine the ownership. And you can see it too, it gives you an idea of what the options are at a certain salary band. Like Xander at 9,300 occupies a certain spot in a lineup. And by playing him at 27% or so of the lineups, really, the statement being made there is that he’s by far one of the most efficient golfers to use at that salary. I mean, you can see even just looking back at this Higgo example here again … I know another golfer and I think we’ve actually got him listed as such, that’s likely to be popular at the exact same price point is Charlie Hoffman, who we have at 13%. Again, that’s the advantage of generating ownership this way, is that Higgo and Hoffman being the exact same price, essentially occupy the exact same role in a lineup construction.

Jordan:
So, even though I think too, both of those players are likely to be highly owned this week, by looking at ownership this way, you maybe see that they’re not potentially as highly owned as maybe as expected because they’re essentially splitting that 7,200 equity. It’s just the way I think about it, but …

Max:
No, that makes a lot of sense.

Jordan:
Yeah. Let’s see. There was one other question that came in from Austin here, made a comment in the beginning of the show, listening or seeing the industry touts, are there analysts out there that you like to follow or listen to?

Max:
I mean, I just look on Twitter because I follow all the top players, so just from that is usually what I’m looking at. I’m curious what you would say, Jordan.

Jordan:
Yeah. Honestly, I’ve gotten a lot of mine from our Slack community, people that maybe necessarily I didn’t follow before, but now I’ve been introduced to. But I think RickRunGood and Pat Mayo both make really entertaining videos. They have strong content. I watched theirs most weeks. And again, I think from my particular standpoint, I am more looking at those guys to see where the chalk’s going to be. I’m not so much necessarily determining my player pool off of what the content I’m out there reading, but I think you often get some initial semblance of where that early ownership is lying. Those videos also come out really early, a lot of times those are out Sunday night, Monday morning. So, I think they establish a tone amongst the general player base of where the best plays are, that then again, creates what I feel like is that positive feedback loop throughout the week.

Jordan:
Or anytime you hear on Twitter sometimes, I’ll see people saying they got talked on to a player, and I’m always like, why? What happened? What do you mean you got talked onto the player if none of the information has really changed? You had the information at the start of the week and at the end of the week. Anybody getting talked on or the public getting talked onto a player over the course of the week is a huge flag for me.

Max:
Yeah. I mean, I feel like there’s a lot of stuff we just don’t know, and if you were actually able to literally interview every player and get a read on their psychological wellness, like how they’re feeling about their game, you’d probably find signal in that. And so, I wouldn’t dismiss it entirely, but it’s just sometimes where it’s like, he loves this course, it’s like, okay. Or, oh my God, the one that’s the best is when it’s someone’s home course, everyone loves that. I’ve been around the block enough to see that that just does not work how you think it’s going to work for one reason or another. It’s like, I don’t know, you’d think it would, but no one’s doing the analysis. I mean, maybe there’s someone, I don’t know, but if there’s actually a signal of this, but from my experience, some of these talking points about someone just knowing the course really well or something, it just sometimes does not pan out at all.

Jordan:
Yeah. It’s not to say that there’s not actual real value that opens up here because there is. I mean, there are signals in these things, and probably some of these typical trends that you see are somewhat predictive of future success. I think even data golf has their short term recency bias quantified in terms of how much that actually builds into the model. But I think the general trend of just what I see over and over again is that these things get overstated. And when the field is searching for an edge or a value play, or a minor inefficiency in the market, they just get overload.

Max:
Don’t overrate a minor inefficiency, I think that’s a takeaway.

Jordan:
Yeah. That’s a good one. Cool. Let’s see. I’m just reading here, we did get another question here from Jack. I think this one came in earlier this morning, post this here. He said, if you guys have time, is there any chance you could go over some reasons as to why SaberSim is projecting so many more fantasy points for this week across the board relative to other places? We talked about this a little bit before the show as well, do you want to take a stab at this one?

Max:
Yeah. The model that I created currently does not [inaudible 00:53:59]. It currently does not take into account course difficulty. That can be difficult in some ways, because you can have the historic difficulty of a whole, but that can change based on how the course is playing or whatever, so I decided to leave it out because in some ways, the data was not great. That being said, in one sense, there’s a way that it doesn’t matter because what matters is the rank of the golfers is still going to be the same, but I think it’s important to get a feel, especially based on weather, whether the course is going to play harder or not, because that’s going to make it so the differences in projections between players is going to get tighter, and it makes it a little more random. And so, it makes you want to take more chances, whereas if the weather’s perfect and the course is really easy, you’re going to want to take less chances.

Max:
So I think there, you have a little bit of both this week because the weather is legitimately perfect, but it’s the US Open, so who the hell knows what they’re going to do, whether the greens are going to be hard as a rock and 12 on the step meter or whatever, but you just want to keep that in mind. If it’s the US Open, you can take a few more chances because it’s going to be a little more random. If it’s the masters on a really easy weather thing, or it’s just a course that subpar 72 and plays really easy, and the weather’s really good, you’re going to want to take less chances. So, it’s something to keep in mind, but in terms of building your lineups, it’s not going to be that big of a deal because we are still fitting to the course in the sense while we’re fitting to par fives and par fours and par threes in length. So, you’re still getting the course fit, you’re just not getting the course of difficulty as accurate as it should be.

Jordan:
Right. That makes perfect sense. Yeah. And again, I think I wouldn’t jump in and feel like there needs to be an enormous amount of tweaking to the projections in response to that. The ranking is still going to make sense, the overall impact to the player pool is relatively consistent. Cool. We’ve got a few minutes left here. I think this has been pretty successful here so far. If anyone has any final questions before we wrap up, feel free to shoot them into Slack or into YouTube chat. I’ll wrap us back around here. Just a couple reminders of the main takeaways, the things we wanted you guys to get out of this is to build lineups that are constructed with upside. When we have multiple players with a really low chance or relatively low chance of making the cod or pulling out a top 20, and then we start to compound those multiple players and do a lineup, we create a lineup that just has very low overall upside.

Jordan:
It’s important to pay attention to ownership, especially some of the signals that we talked about with driving distance and youth and recent form, and … what was the other one on your list? There was another-

Max:
Driving distance-

Jordan:
Course history.

Max:
Course history. Yeah.

Jordan:
Yeah. Make adjustments to our ownership numbers in order to account for that, and bump certain players up that you think are going to get popular. And then the final one was to manage your player pool, to cod out some of those players closer to the bottom by either excluding them or using a minimum projection. We talked about using make cut. I like to either use make cod 50% or top 20 10% as just a starter signal for where that line may need to be drawn. But using that, and then also allowing the high ownership fade and high smart diversity that the builder’s set at is going to allow you to really build strong lineups right out of the box.

Max:
Yeah. Just to piggyback on that a little bit is, this isn’t like basketball or football or something like that where you’re going to get a lot of edge trying to find these guys are going to play high minutes or under projected people or somebody like that. It’s like, focus on the things that are going to provide some value, and that’s going to be ownership, getting a nice diversity of players in your player pool and making sure you’re not taking such a ridiculous chance that it lowers the odds of your lineup getting six out of six, or getting all players in the top 20 so much that you can win big. So, definitely focus on those things. I think ownership in particular is one that I love to focus on, because I think it’s where I can add the value. So, definitely focus on those things, and you’re going to do pretty well.

Jordan:
Awesome. Cool. It looks like you got a couple nice stash comments here across YouTube and Slack. People were appreciating the stash. Cool. Well, anyway, I think this was successful. Thank you everybody that took some time to tune into us live here. I’ll have the recording of this up on YouTube later this afternoon. So, if you’re watching the recording of this, thank you as well. I’ll be back again tomorrow with Office Hours at 2:00 PM Eastern. Good luck in everybody’s US Open lineups, I think it’s going to be a lot of fun. I’m excited to start putting some of mine together. I’ll be around here this afternoon too in the regular channels on Slack PGA Office Hours, wherever you guys want to ask, if you have some follow questions, if that come to mind. Max, anything else you want to add on before we get out of here?

Max:
No, that’s it. I mean, I see a couple more questions that might’ve already been answered earlier in the video. So, definitely check it out. We talk about making the count on top 20%, but I think if you have any more questions, feel free to ask it in the Office Hour Slack, and thanks for joining us. I appreciate it.

Jordan:
All right. Cool. Take care.

Max:
See you.