How to Get an Edge in Sports Betting

How to Get an Edge in Sports Betting

Transcript

Andy:
All right. Hey, everyone. Welcome to our first strategy session as part of the Monday through Friday office hours that we’re doing. So this is something that we’re experimenting with, but we’re going to be hosting these at the start of every Thursday’s office hours where we’ll get a handful of different people in from our team, from our players, just all over the place, who can share some thoughts on different parts of DFS and betting strategy.

Andy:
We’ll still do Q&As while we’re doing this, but we thought this would be a good way to just do some deeper dives into areas that can help you level up your game. So what we’re going to be doing in this video is covering sports betting. So I’ve got Jordan with me as well as Max, and we’re going to be just diving right in. The first thing, I guess, to talk about is actually just the SaberSim betting tools. This is where we’re going to spend a decent amount of our time, but really what Max especially is going to get into is everything around betting strategy and how SaberSim fits into that, for sure.

Andy:
But there is a lot of other information, a lot of other places you can look to improve your betting results. So that’s what a lot of this will be. But yeah, let me just share my screen. All right. Then we’ll get it added here. Okay, perfect. Our betting product right now is very simple, and that’s kind of by design. We’re not trying to just overwhelm you with a bunch of information. We want to show you, just get right to the point and show you the most valuable information that we have.

Andy:
These are things that we’re going to be building on in the future. We can kind of touch on some of those improvements that we have coming down the pike in a bit. But, for now, let’s just talk about what you see. You’ll see all the games here, start time, who’s pitching. If it has a check mark, that means the pitcher has been confirmed. And then-

Max:
It means the lineup has been confirmed too.

Andy:
Yeah, exactly. So we’ll know the order of the batters as well, which does matter. That is an important thing. Max will definitely touch a bit on that, about how to get edges by looking at when things change from what’s expected. But pretty much we’ll pull in the odds from different major books and then compare those odds to the results of our simulations and say whether or not we think the bet is a profitable bet. So we think for this one, that there is a 1.1 unit edge on Colorado for the money line when the odds are +112.

Andy:
A couple things to point out here, we consider a unit to be 1% of your bankroll, and this is calculated using a quarter Kelly, the Kelly criterion, which is a great way to manage your bankroll, minimize the risk of ruin, and maximize the growth potential. The other thing to touch on is that the odds matter a lot. This is something that you’ll get from casual gamblers a lot. They’ll just say, “Who do you like?” That’s part of the question, but what really matters is, at what price? Because we could give you odds on any team, any game that make it unprofitable.

Andy:
So you want to make sure that the odds you’re getting, if they don’t match, they’re close to it. But also, Max will touch on this in a little bit, that you’re finding the best odds available for the different bets. So we just analyze all of these games, show you which bets we think are profitable according to our data, and that’s really how this tool works at the face level right there. We also have a prop betting tool that allows you to look up different prop bets and put in the odds.

Andy:
We’ll tell you if, again, according to that simulation data if we think there is an edge there. Something to touch on as well before we dig in more is just what makes SaberSim so well-suited to this is really those simulations that we’re running because betting all comes down to accurately determining the frequency of an outcome because that’s what odds are. To break even on a -100 bet, you need to win the bet 50% of the time, and the odds go up and down from there.

Andy:
So what a lot of other models are doing, they’re kind of using spreadsheets. They’re using averages. They’re trying to say, based on this historical data, how often are they going to beat this line? That can work, but they don’t have the advantage of getting a massive data point, a massive set of data points for the game that they are betting on right now. Whereas, when we’re simulating out each of these games thousands of times, we can just count literally how many times does this outcome happen? How many bases does this hitter get on average across these games? How frequently do they hit that number?

Andy:
That’s why we’re able to calculate really good ceiling numbers and distributions for the DFS side of it, and that’s why we believe there’s significant edge on the betting side of it. But that’s really the basics of our tools. What I want to really jump into is Max’s betting strategies and just kind of talking to him to hear how he approaches this. Before we do get to that though, there is a video on YouTube that Max and his brother, Danny, put out two Super Bowls, just showing how they used our prop betting tool to wager $70,000 in Vegas and came out, I forgot the exact number. But you won tens of thousands, right?

Max:
I think we won about 20.

Andy:
Okay, all right. Not a bad day.

Max:
Yeah. It was super fun.

Andy:
Not a bad weekend.

Max:
Yeah. We’ve been doing it every year, so it’s really fun.

Andy:
Yeah. So let’s talk a bit about that, Max. I know you and Danny as well both do a good amount of betting. What sports do you focus on? What are you looking for? Just what are your thoughts in general there?

Max:
Well, I would say our bread-and-butter sport is baseball, and that’s mostly due to SaberSim. I think it’s our best model. It’s just something that we’ve both been doing for the last few years. We do quite well. I actually think that especially if you’re not trying to put down thousands of dollars on a game, like we’re trying to do, or you’re in a regulated market, like New Jersey or Iowa or wherever, where you have access to a lot of different sports books, you can make really good and really easy money using the SaberSim model, in my opinion.

Max:
A lot of people don’t have access to something as sophisticated as this, and it makes life a lot easier when you’re actually using the fundamentals of sports betting that make it easier for you. So I would say a couple things that people who are newer to sports betting, I think that they do wrong is a few things. One is not betting early, betting just maybe like an hour before the game’s start, not taking advantage of those early lines, not shopping for lines, just sort of saying like what Andy said earlier in this video, saying, “Oh yeah, I like the Marlins today.” Well, I mean who likes the Marlins any day?

Max:
“But I like the Marlins today, so I’m just going to go to this one sports book and go bet it.” That’s not what you should do or even saying … I mean, our model, we have tested and it does … Over the past few years at least, the time we most recently tested, it did beat closing lines. So you can sort of do that and probably grind out a small profit. But you may have a lot more of an edge line-shopping, actually looking and trying to get the best line and saving -5 or -110, something like that. The difference between -110 and +100 can be a lot, especially in a money line.

Max:
And then also just betting props, I don’t bet props that much, except save for the Super Bowl where you can get down a lot of money, because most of the time I can get down $100 or something. Sports books usually limit me really fast. But if you’re betting 50 or $100, these can be really, really high-edge bets and we have a great tool that helps you make those bets. So there are a couple things that I think if you’re playing it right, you can have a really, really big edge.

Andy:
Yeah. Honestly, Max, when we were talking about this before, one of the things that I immediately thought about was the bankroll management and contest selection video that we put out maybe a month ago for baseball. What we say is to maximize the growth of your bankroll, figure out how much you’re able to safely wager on any given slate. And then, from there, determine the contest to play. So if you’re going to be betting 20, 30, 40, 50 bucks in DFS, you probably shouldn’t be putting that into one of the 150-max $15 contests.

Andy:
If you just want to gamble, that’s fine. That’s going to have the best sweat for a massive payout. But if you’re trying to grind up some profits, that’s just not the best place to do that. You should be looking at how you can get your money down at the places where it has the most edge. For betting, that’s going to be prop bets, but they will have lower limits. Again, if you’re not betting thousands a day, even then, if you have access to a book in a regulated market that doesn’t limit too fast or you just have a handful of books you can pick from, you could definitely get down thousands a day in prop bets for a little bit at least.

Andy:
But those bets are just going to have much more theoretical edge because of how the books work. Without going into a super deep analysis of what’s going on behind the scenes, I mean a book maker, they act as a market maker. What they’re doing is they’re not trying to perfectly balance action on both sides. Oftentimes, that will get close to happening, but that’s not really their ultimate goal. What they’re trying to do is they will put out their best number that they could come up with, but what they really want to do is put that number up, hang that number up, and let the market test it and get information from all the bettors to see which side should we move to.

Andy:
Back in the day when there wasn’t much information on the bettors, they were acting more like just trying to stay in the middle. That still would work pretty well, especially when you have vig on your side. But now, the online books especially have so much information about all the players betting, that they can profile those players and they can know how sharp or how square those players are and adjust the lines accordingly. They’ll know like, “Okay, we might be getting hammered on this side, but it’s kind of from fishy bettors. So we’re going to move it a little bit just so we don’t have ridiculous exposure, but we’re not going to go too crazy with it.”

Andy:
When you look at spreads in NFL, when you look at money lines in baseball, these popular full-game markets, they get an insane amount of action, and that makes it relatively straightforward for the books to set pretty good odds. Again, when they do have the vig on their side, they don’t have to be perfect to still come out ahead time and time again and give you bad prices. So the less frequently a market is bet, the more edge there is theoretically there because they don’t have enough information to set it perfectly.

Andy:
So that’s why, Max can correct me if I’m wrong, but that’s basically why you believe the props are the most specific bets you can make. They’re the smallest limits because they don’t get a ton of action. And then as you move up from there, the limits will go up. They’ll get more action, and the lines will get sharper. Is that a fair summary?

Max:
Yeah, absolutely. I mean anything that is small limits, it means they’re not taking that much time to figure out what the proper line is and it’s not getting that much action, so they can do that. So some of these baseball games, especially the money lines, a lot of sports books will take tens of thousands of dollars for one bet. When you have a market like that and you have big players who are willing to bet a lot of money on it, by the time the game actually starts, the line’s going to be pretty good.

Max:
If SaberSim is saying one way and the line’s a different way and it’s right before the start of the game, that’s a signal to me where I’m saying, “Okay, when I’m playing with my daily fantasy lineups, I’m going to move this team over towards that Vegas line because this is a real signal for me.”

Andy:
Yeah. And it’s also, we’ll talk more about the DFS side of it, but it’s like the bigger the predicted edge, the closer you are to lock or to the close of that market when the game starts, the more skeptical you should be. Just because you could blindly bet all of those picks and come out ahead doesn’t mean it’s going to be the way to come out the furthest ahead. What we really do when we’re betting ourselves using SaberSim is we’re not blindly betting it.

Andy:
It will be, by far, the biggest signal we look at is that’s our starting point. That’s where we’re looking at to get the information, but we’re going to do some research. We’re going to dig in to like, “Okay, what’s happening here? This is a big edge on this bet. Is this something that I agree with? Do I think this is something that SaberSim is off on? What could it be?” So, Max, you want to talk a little bit about when it comes down to actually betting baseball, what your process at a high level looks like?

Max:
Yeah, absolutely. So I think there’s two key times for me or maybe three actually, depending on what’s going on. But I think the first key time is when the lines become available, which is usually going to be the night before. There’s some sports books that open them actually about now.

Max:
But, essentially, the interesting thing about sports betting and what we’re talking about in terms of these sports books being able to see the markets, see who are the big players here betting is as time goes on and as the game gets closer and closer, the book maker is going to have more of an edge setting the line because they have people like me, people with different models, people who are smart who are betting this line, and they have to weigh that information.

Max:
So when they have the least information is the night before. But we have the same information the night before and the morning of as long as lineups are out, and we have our model.

Andy:
Right. When the line first gets posted, no bets have come in. We have the same information.

Max:
Right. But we have our model, right?

Andy:
Exactly.

Max:
So if you look at these overnight lines which, for me, again, sometimes I’ll actually avoid it because I don’t want to move the market. I’m going to wait till the morning, where I can bet more. But if I didn’t care about that, I would just start betting overnight. If I could bet as much as possible, I would because that’s going to be the time where the lines are going to be the most off and when SaberSim is very different than the signs or maybe even a little different and there’s just one reason or another you like the game.

Max:
That’s where you should really be having your process of sports betting. Your biggest focus is the night before looking at SaberSim, looking at the matchups, doing whatever research you should do, and then, as we recommend, price shopping, looking at different books, seeing if you can get that vig down to as close as possible to nothing. I can get into what exactly that means if you want me to go into more detail. But, basically, try to get the best line possible.

Max:
If you do that and you’re SaberSim, I actually almost guarantee you’re going to make money as long as you’re not being a complete idiot.

Andy:
In the long run.

Max:
In the long run, right? I mean, yeah, trust me, in the long run. I’ve had bad runs with baseball. Trust me. Baseball is a killer sport to sports bet because you can have really bad runs. You can have really good runs. So you have to be willing … This is why what Andy is saying is one unit, 1% of your bankroll. You are not going nuts because of a two-unit bet. But over the long run, you definitely can pretty … Honestly, I don’t want to make it sound too easy, but it’s like you can pretty easily make money if you know what you’re doing and you’re really actually shopping these lines.

Max:
So I would say overnight is a really key time. I think another key time, especially if you’re on it and you’re doing daily fantasy and you happen to catch your notification, is when lineups come out. So you could be looking at the lines. You know what SaberSim is handicapping these games and what probabilities they’re signing. And then you see, for example, the Angels lineup and Mike Trout’s not in it. If you’re available for that and you already saw that we were favoring the under in that game, well, suddenly, I would just go to whatever sports bet you can find online and bet the under as soon as possible because that line is going to move and especially towards the end of season or maybe there’s a late-night game.

Max:
Sometimes you can find these lineups that really punted, where there’s just not a lot of good players in it and you can get a really big edge by just being in the right place at the right time and seeing that lineup come out and the line just being way off because it wasn’t expected.

Andy:
You say there was the third part. Is that just basically …

Max:
I would say the third is literally on a-

Andy:
Right before the first pitch.

Max:
Yeah. Well, no, no, no. I wasn’t saying. I mean you could do live as well, but I was thinking the first part is literally if you’re looking for the sports book the very moment they open their lines because those are going to be the worst lines. Sometimes I honestly do that.

Andy:
Right. It’s like truly as soon as the [inaudible 00:18:32]-

Max:
You’re refreshing the page. But yeah.

Jordan:
Max, you had talked about how important it is, the line-shopping component of your process. Can you expand a little more on what that looks like, what you’re looking for when you’re shopping lines?

Max:
Yeah, absolutely. I think the common thought with this is you want … Let’s say, for example, you had a game, like National/Reds, and it was Nationals -110, Reds -110. That’s the vig. They’re taking a rake. As long as they get equal bets on each side or they’re on the right line, they’re going to make money. But if you have access to a lot of different sports books, you can find situations where one book has the Nationals +100 and Reds -110. And then another book has Reds +100 and Nationals -110 or 120.

Max:
In that case, you essentially are having a game where there’s not actually any vig, which translation means there’s no rake. It would be like playing on draft games and they’re not raking anything. As long as you’re an iota better than the field, you’re going to win money. The same way if you can just find any edge possible to play this line slightly better than what the book maker has it on either side, you’re going to win money. So one of the real important things in sports betting is finding that situation.

Max:
When I see a bet that I like, I don’t just go to one sports book and look at it. Let’s say I liked the Rays/Royals under today. I wouldn’t just go to one sports book and look at it. I would say, “Okay, let’s look at every sports book I have access to. Let’s see if there’s an over/under that’s maybe eight and a half instead of eight or maybe it’s eight under -105 instead eight under -110. If I can’t find that all the books are uniform, I honestly might wait because the thing is is that maybe I’ll wait till lineups come out or I’ll wait till maybe there’s some market that’s moved because that says to me, “Okay, this number could be right.”

Max:
But if I can just get a situation where I can still bet this number where another book has gone down, I’m going to feel better about that bet because that’s a signal that this bet is good. So we’re trying to balance all the information that we can find to find these bets aren’t just good because SaberSim is telling us they’re good or that we’re telling ourselves that they’re good, is the market is also telling us it’s good. If the market’s telling us it’s good and we’re getting that rake down a little bit in theory, then that’s going to be a really profitable bet.

Andy:
Right. It’s kind of like a weird concept to wrap your head around because it’s not as though one book is offering you those two sides. It’s not as though one book is giving you that big free bet, but it doesn’t actually matter. As long as there is that artificial market, there are those two sides that exist that you could actually bet on both sides. You want to make sure that it’s a book that people can bet on and can get a good amount of money down on because it’s one thing if there’s a book with a 10,000 limit on a money line and another that just has a really local shop and you get down 100 bucks.

Andy:
Those are comparable unless you are just truly betting $100, and then you can consider that. But a book actually, one sports betting book that’s actually good, there’s a ton of just junk out there. A lot of the books you can get some decent information from. But the best book I’ve read in the last five years, I thought solid through and through, is The Logic of Sports Betting by Ed Miller and Matthew Davidow. They talk in-depth about this artificial market component.

Andy:
This is something that it’s hard to overstate the importance of because if you can just figure out whether you’re in Vegas and there’s sites that will show the live odds from all the different books, or whether you are in a jurisdiction where you have access to a lot of different books, that alone if you’re diligent about shopping for the best lines can be enough when combined with a basic model, not even the best one, just a decent model. That can be enough to give a significant edge, and it’s something that no one on the casual side of it really, really thinks too much about.

Max:
Yeah. I mean I can’t recommend that book enough. But I mean just to hammer this point, I listened to a podcast recently where there’s a guy on who is a professional sports bettor and he straight-up does not have a model. He doesn’t have a model for any sport, but he can actually make money by just looking at the sharp books, looking when the lines move, and doing what’s called steam chasing, which is basically looking at a book that you think takes really sharp action, seeing them move the line and then betting at a different book.

Max:
So there is I would say honestly almost just as much edge to be had line shopping currently and finding lines that are good just as much as having a good model. So if you can master those two things, you can really find a very, very good edge.

Andy:
Yeah. One thing to keep in mind is that this steam chasing and that sort of thing is not as easy as it used to be. It used to be in the beginning of especially online betting the books were, frankly, bad. It wasn’t clear who the best books were, so they were kind of all setting their own prices. A big bet would come in on one site and the other ones were either slow to react, wouldn’t react, whatever it may be. If you did some research and were able to figure out which types of books take the sharpest action, you could get a good amount of action down with real edge.

Andy:
If you see at Pinnacle or something like that, it goes from -110 to -115 or 120. You can be pretty sure that someone sharp is hammering that side of the bet. If someone else leaves up the lines still at -110, you would want to take that. But this is where you do want to be careful if you are betting on regulated books is that they don’t like that. The ones now that are slow to move their lines are the ones that are quick to ban anyone that they think have an edge.

Andy:
So one of the questions we get a lot is just, “What book should I bet with?” So one, as many as you can because that’s how you’re going to get access to the best lines. It might seem tedious, and it is. But that is, again, as Max said, one of the biggest parts of being a profitable bettor. But, two, we just strongly say wherever you can bet legally. I won’t pretend there aren’t reputable offshore books, and we are familiar with some of those. We’re not going to explicitly recommend them. But I would say if you are looking into some of those offshore books, there is a site called Sportsbook Review.

Andy:
Go to them and, frankly, I would only bet on any book at probably A+, and there’s a handful of those, maybe A. But in the offshore market, they’re banning people. They’re not paying people out. You just have to be really careful with those. So you don’t want to go too far with getting access to all the different kinds of lines and pick books that aren’t going to pay you if you win. So that’s why you just have to kind of straddle that. It used to be a lot easier than it is now. But there is definitely still a lot of edge there.

Andy:
I think as more and more states roll out regulated betting and also just as more and more books come up because of that, I think it’s going to get better in the upcoming future as well. So it’s something definitely to keep in mind. Max, are there other points that you want to touch on? You’ve talked about line shopping. You’ve talked about choosing when to bet. But when you’re actually using SaberSim, what is going through your head when you’re determining which edges to actually pursue, which bets? You’re not just blindly betting the board. How are you deciding which ones are worth it?

Max:
That’s a good question. I would say I mean can we actually put up the sports betting page again? Because I really want to highlight something that is something that I look at too.

Andy:
Yeah. I can get that in a second.

Max:
Because I mean I think you can use it for two things. You can use it from a model perspective, and you can use it from a line shopping perspective actually too. So I’d say the first thing is so one thing that I discovered through betting quite a bit and I mean there certainly is a reason why this is and we can get into it. But over the course of my history betting, the most money I make is on unders. I think the reason for that is despite the fact that these book makers don’t try to have a bias one way or another is baseball games are not normally distributed.

Max:
What that means in real time is is if you have a game that’s an over/under of eight, you’re going to have a lot of outcomes that are five. The total will end up like 3-2, 4-2, 5-3, 5-2, so on and so forth. And then there’s going to be those games that are 18 total runs, 15 total runs. So if you’re trying to imagine the distribution of these games is you sort of have a clump that is that these low run total numbers and then it sort of tapers off. But it raises the average runs because you’re going to have some really blowout games.

Max:
Because of that, unders … By the way, this was also true during juiced ball seasons. Literally, in 2019 or 2020, I forget which one it was, but during seasons where overs were crazy, home runs were way up, I was still making money on unders. So one thing that I’m always looking for is when is SaberSim telling me that an under bet is good? Because given my history, that’s the ones that I’m trying to favor. Usually when SaberSim likes an under, I usually see line movement towards it as well.

Max:
So I’m sort of looking through this to just see if I can find any under bets, basically. That’s the thing that I’m really focusing on. Danny and I also love looking at Statcast data. So we’re looking at different things. We’re looking for velocity increases or decreases with pitchers. Sometimes Danny just is looking at data and he tells me, “I really like this pitcher.” I say, “Okay, I trust you. You’re very smart.” So there’s some things that outside of SaberSim we’re just having sort of favoritism for.

Max:
But so we’re looking through this. And then, let’s say we find a bet on the under. So actually, this is a bet we didn’t scrape for because it’s a double header seven-inning game and it’s confusing to do it. But one game that I actually bet the under on today was Robbie Ray, Jordan Montgomery, Yankees/Toronto for the second game of the double header. That was a game that Saber, if you have the total of the teams, we’d say it’s about seven runs. I like both of those pitches.

Max:
So it was just sort of a bet that I just liked. And then I went and line shopped, found a lot of the books were seven under -110. I found one that was seven under -105, bet it, and that was the process that I went through. But so one thing that you’re really looking for is you find that you’d identified a bet you might like. And then what I like to do is let’s say you have something that’s Milwaukee/San Diego where on Pinnacle we have the over/under is eight and a half, and let’s say you want to bet the under of that game.

Max:
Well, for that game, we have about 0.2 units for under eight and a half, -115. But if you go to, let’s say, Bovada and you look at the game, they have under eight, -105. I don’t know if you have this up. I’m just looking at this on my screen. But if you went to Bovada, they have the total’s actually eight and it’s -105. We recommend zero units on the bet. So what that says to me is, okay, it’s not that clear at face value is which is better, under eight, -105 or under eight and a half, -115?

Max:
Well, according to our numbers and looking at our simulations, which is information that not a lot of people will have, that eight and a half over/under line is actually a little better. So I’m going to then, if I have access to both these books, bet on the Pinnacle line and make a little theoretic money on betting the better line. So that’s something in my process as well.

Andy:
This is actually something that I can’t give a timeline for when it’ll be ready because it’s weirdly more complicated than people might think. But we are going to be allowing for people to change the odds. We have a strong date source for the odds, so they’re very quick to update. But still, lines can move faster than we can update. So you want to make sure that we are … When you are betting based on one of our projections, that you’re looking at how we evaluated it against the odds that you actually have access to.

Andy:
As part of that, we’re also going to allow you to change the total so you can see how that changes things because, one, different books will have different totals, at least temporarily. But, two, a lot of books will also have alternate lines that you can bet. They’ll put it, especially at a book like Pinnacle, but also on DraftKings and other books, they will have … For this, they’ll have eight. They’ll probably have eight and a half. They’ll probably have seven and a half as well.

Andy:
They’re going to have different odds for that and you can see which one is the best bet because they’re not all going to be the same. Max, I’m curious. Do you think the edge on unders, is it purely a market thing, like people don’t like betting unders? Or do you think there’s more to it than that?

Max:
I think people don’t like betting unders. I think that these books might skew it a little bit and people don’t realize. I think maybe people have model bias because they’re looking at some model or they’re looking at something and they are looking at averages instead of actually the distribution. Since we have access to just actual distributions or simulated distributions, I think it has something to do with that.

Max:
It really surprised me because even in years where scoring was up, it was still the case. So I think that’s that. I guess the other thing which I think is really important is usually the things that can go wrong in terms of what happens leading up to the game is to the downside is you can be projecting an over/under for a Yankees game and then Aaron Judge might not be playing. There’s a surprise to the downside. Usually, the lineups when they come out, they come out in a worse way than you’re expecting. So that’s something that I’m looking for too is a lot of times in these double header games, you have the first game of the double headers.

Max:
These games play their normal lineups. And then the second game, they just punt their lineup. So that’s something I’m looking for too is trying to predict, “Hey, when is this team going to actually come out with a worse lineup than people might be thinking?” That’s going to be a really good time to bet the under.

Andy:
Because almost no model, no book is going to put up that original line based on just a shitty lineup.

Max:
Right.

Andy:
They’re going to assume it’s going to be at least average.

Max:
The normal lineup, yeah.

Andy:
Yeah, exactly.

Jordan:
Max, is this under trend that you’ve noticed something that’s baseball-specific, for the most part, or is this something that you’ve seen successful in other sports?

Max:
I have not betted enough in other sports to know. I have enough of a track record in baseball and I’ve spent enough time tracking it that I’m very … It’s like I can’t believe it. It’s every year. It’s always the case.

Andy:
Just to clarify as well, this is not Max saying, “Bet every under.”

Max:
No.

Andy:
It’s that you’re going to be more likely to find edges on unders than on overs. He’s specifically looking for those, but he’s not just going out every day and hitting the under.

Max:
No. Not at all. That’s absolutely not.

Jordan:
Another question we get pretty often here while we’re looking at the board is how to handle some of these split bets. I don’t think we see any here, but sometimes you’ll see a suggested bet on the money line for one team and on the spread on another, or on Pinnacle I noticed we had a couple points on the under for the full game for the Padres/Brewers, but a tenth of a point on the over for the first five innings.

Andy:
Yeah, I think-

Max:
Sorry. Andy, you can go.

Andy:
Yeah, no. I’m curious to hear your take on it too. But just having worked a good amount with Matt and now with Wil on the model and following it a lot myself, this is you’re getting SaberSim’s opinion a lot of times on the bullpen. Also, because we’re simulating every game out, in those simulations, we have logic to determine when a pitcher’s getting pulled, how often that happens, and what happens after they do get pulled. So that’s part of it is that we are actually looking at how all these things can shake out and what happens when they do.

Andy:
So the first five bets are going to just be much more head-to-head evaluation of the two starting pitchers. There’s nuance, but that’s a good idea of what it’s actually measuring. Beyond that, it changes a lot, if there’s pinch hitters coming in often, how people manage their bullpens and a lot of that. So that’s what can cause that theoretically. But then in practice I mean, again, these markets move independently as well. That’s something that I know some bettors … Max talked about people being profitable bettors without a real model.

Andy:
A lot of it can be steam chasing, but another way of doing it is just what they call derivative bets is they’re saying, “Okay, this full game line is moved X. That means the other lines for the first half or whatever should move Y, or the other way around.” They’re trying to find when different bets, different types of bets get out of sync with each other. The reason there can be edge there is because the books almost always move bets independently.

Andy:
So if they take a massive bet on the money line, they’re probably not moving the total. They’re probably not moving the run line. They’re probably not moving the first half money line even. In extreme cases, they will be. But it doesn’t happen that frequently. So the different bets can just get disjointed. So I wouldn’t read into it a ton unless it’s a pretty big unit bet on both sides because it is very common that you’ll see 0.1, 0.2, and it’s a one side for the first half, and then one unit on the other side for the full game.

Andy:
Yeah, I wouldn’t read into that a ton. But if it was bigger bets on both sides, then that’s something where you can be a little bit skeptical and do your own research to see is there a reason that SaberSim has this view that makes sense? Is there a narrative I can think of that plays out this way? Or do I just think SaberSim’s off on this?

Max:
Yeah. I mean I would just add I think our best, the biggest advantage of our simulations are going to be probably the full-game simulations. So with first five, you’re going to be making a bet where people are evaluating a very straightforward situation, which is starter versus starter. It’s the easiest, much easier than trying to evaluate a full game. So I love first five bets, but I usually am betting them having nothing to do with something I’m seeing on SaberSim, just having to do with my own personal bias about specifically liking the starters versus the bullpen.

Max:
Usually I’m going to bet those as soon as they come out because usually what these books do is they open first five the same as the full game line and they might not be understanding that the full game line is the way it is because of a team’s really strong bullpen or something like that or the home field advantage or whatever.

Andy:
Does that make sense, Jordan?

Jordan:
Yeah, yeah. Absolutely.

Andy:
All right. I’m trying to see. Are there any other questions we should get to?

Jordan:
Chris chimed in in our office hours channel and he asked, “What type of bets would you recommend for someone who’s just dipping their toes in and hasn’t bet before?” I know we get a lot of crossover users that are pretty familiar with the DFS side and just trying the sport betting add-on for the first time. What’s a good starting point for users like that in terms of types of bets?

Andy:
I think prop bets.

Max:
Yes. I mean 100% prop bets, especially if you’re playing daily fantasy because I assume if you’re playing daily fantasy you’re not just looking at our tool. You’re doing research. You’re looking at what smart sharp people are saying on Twitter. You’re having your own opinions. You’re looking at matchups. You’re really going into it. If you’re really going into it, then you can, beyond the numbers that we’re already giving you, you can use your own judgment and use that information.

Max:
You can pretty easily dip your toes and make some good money. And again, that’s another thing to favor unders is you like the under of some sort of prop, that’s probably going to be a better bet. When I’m doing props, I mostly am betting unders.

Andy:
One word of caution on prop bets is that … I hope this doesn’t get too theoretical. But to Max’s point on there being more potential edge on full game versus first half bets or first five bets, I think part of the reason is the first five in baseball specifically is an easier matchup to measure. So the average bettor is going to be relatively sharp. Even if we evaluate it better than them, there’s only so much better we can be. But when we are looking at all the other factors that can come into a game, that’s where a sophisticated model like SaberSim can really differentiate from that.

Andy:
So that’s part of it. But I also just think when you’re looking at higher-level bets, so when you’re looking at entire games, it’s easier for not even biases, but just wrong assumptions to average out. Whereas, when you get down to the micro level, when you get down to how does one individual player perform, sometimes wacky things can happen. We’ve really dialed in our models and in the next few weeks, we’re going to start talking about how every day as a team we go through, evaluate the place for the day, tweak things, and just get it even more dialed in.

Andy:
But you just have to be a little cautious on the prop bets because while the odds aren’t going to typically be that great from the book side, they’re going to be favorable for you. If you just see a massive bet, like a 10-unit bet or a seven-unit bet, you want to be skeptical of just maybe there’s something. Maybe this pitcher’s limited and Saber hasn’t fully accounted for that yet. Or maybe this batter has significant pinch-hit risk and we don’t think SaberSim’s accounting for that.

Andy:
These are all things that we do account for, but just one wrong assumption will have a much bigger impact on a prop bet when you’re literally betting on the outcome of one single player versus averaging all that together for an entire team. So, broadly speaking, it’s the same thing as saying, “Yeah, I would play single-entry contests below $3 if I had not much to wager on DFS.” For betting, I would play the most inefficient markets, which are props, but I would always recommend exercising some caution so that you’re not just finding yourselves trapped in something where it’s like we weren’t picking up on the right thing.

Max:
Yeah. I’d love to add to that too, which is essentially if I had a choice and the choice was I had a prop bet or a bet that SaberSim recommended that either recommended one unit on and another bet that SaberSim recommended two units on, the bet that SaberSim recommended one unit on made a lot of sense to me. I got the logic. It was early in the day. I believed in the bet myself for other reasons, just even intuition, just considering it for my own intuition.

Max:
I would rather bet that bet than bet a two-unit bet that I didn’t understand or it didn’t make sense to me because it’s just sort of that Bazian thinking that we want to be doing is you say, “Okay, if I have two sources that are confirming that, okay, this is a good bet, it’s a lot better than just having one source, that it’s going to increase the probability that that bet is a good bet.”

Andy:
Yeah. So there’s a few questions I’m going to try to knock out quickly. So Gun Ranger on YouTube asks, “I found the most successful tool in MLB and NHL. Is there any specific reason why it’d be better in those sports?” Honestly, baseball is, at this point, our strongest model. So that’s probably one reason there. For NHL, I would just guess it’s the least efficient market out of the major sports. It doesn’t get a ton of action.

Andy:
So if you’re able to do some line shopping and not just blindly bet it all, I think you can find better spots just because the markets aren’t as efficient. Do you think there’s anything I could be missing there, Max?

Max:
No. I mean our company’s called SaberSim because of sabermetrics in baseball. So I agree with the baseball one. It’s our strongest model. The hockey point is very good. NFL gets a ton of betting. It gets a ton of action. There’s really sharp action. NHL is definitely a lesser market, so I could imagine that we might be good at it, although I don’t do it myself.

Andy:
And then this is a good one from Marpren that I don’t know if we’ll actually know the answer to. “The Cincinnati/Nationals game is seven innings. If a pitcher pitches all seven innings, will that count as a complete game on DraftKings?”

Max:
Yeah. The answer is yes, I’m pretty sure.

Andy:
Interesting. Okay.

Max:
Although I mean I don’t know this. But I guess is in a seven-inning game the team as probably more willing-

Andy:
You have a shorter leash.

Max:
Yeah. They’re like, “Well, we only need …” After five innings they might say, “Okay, we’re going to our bullpen. We don’t need this.” Especially Washington and Cincinnati, which I think is not a double header.

Andy:
Correct, right. Jordan, were there other questions that people had submitted earlier that we should get to?

Jordan:
No. I think that puts us in a pretty good spot in terms of questions we’ve gotten so far. I know one thing that a lot of users ask about pretty often, on the DFS side, are there signals that we can use from the sports betting app and our sports bets to improve our DFS process?

Andy:
Yeah. That’s actually as we shift into the DFS topic there’s a few points I can hit on broadly. And then Max can get more into the details of what he actually does because he plays much higher volume and higher stakes than I do. But, to me, the fastest way … I’ll share my screen again. So the fastest way to figure out what games and players to adjust is going to be looking at the betting tools. So for games, we actually don’t have this information on our site, but it is something we plan to add.

Andy:
FantasyLabs, they are one of our competitors but they have some great free data. You can get this information without logging in. They have what’s called the implied total, the implied score. You can compare that to what we are projecting each team’s run total to be. If you see us being way off and, again, you can’t find a reason to explain that and you want to play it safe, you can move it towards Vegas. What I typically recommend people do is rather than just assuming Vegas is right is split the middle between them, is figure out, okay, SaberSim might pick up some things that Vegas isn’t and more the other way around.

Andy:
So let’s split the difference and move it towards Vegas without going all the way. So this is just a really quick way of seeing what teams are we out of line on. In the betting tool itself, it won’t show you the exact implied run total, but you can just see any teams that we have a significant bet on probably have two units or above on the totals of the over/under. Those are ones I’d be looking into and checking what the implied total is to potentially raise or lower it.

Andy:
But it doesn’t look like there’s too much discrepancy. Maybe this Arizona game, I’d look into that-

Max:
Can I just add a thing too?

Andy:
Yeah.

Max:
Is you’re looking for if something where it’s both the over/under and the money line. So I’d say Detroit is one, is we have some money-

Andy:
True. Right, right.

Max:
… obviously, on the over and almost a unit on Detroit. So that would say to me, okay, maybe we’re a little low on Detroit, which actually surprised me because Shane Bieber is a beast. But hey, that’s very interesting that the sports book landed on a bigger number there.

Andy:
And then the other thing for players is what I’ll be looking at is Pinnacle is one of the sharpest books just across the board. What they will do is if you go, you don’t have to log in or anything. You don’t need an account. But they have props on everything. DraftKings, for regulated books, has great props. Most of the regulated books have decent props, but there’s one I haven’t found elsewhere and it’s for pitchers. Let’s think of a good one. Who did we say is starting for a game coming up later? Let me look at this one.

Max:
How about White Sox, Dylan Cease?

Andy:
Okay. So we’ll find the game right here. You click over here to open up all the other bets. Go to player props. And then you can find Dylan Cease. What’s really interesting to me is this pitching outs. This is a simple approximation of it. But I’ll say, “Okay, the total’s at 15.5. Divide by three.” So this is saying the middle is going to be about 5.2 innings, but there’s a lot more on the over. So Vegas is saying this is significantly above, probably closer to six innings is reasonable.

Andy:
So then you can go over into the app and see … Let me get to a slate with it, and see what we have as the innings pitched.

Max:
I didn’t even realize you could do this. So I’m going to be using it.

Andy:
Yeah. If you go to view detailed player projections, you’ve got Dylan Cease right here. Innings pitched is about 5.95. So that’s kind of spot on with what Vegas is projecting or at least Pinnacle is. They take a good amount of action on props, so you can trust the number relatively. But if that was way off, if we were at five or we were at six-point something, you could use that as information to just quickly dial it in. We talked about reviewing and adjusting projections.

Andy:
That is going to be one of the fastest ways of doing that. You can sort by value and see if there are any players that we have exceptionally high value that you are surprised about, research their props, and check that out. But, Max, on your own, what does your adjustment process look like when it comes to this sort of thing?

Max:
Yeah. I think if you’re paying enough attention to the lines, that this can be really valuable. Something that I’ll do is let’s say Arizona and St. Louis is something that stood out to me because we have over a unit on the under there. Let’s say in the morning you saw the stat and the over/under was nine and we had a unit and a half on the under. And then it comes an hour before game time and the over/under’s 10. I would say, “Okay, there is a lot of strong action on this over. I don’t believe this is accurate for one reason or another.”

Max:
So then I would look to adjust up the hitters of a team or just use the tools that we have where you can just adjust the run totals and we’ll do that for you. So essentially, I’m just looking for, okay, if I was looking at this line and we had some units on a bet and there’s just real strong line movement towards the other side, that’s where I’m saying, “Okay, I want to take this into account quite a bit.”

Andy:
Yeah. This is actually one of the questions I see is that, how do I know if SaberSim being different from Vegas is a sign to edit projections or it’s potentially edge on the field? There’s no surefire way to always know this without doing your own research. Even that’s a pretty vague answer, but it is helpful to look at how a line is actually moving over time, how we’re changing over time, and use that as a sign.

Andy:
The further Vegas gets away from us, especially as it gets closer to the close, the more skeptical I would be unless I can pinpoint why I just disagree with the market and why I think they’re not getting it right. But I think, yeah, the line movement is something Max and I haven’t touched a ton on. But seeing those trends is incredibly valuable of knowing what to trust. If you don’t have that much of an opinion, that’s okay. You could still look at these as signals.

Andy:
I think the good rule of thumb is the closer it is to lock, the closer it is to the game starting, the more you can trust Vegas, especially when it’s the larger, the projected unit bet is on one side or the other. That’s just how I would think about it. But it’s always important to add your take as well.

Max:
Yeah. I mean if you’re following it closely, it can be a really easy and high-value way to change the projections in a way that is going to be helpful to your lineups.

Andy:
Are there any other tips we want to cover on using betting for DFS? I know I kind of ran through it quickly, so did you. But I do think we got the important stuff out there. But is there anything else you think we should add there?

Max:
I don’t have anything. I mean it just really is as simple as just paying attention to the … If you’re just paying attention to the information that these markets are giving you, what you described, Andy, is very clever and I’m sure with the Pinnacle props, I’m sure not a lot of people do that. So finding ways to just incorporate that into a model that doesn’t incorporate that. We’re a completely independent model, which is very valuable, especially for daily fantasy because that means we’re going to be different than the field in ways that are going to be very valuable to you.

Max:
The edge in daily fantasy is not just picking the best team, picking the team with highest run projection. It’s picking the team that has a combination of they’re going to be low-owned. They have high up side. Their prices are good, all those things. So that’s very valuable and you can add to that value by just actually locating the places where we might be a little off.

Andy:
Yeah. I think there’s a couple of questions. They’re not specific to DFS. We’ll just try to hammer out quickly. And then, Jordan, let me know if there’s anything we missed. But so one was just, “Should I cash out bets I’m winning early if I can make a profit?”

Max:
No. No, I mean the reason they offer that to you is because they want you to do it. If you’re that risk-averse, I would just bet less.

Andy:
I think that’s the important part is you should be … It all goes back to bankroll management. For sports betting, following Kelly criterion and probably not a full Kelly, we can post some links later. But just Google Kelly criterion and Kelly calculator. I recommend a quarter Kelly as a safe balance. But if you’re betting within your bankroll and it’s not some … I guess I could see a case where it’s like some wild long shot and you were betting a right amount. It’s money that would make a material difference to you.

Andy:
Then, sure, cash that out. But, as a rule, it’s not good value. That’s not what you want to be doing. I’d rather bet less, something I can just let it ride. And then-

Max:
Yeah. I see a lot of … I was just going to add, I see a lot of people making the mistake where they’ll do these real long shot parlays and then it will have a lot on the line in the last game and they’ll want to cash out. I’ll just say to them, “Why are you doing this parlay if you don’t want to win the full money from the parlay? You’re just lighting money on fire.”

Andy:
If you get surprised when you actually get there like, “Oh shit, this is too much.”

Max:
I mean like you said, if the money really matters to you, it’s like, sure, cash out. But I just would recommend just making a smaller parlay or something like that and not just making a horrible bet and then trying to cash out when it wins. That’s just a recipe for disaster.

Andy:
You’re capping your up side and that’s not a good thing to do because a book’s not going to let you buy out if you’re losing.

Max:
That actually made me think of one other thing to add, which is there is a little edge. Sometimes you can find parlays that are correlated. One thing that is correlated in baseball in particular, and this is just a fun little tip, is the home team and the under because if the home team wins, it means that you play eight and a half innings instead of nine innings and that means the game is a little more likely to go to the under.

Max:
Some books won’t let you do that parlay. A lot of books will though. So if you have a situation where you’re betting on a game and you like the home team because of the pitching basically, that’s a spot where you could make it get a little bit more of an edge as long as that parlay’s not being raked too hard by just parlaying both the home team and the under.

Andy:
Yeah. One of the questions that comes up a lot is, should I parlay bets that SaberSim likes together? Broadly speaking, no, because you’re going to get worse odds for doing that. But the exception to that is when there is some correlation between the bets and it’s a strong enough correlation where it overcomes the hit that you’re getting to the theoretical odds. The reason that is the edge is because the parlays are priced as individual, independent events.

Andy:
So they might say, “Okay, these two events are 50% likely to happen.” So that would mean that the actual odds of them happening are 25%, but the book is going to price it as though it happens 30%, something like that. So you’re losing money versus that. But when events are correlated, what that means is that if one of the events wins, they’re not independent. If one of them wins, it’s significantly more likely that the other will as well, and that’s not being accounted for in the parlay pricing.

Andy:
So, generally speaking, someone says, “Should I parlay all these bets?” The answer’s no, but you can get smart with it by finding some spots, like Max said. If you just want to de-gen and you truly just want to put down 10 bucks and have a sweat on a Sunday, that’s fine. I’d rather you do that with bets that Saber likes because I guess you have better odds of hitting it, but it’s not a winning strategy.

Max:
Yeah.

Andy:
I think, yeah, we’ve touched on specific sports where there’s more edge in betting and the smaller the markets, the smaller the action, the more edge there is. But something to keep in mind is that those bets almost always have much larger vig and much lower limits and that’s how the book is going to account for the fact that they don’t have a full-time CSGO trader on staff. So, yeah, those odds are going to get out of whack a bit. But they’re definitely going to be softer than a major market.

Jordan:
One other question that came in in Slack more on the app side, Jay was asking about when we might expect to have game totals for the DraftKings book.

Andy:
Yeah, that’s a weird thing from our data provider. We are going to be either switching or adding, in addition to our current provider, adding another provider who has faster and more accurate data from all the regulated books. Frankly, it hasn’t been a major priority because it’s not that straightforward to do. We’ve been focused on some other aspects. But I’m optimistic that by football season we’ll have that in.

Jordan:
Awesome.

Andy:
I’m trying to think. Anything else we should go over? I feel like we covered a lot. I hope it wasn’t too much of a brain dump, but I think there should be some value in there. If anyone has any questions at all, don’t hesitate to reach out. We’re always happy to answer your questions on strategy, really about anything. That’s what we like to do to stand apart from the rest of the market is we know there’s a lot of questions that come up about what to do, why to do it, how to do it, all that kind of things.

Andy:
We want to be there to help. So if you have any questions on betting, DFS, you can always reach us at [email protected] or right in the Slack channel. If you have any suggestions on other strategy sessions, other topics we could really dig into, let us know. I think we’ll have some on smaller slates for NBA coming up either in a week or two. But anything else you can think of, just let us know. But, otherwise, thanks for tuning in everybody.

Jordan:
Thank you.

Andy:
See you guys.

TRY SABERSIM FOR FREE

Start building better lineups right now!

Start My Free Trial