The Risk Takers Podcast
Willing to bet this is the sharpest podcast in sports betting
https://www.goldenpants.com/
The Risk Takers Podcast
Bet Like a Bayesian & SP's DraftKings Beef | Ep 108
This week we learn how to make more money gambling from the lessons of "The Reverend" Thomas Bayes.
His theorem is the backbone of every successful sports bettor (even if they don't know it).
This week we walk through examples of priors, posteriors and general Bayesian betting hygiene. It's more electric than it sounds!
Andrew Mack's Book: Amazon
0:00 Bayesian Thinking Intro
10:05 Bayes in Sports Betting
51:23 News
1:07:30 SP v. DK Pick6
1:18:45 Q&A
Welcome to The Risk Takers Podcast, hosted by professional sports bettor John Shilling (GoldenPants13) and SportsProjections. This podcast is the best betting education available - PERIOD. And it's free - please share and subscribe if you like it.
My website: https://www.goldenpants.com/
Follow SportsProjections on Twitter: https://x.com/Sports__Proj
Want to work with my betting group?: john@goldenpants.com
Want 100s of +EV picks a day?: https://www.goldenpants.com/gp-picks
It does you no good to be 100% accurate with things that agree with the market. All that matters is how accurate you are when you disagree. And so that's where, again, there's a premium on being right on outlier people, just like there's a premium being right on things early.
SPEAKER_02:Hey, what's up, everybody? GP and SP talking about our favorite old reverend this week. The reverend, as we referred to him and his theory, Thomas Bayes. And we're talking about, basically, this is going to be a very, very light to not at all mathematical theory. on Bayesian thinking in sports betting. It sounds intimidating, but it's not at all. I would say it's incredibly intuitive compared to even frequentist statistics. I know that's a big war. This is more SP's world than my world. He's come from the stats wars. I think you've started with the Bayesians, but are you a closet frequentist and what are we going to be learning today that's going to differ from everybody's college stats class.
SPEAKER_00:Yeah, I'm glad you introduced it as a war between the two camps. I actually, in terms of like in the actuarial world or statistical world, I think I would actually be, I know there's some, lots of actuaries are very passionate about Bayesian thinking. I think I'm generally less than that. And maybe this is like a hot take, but in general, I think either can be appropriate and they will converge to similar points depending on the problem you're doing. So we could talk about that a little bit more. But yeah, I think it's a very helpful Bayesian thinking. I think it's just a helpful construct to think about life and embedding. The idea here, if anybody's not sort of familiar, is is really, when this is talked about, it's usually referred to as like incorporating some type of prior belief into like a probability estimate of something going forward. So that's sort of like the crux of it. And this is really important in betting or any sort of like prediction type thing, because oftentimes you have relatively minimal data. And that is where Bayesian thinking will be at its strongest, when data and data points are more limited. I
SPEAKER_02:agree. When we were briefly chatting before the show, I said this reminds me of Kelly, vibes-based Kelly, as you coined it, a while back. And how... If you were to use Bayes' theorem, and we can might as well get to that right after this, but it's kind of like it works really well in a vacuum, but conditions have to be really nice and tidy for it to apply perfectly to the problem at hand. But the general concept of the theorem is, And the ways that you do things that you would say are Bayesian and staying true to the logic of Bayes' theorem is, like you said, very useful in betting and in life. And it's the same way in understanding what Kelly is, is really helpful. And it's less so applying Kelly religiously, like the exact formula to your bet sizing. And it's more like understanding why what makes Kelly tick, why it works and how to kind of use that as a framework to think about just like risk and position sizing, right?
SPEAKER_00:Yeah, exactly. So maybe just to give maybe like a context on like, A non-sports betting example, and I didn't necessarily think of this ahead of time, but just to give how the thinking can work in real life. Let's say you had a friend stay over your house, and then a week later, you notice he's wearing a shirt that you have the exact same shirt of. A non-Bayesian way of thinking about that would be sort of like, Let's say you're trying to estimate what's the probability this dude stole my shirt and took my shirt and now has my shirt. I think the right way to think about these types of problems through a Bayesian lens is this person is my friend. How trustworthy are they? These things should be part of the decision-making calculus. You should have some sort of prior view of this person, and that should sort of weigh into the ultimate conclusion. probability that you thought this person stole your shirt or whatever. If it was a random person, you should give them less, um, benefit of the doubt. It actually, you know, it reminds me of some of the stuff going on on Twitter of, of people complaining about like, oh, you're defending your friends on these things. Like in this whole poker scandal, what there was, that was a lot of
SPEAKER_02:this,
SPEAKER_00:you know, like you're, you're, you know, there was a lot of the commentary of, of like your step, you know, um, you wouldn't be saying this if, if it wasn't so or so person. And to me, that's just like a gap in Bayesian think, or, you know, that type of thinking, like you should have a prior on someone, you know, and if it's a totally random person that you should view, you know, the likelihood that someone, someone like that is acting nefariously different than your friend or someone you trust, even if they do the exact same thing. Right. And so that's, that's like maybe the simplest real life example is, you know, to incorporate of like, What you view of the person, the team, the event, the situation should inform your ultimate probabilistic estimate of the event. Yeah,
SPEAKER_02:yeah. I think that that's spot on. And just to go over the core parts of Bayesian thinking, You have basically– what you're saying, if we were to use that example, your opinion of your friend is your prior, right? So there's a couple key terms that we'll probably use in this episode. So you basically have your prior, your prior belief, and we'll touch on priors more because I think they're the most– controversial or non-frequentist part of Bayesian statistics because they are or can be subjective and that's fine. But in this example, SP's friend, you have this opinion of him. You think, I've known him since college, never seen him steal anything. You have all these all this prior data. So I haven't seen him steal anything. He's been very honest on all of these examples. Then you see him wearing the same shirt that you have. Now, what we didn't know is, did you check to see if you still had this shirt? I don't know. That
SPEAKER_00:would seem like an easier
SPEAKER_02:way
SPEAKER_00:to solve the problem. In this example, let's pretend you can't. Okay, you can't
SPEAKER_02:check. So seeing your friend wearing the shirt is what in the theorem is called the likelihood, but that's just like the new data, the data that you observe, right? And so the new data is you see your friend wearing the exact same shirt that you have, and maybe it's even kind of a weird shirt. So like that, and that's important, you know, is it a, just like a gap t-shirt or is it like this, like a third division, like Russian soccer team shirt, you know, which that would matter. And then you combine those two and you get what's called a posterior. which is your final belief, taking your prior. And then people will say like updating it with the new observed data and then have your posterior belief, which is your new belief. So in this example, you're saying like your prior on your friend is very, very strong. You think, you know, that's a honest person. He doesn't seem to have a reason to have stolen a shirt, you know, um, Whatever it is, and just seeing your friend wearing that shirt is not going to, what you'd say, adjust your prior so much. So maybe your prior was, I think there's an only 5% chance that this person or a 2% chance this person would ever steal from me. And then you see him wearing the shirt. And then you update it and maybe you're like, okay, now I think maybe there's like a 5% chance. I still think it's very low because I trust this guy, but that's the new posterior probability in that example, right?
SPEAKER_00:Yeah. So maybe to get out of my contrived shirt, that not well thought out example at all, where you could just check your closet or whatever. Like in sports, I think to me, the easiest way to sort of explain this is is a simple sort of football power or any sport, but let's just use football, like a football power ranking or rating either or sort of model where you think team A is maybe three points better than average, but you don't know exactly how much better than average they are. Let's say it's the first game of the season. You think they're going to be better than average, but you're not sure exactly. But you think on average they're going to be about three points better. They could be four, they could be five, they could be two, they could be one, but you think they're better. And they're playing like an average team. So their sort of mean performance is distributed around zero. And those two teams play week one. And let's say the team that is supposed to be average beats the team that's supposed to be a little bit better than average by 30 points. Well, you wouldn't just immediately say, okay, this team is 30 points better than this team. That's the one data point we have. We need to adjust the rankings. No, what you'd likely do is you'd want to move the distribution of where the supposedly above average team a little bit to the left. So maybe they're now distributed. Maybe you're guessing they're one point better than average. And the team that beat them, maybe you're guessing is two and a half points better than average. Something like that. The movement is the key and sort of the science of this. But the idea is that you don't solely react to the data you see, right? You're reacting and moving some amount of weight. And the other component of that that I just wanted to mention is one of the big benefits of Bayesian thinking as opposed to some of the frequentest way of thinking, and we actually ran into this problem a little bit when we were discussing the modeling episode and the distribution episode, is Bayesian thinking is like... it's always probabilistic in nature. There's much less you have to do to try and infer distributions or probabilities. It's just core to the methods. And so that always gives you a whole distributional curve of it rather than just point estimates, which is generally more how some other methods work.
SPEAKER_02:Yeah, actually, I had a real-life example of this where I was trying to judge, basically, if my handicap in golf was low enough to enter this one tournament. It was the South Carolina Mid-Am, and they say you can apply if your handicap's up to a 10.4, but they'll only take... a hundred and something people. So then I was like, well, will I, if I apply, get in? And then I had to be like, well, okay. I think like a scratch would definitely get in. Like that is my like a hundred percent. And like a 10.4, a 10 probably would not get in at all. And then I have a curve, right? Which like probably is like, starts to kind of coalesce around like a six or a seven. And I'm like, okay, that is I think the most likely probability, but there's a chance that it's actually a six or a seven or an eight, you know? And you think about it in terms of like the curve of, instead of like you said, a set point of like, this is just the answer, but you're thinking like, well, yeah, 95% of the curve is to the left of 10. A 10 handicap, I think, is only getting 5% of the time. Then you start to build your estimate that way. That was just a really good look into how much of a nerd I am when I just take up regular problems and overthink them drastically using some type of probabilistic thinking.
SPEAKER_00:You also probably get an aspect of People doing what you're doing to some extent and saying– you actually might get these spikes because there's probably people who don't– like certain handicaps who don't even apply because they think they have like no shot and it could actually be like a game theory type of thing. Because like if I'm a– I don't know what you said, 10.4 or something with the
SPEAKER_02:– Yeah, I think that was the ceiling, 10.4.
SPEAKER_00:Yeah, so if I'm a 10.4, like there's no chance I'm probably applying. Right. And similar– and you could make that– that's sort of like induction argument down a white ways. Like if I was a 10, um, I wouldn't apply probably. But, and so it's actually, so like saying
SPEAKER_02:like nine could be like, really?
SPEAKER_00:Yeah. It's probably not even like cut off. Yeah. Yeah. Interesting. The way you, you probably think, cause you have like a selection thing. Right. I do want to get to, to some of like the, um, betting stuff yeah like how to actually use this stuff because i think so far we've been fairly theoretical and just talking about it but maybe before that i just want to give like maybe a couple more really one more example that that hopefully people are somewhat um familiar with that um hopefully makes it less intimidating all this stuff we're talking about if it has been uh confusing or intimidating up to this point but um for anybody who's like familiar with chess and or lots of video games like use a lot of the rankings and that uses like some type of elo model um which is like a way to basically you know you have two people play each other in something chess is the the classic thing this is applied to everybody has a chess ranking um like a numeric ranking and you you if two chess players play each other um you compare the ranking and basically update the ranking based on who won in the relative ranking. So basically if I'm not an expert in chess rankings, but if like a thousand rated player beat like a 1200 rated player, that thousand rated player is going to get a bigger bump than if he won versus another thousand rated player, right? He beat up a more competitive player. And so this isn't necessarily... the direct application of Bayes. And I think that's, but I want to talk about this because I think there are complicated, in my opinion, like statistical methods to apply Bayesian like inference and modeling. I think that's probably like not the, what we want to be talking about for the most part here. I think, you know, from my perspective, what I want to then move to is like, How can people actually be doing this in the context of some of the other episodes we've done and more basic start to thinking this way?
SPEAKER_02:Yeah. Yeah. I think this will pair very nicely with both of the modeling episodes. But I think also the first episode, when you start to apply that to low data problems, which I think we both agree is a great important caveat to kind of split your techniques because low data problems are, you know, a Bayesian way of thinking will be a lot. If there is an area for it, like that's the area. I think I wanted to maybe, I don't know how you wanted to kick this off, but there's like a very common low data problem in golf. that I could lead off with in
SPEAKER_00:terms of like how does- I think examples work better for these types of things.
SPEAKER_02:Think about it. Okay. So in golf, you have like there's different tours. So before you get on the PGA tour- or the Euro Tour, although the Euro Tour is now kind of like a development tour for the Euro Tour, you'll play on a tour called the Corn Fairy Tour, or even below that, maybe like a Latin America tour. And there's all these minor league golf tours, right? Or you play in college and you go right to the PGA Tour. So the PGA Tour has a very rich... database of shot data and all this stuff. But the lower tours don't really have that data. But every year, there's a graduate class of depending on... depending on a couple of things, but there'll probably be like 20 to 30 new PGA tour players a year. And that's a significant amount of players. Now, some of them might've played a bunch of tour events, like a Luke Clinton while he was in college or whatever, but most of them, you kind of don't have a ton of high quality data on. So you need to know what to do with these guys when they first come up and you'll have some, some understanding of like where they rank, but you might not know their specific, uh, skill sets you might know their overall skill kind of so how do you go about quickly um identifying like what they're good at and what they're bad at well you could just right when they play their first pga tour round like let's say they gain um two strokes off the off the t or two strokes on approach now you could be like well that doesn't matter i'm gonna wait to 10 rounds from this guy because I need some kind of big enough sample. And the problem with that is that while you're waiting, you're running a sim on the full tournament. So there's better things to do than wait, but not necessarily the Your intuition when you're saying, I want to wait is not wrong because what you're saying is like, well, that's one round. There's a lot of variance and it might not actually mean they're that good at approach. So what you can do is you can say, well, my prior belief of how good a person that comes up from the Korn Ferry Tour onto the PGA Tour is at approach is like a negative point. negative 0.8. Then you can update it. How much weight you want to give the prior and the likelihood is... How much weight you want to give the prior and the observed data, that's going to be something that you can tweak a little bit. It's a better place to start because it's not random. Picking your prior a lot of the time is going to be based on a bigger kind of general sample that you can apply to the single player. So like, what is all the Korn Ferry Tour graduates? What's like their average approach scale? Okay. And then you can start updating it. And the difference between doing that and just waiting to collect your data is that like, by the time you get to enough new data that you would start to feel comfortable, if you were using a Bayesian method with a prior that made it any amount of sense you would have gotten to a point quicker where you had useful information
SPEAKER_00:right because because in sports betting so much of the betting opportunities that are good are exactly like relatively early like it is when there's a limited amount of data for for people generally it's not something like in different
SPEAKER_02:oh sorry
SPEAKER_00:Shoot, I realized I was on mute.
SPEAKER_02:No, I muted you by accident.
SPEAKER_00:Oh, okay. All right. That makes you feel better. I'll just keep going then.
SPEAKER_02:Yeah, go
SPEAKER_00:ahead. Okay. In different domains, I think it's totally fine to not use Bayesian statistics. This is where I'm more of in the middle camp that each has a role. But in embedding, Because there's just a large percentage of the betting board has to do with relatively small data problems, it's really important. So I think using your golf example and sort of jumping off of that, if you think about how a lot of people model things right now, and I'm going to use baseball just because I know more about that. I don't know the statistics as much in golf. But in baseball, let's say you were trying to put together a model that's going to predict strikeouts for a pitcher. A lot of how people would do this would be they would get some sort of strikeouts per game metrics or K per nine or all different metrics into a data set and then fit some regression model. And then they're going to use that to predict strikeouts. The problem with that, what you're going to end up happening if you do that and you don't make any considerations for some of this, like this prior belief stuff is you are going to have a disproportionate number of bets on people who don't have a lot of data because you're going to fit a relationship between say like strikeout rate and how many strikeouts a person has like historical strikeout rate and how many strikeouts they have in each game. The problem is like, if you think of like someone who has 10 starts and, just by volatility itself, like those people are going to have strikeout rates that are outside the band of like a quote unquote, normal, stable, um, strikeout rate. So, so what you're going to end up having is you're going to have some rookies who have three or four really good games and you're going to think they're like the best pitcher of all time, you know, and, and constantly be betting their overs. And you're going to have some, some pitchers who get, um, beat up pretty bad in the first couple of games. And you're going to think they're never going to strike out a pitcher again, because you're not, you're not accounting for the fact that like, there's a, a credibility in there, or maybe a better way of saying a lack of credibility in their underlying metrics that you're using, um, to predict whatever you're predicting. And so you need to account for that some way or else you're your bets are, again, because we're betting, like you're betting where you're different. And if you're not accounting for this, this is where you're going to be different.
SPEAKER_02:Yeah, I think actually the key point of the whole episode is like a lot of the opportunities in betting are a lot of the opportunities your model might fly to you in betting. And basically a lot of the interesting things in betting are low data spots, right? Like SP said. So I think you're right. And it's not a frequentist, first Bayesian episode or whatever, but I think you're right. There's tons of frequentist statistics that I use all the time. And it's not like one is technically the be all end all. I think they kind of are super related and whatever. But the reason where... I kind of had this belief that sports betting in a way is like sports betting and trading. These are things that will make you into a Bayesian because it's just such the clear use case of it where you've used the term inference a couple of times. You're not... trying to describe data, you're trying to predict data, you're trying to take small amounts of data, and that's where the prior is going to actually matter. One thing we can talk about is if you have a prior, and this was an example that I think we talked about before the show, but if you go back to the golf example, if I have Rory McIlroy's approach prior, because you might be like, well, What about a player that came off of the Korn Ferry Tour like Justin Thomas? And now he's a great approach player. He's not a negative 0.8. Wouldn't that affect him negatively? Not at all. Because Justin Thomas now has been on the PGA Tour for 10 years. We can give him that negative 0.8 approach prior. And at this point, there's so much data on him that won't even... So a lot of where Bays matters, like SPO is talking about, and it syncs up with where there's a lot of opportunity in sports betting. And I think that's why it's such a match made in betting heaven, because it just threads this needle of that's where the opportunity is, and that just happens to be the area where Bays matters. is clearly better than like a frequentist method or like methodology.
SPEAKER_00:Yeah, exactly. I mean, just this is, this is, this is sort of true in all sports is like, if you just look at it from like, how much do lines move in anything, whether it's the actual like game lines, totals, player props, anything, you know, ignore injury news, ignore all that part, but lines are just going to move further forward. from the open in most sports earlier in the season. And that's just because there's more uncertainty for everybody. And so if you are able to, what that means is if you're able to have, because early in the season is really all priors for a lot of things. It depends what we're doing again, but it's much more priors. Like week one, college football is is basically who modeled the transfer stuff these days. Who modeled the transfer stuff the best would be my guess is effectively what that game is and who's the most on the news and stuff. College football in general, there's 11, 12 games, maybe more now with all the playoffs, but you're going to need the priors the whole season. You're just not going to have enough data points. It goes to show that it is really important in betting. And what I will say is it's okay to think you can't do the prior stuff well. If you want to bet college football, but you're not grinding or have some process to understand transfers and everything. If you had some model from that one six years ago, but it is not equipped to handle all that movement and stuff, that could be okay. But you just can't be betting in spots where you need aggressive priors, like week one, week two, those types of things. So it's okay to have different... If you don't feel like you're capturing priors well or... In the baseball example, for example, let's say I had a strikeout model, but I don't feel like it's properly accounting for actual pitcher skill coming from the minors. If I just assigned... everybody the same skill level, let's say I had no minor league data, and I just wanted to assign everybody as the same skill, like average minor league, that could be okay. But probably in that case, I wouldn't want to be betting like people's first starts or whatever. Because I have no, no, like that's a pretty uncompetitive prior in that case. So I think you need to like sort of have a little bit of a self reflection on like, is what I'm doing for the prior actually better? And it's true on both ends, right? It could be like, is what I'm doing late season better than these people who have reams of data and are like really good? Because maybe you're just like really into college football and you follow the news and whatnot. Then maybe that's a place where it's okay to bet certain things early. But when people who are better with data have eight weeks of data, like maybe you don't want to be betting against those people. So, you know, it's like... attacking where you think your strength is. And
SPEAKER_02:this goes back to our comment, what is your edge? Where does your edge come from? Which, you know, the question that we should always be asking ourselves. But priors, and why I like this is because it's like not, it gives an avenue to the ball knowers to compete as long as they know or where this is where my edge comes from. Because to me, priors disproportionately rewards actual sport knowledge compared to collecting observed data and making predictions based off that. Setting a prior, it's going to always reward you for understanding the actual sport. And you could... Your prior could be like, yeah, I think SP said it perfectly with college football or trades like big off-season moves. Like, okay, this team now has this quarterback. Okay, so how do I adjust this team? Because you know quarterbacks play a massive role in team rating, so that's important. And then you have some belief on not just the skill of that quarterback, but how... they're going to fit into their new team relative to their old team and relative to the team's old quarterback. And that's a situation where going into week one, if you were an ex-NFL coach with some kind of basic knowledge of statistics, I wouldn't think it's that insane for you to come to me and be like, I really think... this could be a spot. If there was just some touch of Bayesian thinking involved in it, but because you understand how basically a prior could be built better than somebody who's really a data analytic specialist, week one, week two, I'm on board with with maybe hearing you out and thinking like this is an actual edge whereas if it's you know week 14 i'm less inclined to unless it's like a big trade or something like i'm less inclined to think like your approach is going to beat the big you know the star lizards of the world or whatever
SPEAKER_00:right right i mean if you're not competing with a data advantage, you want to play games where data is sparse. You want to play in the pools where you have as many weapons and tools as everybody else. If you don't have a lot of data, tools, or ability, you don't want to compete in data-heavy endeavors. This is my opinion, nobody could ball knowledge their way to projecting I don't know, some Max Scherzer's strikeouts price better than someone who's using very data-intensive methods in his 25th start this season. There's no ball of knowledge that's going to outdo that. Now, rookie coming up, playing in, let's say, early in the season, playing in Sacramento where the new team is and nobody really knows how the field necessarily is going to play out. I could buy that. I could get more behind that type of thing. I think just competing where you actually want to be is important. The other thing I wanted to say was, this is maybe a stereotype, but in general, whenever I think about this, And I talk to anybody like that and it starts to go in this conversation. I think in general, people who are like new in their modeling, Bayesian thinking, sports betting journey almost always don't rely on priors enough. Like they're almost always too quick. And you see this in like, media, like the way that like non-betters talk about sports, right? Like team wins a game, they're the best, they're going to win the Super Bowl, right? And I think that's like how novices, you know, or people who are new to sports betting, that's obviously like an extreme version. But I think the idea of like what I was saying where pitcher has 10 good games, you're just, you're always betting the over on them. In general, I think people early on do not regress or rely on priors enough. Yeah. And I think the opposite, I think in general, bettors who've bet for a long time or more advanced bettors, obviously, if they've had success, they're doing well. But if they're going to fall on the wrong side, I think they're more often to fall on the side of relying too much on priors. Or another way, I know this is slightly different, but it's similar, at least in my head, is regressing too hard to means. Because I think we've talked about this a little bit, a lot of modeling... techniques that that that people use in sports like the most common ones are not gonna pick up the the outlier when it's actually there very well um and that's sort of by design but um you know it's just i always think about that because i try and think of like where am i What outlier or what am I possibly missing in this situation? Because for me, I think in general, if I'm missing something these days, it's much more I'm relying too much on a prior belief or I'm regressing someone too hard or a team too hard to quote unquote average.
SPEAKER_02:Yeah. I saw that in the show doc that you sent and I was thinking about it where I fall and I'm
SPEAKER_00:curious what you've
SPEAKER_02:done. Yeah. I've kind of tried to, I guess, probably fall on... I think probably fall towards maybe regressing a little bit because you feel like it's safer or whatever. And also, if you're doing something... at scale and you're trying to have everything work really smoothly without having a ton of human input, a lot of times you may sacrifice a little nuance at the edge cases. But I think what I try and do is I try and get off priors as quickly as I can by getting really useful observed data you know doing everything because like you could have what's a good example of this okay let's say you had a a guy come up in baseball and you want you think like oh this is like a power hitter but you don't know if he's a power hitter so you're just like watching how many home runs he hits well that's gonna be like really spiky and like noisy and you might miss like you might just like have a couple of balls caught at the warning tracker or, or whatever that were like hard hit balls. I, again, I don't know the baseball terminology, so I know there's some stuff like this, but you could be like, okay, well, what, what is like the velocity of the ball leaving his bat when he connects? Like now I'm thinking like, okay, I don't need to, I don't need to have like this prior that like, he's everybody who comes up from AAA is like a 10 run hitter, 10 run, 10 home run hitter. When this guy has like a top 10 velocity, even though he hasn't hit a ton of home runs, like his velocity is just clearly at an elite level. Okay. I'm off my prior or much faster. Like I'm off my, my 10, my, whatever my made up prior is from somebody coming up from AAA. It's because instead of using like, basically data that takes a slower time to realize something useful, I found something that does it faster. And I think that's always my battle, is to find that thing that does it faster, basically. Or to create some metric that does it faster than just the raw data. basic data that's coming in. So I'd say I am less of a prior guy, but I'm always cognizant of it and I'm just fighting the fight of not having to be super tied to my priors or tied to them for as short of a time as responsibly possible.
SPEAKER_00:Yeah, no, I think that's a really good way of thinking about it, of trying to identify In my example, and I think we've talked about this before, if certain things are stickier and stabilize faster than others, like in the pitching example, maybe a better way to inform a prior rather than strikeout rate where that's going to take maybe a longer time to get a credible number. you can look at just how fast a pitcher is throwing. That's something that's going to stabilize in one game. And that can help inform a better, more accurate prior, if nothing else. Because I do think, just to be clear, I do think that's the, in general, the side you want to be on probably is the over-regression, the over... reliance on like, quote unquote, like average and whatnot, and not assuming everybody's an outlier. But the challenge with betting again is like, you're betting when you disagree with the market only, right? So it's not like enough to, you know, you have three guys, they're all quote unquote outliers, maybe two actually aren't, right? but you're going to have the most betting opportunity on the one who actually is an outlier, right? And like someone else is capturing that better and you're going to have the most like market disagreement with that. So it's like not enough to like, that's what I think is missed on some of these like modeling, like these public models that I see is like, they might be right on average or accurate on average. Like people will talk about like, whatever, like mean squared error or whatever, like in a model, that really doesn't matter in betting because the only mean squared errors or differences to your, you know, that matter are the ones you're actually betting, like to the market. You could be, it does you no good to be 100% accurate with things that agree with the market. All that matters is how accurate you are when you disagree. And so that's where, again, like there's like a, premium on being right on outlier people, just like there's a premium being right on things early. Cause that's where the opportunity and disagreement is.
SPEAKER_02:I'm probably going to make that the intro. That was good.
SPEAKER_00:Okay. Good. I'm glad.
SPEAKER_02:Do you want to talk about Markov chains quickly or any of the, you, any of
SPEAKER_00:the, yeah, we can talk about something just to introduce that for like maybe people who are more advanced, um, Maybe just before doing that, I just wanted to maybe give a couple warnings if you're going to go down that I wanted to share with Bayesian thinking. First being, and this is sort of what I was talking about, is depending how you're doing it, depending if you're using formal techniques or whatnot, it can be relatively hard to overcome a bad prior. And Bayesian thinking is generally slower to react to changes than other methods. And so... you should want to be sure you want that in what you're doing. There's some cases where that's good. There's some cases where maybe that's not. I think you were talking to good ways to maybe move your prior faster, even if it's manual or looking at things, which I think is important because in something like college football, if you have a bad prior and you're using a classic sort of Bayesian framework, like if we want to talk about which I'll talk about in a second, you're just going to be betting on that team for a while and it might not be good. So the prior is very important on certain things.
SPEAKER_02:Actually, wait, I wanted to put out a disclaimer because, yes, my actual Bayesian modeling approach is that I see what it spits out and then I decide if I want to move the prior or not.
SPEAKER_00:That's, that's, that's the true vibes based Bayesian. That's the
SPEAKER_02:way to do it. Yeah. Anyway, so now
SPEAKER_00:continue. Um, we, we covered the only other things I had, I, we covered like, I would just ask yourself, like, do you actually want to be betting your prior? Do you want, like, do you think your prior is better than the market? We talked about this, like, where does your edge actually come from? It's fine to like have a Bayesian model with a prior and get a nice, pretty plot of your distributional like prices, but if you didn't actually put a lot of effort into the prior, it may not be worth betting when you're very dependent on the prior, which would be early on. So that was, that was one warning. And then I think what comes with that is, well, we'll talk about this cause I'll talk about some of the more complex methods and I'm not gonna get too into them, but like some of these more complex, like Bayesian methods give you, I can, I feel like they can give you a false sense of like certainty because they give you again, like a nice sense, you know, distribution and you're like, Oh yes, exactly. 72% of going over 16 and a half points. Um, but it is entirely dependent on some of the, you know, the prior you choose and how you're actually modeling it. Um, so those were, those would just be some warning signs. Whereas if you're doing like, dirty math, as I would call it, um, or like dirty Bayesian
SPEAKER_02:family
SPEAKER_00:podcast, like a mix of like, you know, just like stuff like that, like more of the napkin back of the napkin math we talked about or modeling we talked about with, with trying to incorporate some of these ideas. You're, you usually have more humility around that. Um, so, um, Yeah, those are some warnings. I guess I didn't really have much to say because there's a book I think that is fairly good that describes this if you want to understand some of the actual– I know we've talked about regression and whatnot. There's other methods to use sort of Bayesian inference and modeling that sort of stuff through a Bayesian framework. I don't actually know the name of the book. It's by Andrew Mack. I don't know if you know the name of the book.
SPEAKER_02:Yeah, it's– I think it's called Bayesian sports models or something. I don't know. But anyway, it's the one book by Andrew Mack that has the word Bayesian and sports in the title. And it's really good. Actually, I'll link it in the description. We should probably have Andrew on. We have no sponsorship dollars for this. Exactly. As good as the book is, I don't think Andrew's getting rich off the BSN. I can't imagine that's the best seller. That was a book for us, maybe for him. We'll link it. Obviously, no sponsorship dollars, but we should maybe reach out to Andrew. He's written a lot of good Excel books. I was actually going to bring that up because it's really not like... for an audio, like a short audio podcast. So it's really hard to, it's basically go into the actual implement, like the true in-code implementation. You're going to want to maybe read that book. And he gives good examples of what we talked about in the beginning that are more kind of fleshed out. And I think he has one about like a hockey goalie in the backup. I forgot. That was pretty good. But yeah, read that book for sure.
UNKNOWN:Yeah.
SPEAKER_00:Yeah, it's not necessarily a light read. It's very technical just to be straight with anybody. If you're not into coding and whatnot, it's probably not for you. But if you are into that, it's helpful. It provides some of the framework about how to think of these problems. And really, to me, it just adds another tool into your toolkit for certain problems of how to estimate some of this and you know, there's different, there's different sort of ways to actually do Bayesian inference in, in coding or whatever. And then he goes through a couple of them in there. So,
SPEAKER_02:right. And it ends with the Markov chain as like,
SPEAKER_00:yes, which I think is, you know, by far the most common or popular way of using, right. Like actual, if you want to not just do sort of like the dirty way of, right. I do think it's worth calling out. Like, I think you can, hack frequentist methods, which frankly are much simpler to most people. I think there's ways you can hack those to capture Bayesian type thinking. One of the things that I wanted to mention, I guess I forgot, was let's say you were doing that strikeout idea again. You could, instead of just having strikeout rate in your linear regression, you could have strikeout rate interacting with number of starts that the player has actually had. So what that's effectively going to do is you're going to place more weight in your regression model, like on the strikeout rate, you know, like for every start. So if it's someone's 10th start, it's going to get a weight of 10 times the strikeout rate in the regression. And if it's someone's 100th start, it's going to get 100 times the strikeout rate. So that's like a dirty non- scientific way of capturing the same sort of things. And you're able to do other things with... I mean, that's a very simple method in regression. There's other things you can do with some other non-Bayesian methods of more non-linear modeling or whatnot that can capture some of the fact that What is the prior? What is a sort of regressed distribution or number versus how much can you rely on the data as of the 10th game, the 20th game, et cetera, et cetera?
SPEAKER_02:Yeah. Yeah, definitely. News? Yeah. Okay. Here's the news. Okay. Okay. Some cool news this week. So we'll kick it off with a couple news. of our classics. Uh, the first is there's hints of that. Fan duels kind of kick the can, kick the tires, sorry, kick the tires on a call. She partnership slash investment. Um, this was reported by, uh, already, uh, front office sports, uh, They actually had some good news on that website. I think it's interesting. I'm not shocked, but obviously it doesn't go into a ton of detail as to how this would kind of play out. I think it's kind of funny if Kalshi's like, we're not a sports book, but FanDuel now owns us. It's just kind of like, we're just playing... this charade that everyone knows is just fake. And like every time something like this comes up, it's just so funny to me because like, why is FanDuel interested? I don't know. Probably because they're offering a bunch of sports bets.
SPEAKER_00:I don't know if you've used any of like the, uh, the sweepstakes books, but like on those, you, you can click the little flip or this switch the flip to go, um, Flip the switch to go from like real dollars or whatever they call it to like fake dollars basically. And I'm excited for this because it'll be like, you'll go look at your bets and you'll click bet tab and you'll have Lakers minus four and then you'll go to the predictions tab. And you'll also have Lakers minus four, but it will be totally different. One will be bettering society and one is just having a bet. So yeah, I think it's a match made in heaven. I mean, yeah.
SPEAKER_02:I think it's important to know FanDuel isn't really FanDuel, it's Flutter and they have Betfair and they have... which I guess is that prediction market. I mean, I don't know, but this is like, this will be interesting. I do think it's interesting. Like what, because FanDuel and DraftKings, and we'll talk about DraftKings and FanDuel in a second, the next item, but FanDuel and DraftKings like have the decision to make as to if they want to, because they have, there is support at the state level of like not wanting to, and the prediction markets to operate just federally with no state licensing or none of the burdens that come with operating the sportsbooks in the state and none of the money flowing back to the state that comes with those licenses. So there is a group that's fighting these and there is a lot of resistance. But it's like if the Andal and DraftKings then decide they want to side with the prediction markets... To me, that would just be kind of a telling sign of where things will go. Because you have to believe that FanDuel is operating with some of the, probably the most complete set of data right now on this of where they want to be. And if they're kicking the tires, I don't know. The only thing that I'm taking away from this is like, it's good news for Kalshi just in terms of like their future outlook because FanDuel has the option to decide to like really put their flag down and fight this. So if they're even like in talks, that's pretty good news for them.
SPEAKER_00:Yeah, I would agree because it seems like one of the bigger oppositions has been like, I think it's called the Sports Betting Alliance, the SBA. And I think that's mostly funded or entirely funded by like made up of like, you know, the, the, uh, quote on the, the normal sort of sports books, uh, for lack of a better word, uh, in the U S. And so if they're jumping sort of ship and saying like, we're not going to be able to stop this, let's just get like on, on the ride. Um, yeah, I agree that that is a pretty good tell of what, uh, where FanDuel thinks, thinks this is going.
SPEAKER_02:Yeah. Yeah. I, I was surprised that, of this, but it did seem like DraftKings was... Wasn't DraftKings like... They had put in some kind of copyright for their own DraftKings.
SPEAKER_00:Yeah, right. I had
SPEAKER_02:the
SPEAKER_00:same thought when we talked about this, using Bayesian thinking. DraftKings probably knows, like I said, more than most people on the actual temperature. If they're getting into this market, I would assume they have a good handle on that it's going to be a good idea. Now, ultimately, they withdrew that, so maybe that wasn't right. But yeah, I just see it as FanDuel getting ahead of this before it continues to grow.
SPEAKER_02:Well, that's a classic DraftKings move though, right? They do something, they withdraw it. They do something, they withdraw it. They do something, FanDuel says they're not doing it, they withdraw it. But in the next news item, it's the flip side. FanDuel decided to do something and DraftKings copied them exactly. DraftKings is following suit with the 50 cent Illinois tax. I think like we talked about last episode, this had to be done eventually. It had to be done eventually. There needed to be a point where the sportsbook said, okay, well, we're going to have to pass this on or whatever. Because until there's resistance, there's nothing to stop the states from just milking the cow as much as they can. So I kind of had the sense DraftKings was going to follow. I think it's pretty clear FanDuel's the big dog in DraftKings is the number two at this point. And certainly, you know, with FanDuel's kind of either implicit or explicit blessing, like falling is probably a– it can't hurt. You know, I don't think they were ever going to– like I think if you remember when DraftKings had imposed something and FanDuel said no– I don't know if DraftKings passing on this one would have had the same effect to FanDuel. So I think they're probably just doing what's best for them in this situation. It'll be interesting to me. I'm most interested to see if this has any effect on Illinois law. But even if it doesn't, it'll have some effect on other states. So there will be... I think a positive for, from this, for the books pockets at least.
SPEAKER_00:Yeah. There's, there's two aspects to this that I think are interesting. Cause I, I also was not surprised by DraftKings. I think we've mentioned that on last episode that we, that we expect them to do that. What I think is going to be interesting and they're sort of related is will any other sports book in Illinois be do this I know the tax is like very targeted towards DraftKings and FanDuel so it'll be interesting because I think if any other book does it and this is sort of to my other point of will any book do it in Illinois but also is DraftKings and FanDuel going to just do this in Illinois and realize nobody cares and then just do this in every other state because I think you know I I If they implement it, it could either be very good, I think, for betters, this whole thing, how this is unplaying, or it's going to be very bad and there's probably not going to be an in-between. If it's very good, I think enough people will complain, Illinois will repeal this or rescind this, and states will get the message like, okay. We've hit our line. We can't be doing this. That would be a good outcome for consumers generally. The very bad outcome, which I actually think is probably more likely, is they implement this tax. Nobody cares. They still place their$10 parlay. It never wins. So it doesn't matter. They don't care about the 50 cent tax. And then DraftKings and FanDuel say, let's implement this in New York and let's implement this in Tennessee and let's implement this everywhere. And there's really no, you know, this is like, they basically got like a free pass to increase the, effectively like the rake, vig, whatever, the takeout. They got a free pass to do it without getting any sort of like backlash, I think. Cause now it can be pointed at the state and it's giving them a free pass to test this. And if it works, like they're just going to do it
SPEAKER_02:everywhere. So that was my, I have a question for you. What do you think would be more, would consumers get more frustrated by just the 50 cent flat tax? Or let's say like the juice, the like the even juice going from like minus one 10 to to minus 115 if their bet size, like if the effective, whatever their bet size is, the effective like increase is 50 cents. So like let's say you normalize them. Like what would be more, what do you think would be more frustrating for the consumer or would cause a bigger backlash?
SPEAKER_00:So that's a really good question. And I honestly think, you know, FanDuel and DraftKings doing it this way, I think their goal is, is to fight taxes and send a message across the country that they don't want this and they don't want to raise prices. Because otherwise, they would have done what you just said, in my opinion. And they wouldn't have even done it in a transparent way at all, like you said. What they would have done is they would have just buried, they would have cranked up the SGP margin from 30% to 35% or whatever. And um, just collected enough there and nobody would be, nobody would even have a clue. Right. So like that's, if they just wanted to like sweep it, like, okay, we're just going to pass it onto the consumer and we don't want the consumer fighting on behalf of us. I think that's what they would have done. But I do think they want, and I understandably, so they want the consumers in this country, um, the betters in this country to fight on their behalf basically. Um, and, uh, and send a message to other states. And so that's why they're trying to make it as transparent as, as possible. In my opinion, I do think Dave, maybe answer your question. I think if you buried it, nobody would know. I do think if you, if like, I do see people angry about like in certain monopoly States where it's like run by the lottery, it's like minus one 30 each way. Like, yeah, I don't think people would be happy about that, but I do think you could bury it in ways that, There's no real uproar.
SPEAKER_02:Right. Right. Right. Exactly. Because their SGP product is such a... It also represents so much of their revenue. That's turning the knobs there. It's all beneath the surface. So
SPEAKER_00:it's not hard. Not only represents so much of the revenue, it represents so much of the revenue that's impacted by this because... Small bets, that matters for small bets. It doesn't really matter for$500 bets. That's quite a small percentage increase. It's like when people are betting$10 or something, that's a problem. Just bury the extra 50 cents in there. Again, nobody would know. It's funny to me because If this plays out this way and nobody actually cares about the 50 cent tax, I don't know what all these people who are trying to proclaim the financialization of sports betting are going to do. Are they just going to put their head in their hands and take their ball and go home? Because they've been fighting this battle of consumers getting ripped off. It's ridiculous that they're getting charged. And if nobody cares... you're in a tough business then because you're selling.
SPEAKER_02:Well, isn't that what happened kind of like with Sportrade at least?
SPEAKER_00:A little bit? I mean, I think that that's the challenge. The exchanges like have or any low margin like Sportsbook has faced is it's just people don't– price is not enough. Like people don't care enough. So this could be like a real nail in the– it could also be like a big– boon, right? It could go either way. But like, if people don't, if you add 50 cent tax to like,$10 better is effective hold, you know, it goes. Or
SPEAKER_02:it goes up because they bet more, like you said, because they just want to reduce the percent of their bet. That is in fact the tax. So it could, it could just be like this weird, just absolute boon for DraftKings and Vandal.
SPEAKER_00:Oh, for sure. But, but I guess what I'm saying is like, if, If you're a sport trade and you see people like FanDuel and DraftKings are the only ones that enact this, and FanDuel's market share doesn't go down, DraftKings' market share doesn't go down, you're in a tough position. That doesn't bode well for you because it effectively means the hold can continue to be raised and nobody cares. They're still going to play on the place that has the best product, in their opinion. If my idea was to compete on price and the price raised on my competitors and nobody cared, I would have to take a hard look at the mirror and say, is this actually what I want to compete on? I
SPEAKER_02:mean, I think you don't ever really want to compete on price. Just
SPEAKER_00:as a rule of thumb. I think if you're selling butter, maybe. Yeah, true. But for what is an entertainment product, I don't know any... It's like saying there's this huge movie coming out this summer. Everybody's excited about it. But there's this homemade movie that I made that's half off. That's not how people consume entertainment.
SPEAKER_02:Right, right, right. And I mean, it seems like Netflix realized and now they just keep raising the prices and no one cares. Exactly. Okay, well, we talked about We teased DraftKings in, but now we're going to talk about an internet shattering story that you were right in the heart of. DraftKings voided some of your and other bettors pick six contests. I guess it was last week. You want to let the audience in on this one?
SPEAKER_00:Yeah, we're not going to turn this podcast into my personal grievances with everything that I've been wronged with. But I just wanted to talk about this really because I think it's related to some of this peer-to-peer or exchange voiding we've talked about where someone accepts or takes a bad side, basically, and do you void that or not, and some of the considerations there. Yeah. I think it's sort of a quick summary. Basically, there were some contests posted. There were lines in the pick six contest that were the highest win rate, I guess, if you want to view it that way, based on what I saw. It was maybe minus 200 to minus 230 price, so maybe around there. What's your
SPEAKER_02:effective price you're getting? On pick
SPEAKER_00:six for that sport. Yeah, right. You're getting maybe like 66%. Yeah.
SPEAKER_02:Oh, no, I mean, sorry. What's the implied on that sport? On pick six. Yeah, for that sport.
SPEAKER_00:It depends on pick level, so I'm not even really sure. I'm guessing it's like somewhere around like 140, minus 140, somewhere like 135, 140. At the lowest, it's probably even higher for other ones. You need to win– like the lowest at like 57, 58, 59%, something like that. So we're talking like you need to place picks at like almost 60% to win and these were priced at like 66% or something. Yeah, God forbid. To be fair, it is outside the norm of like what's usually available. For anybody who doesn't know, what's usually available is just like what's on DraftKings lines. And so I'm sure there's like that– that win margin if you're modeling or something, but it's not like that far. It's not like they're intentionally including stuff that's like minus 200 in their pick six set. So long story short, they, they voided the contest because I think they, they had this issue. And I just think, um, you know, and I, I think I discussed this when we talked about the, this, uh, exchange stuff, I totally get voiding, um, you know, like when you're playing against a sports book for the most part, like I wish there was clear rules on what like a palp or whatever was, but like that is part of the game. It's just like a bad look and bad decision. I think when you're voiding like peer-to-peer type stuff, cause like that's just a bad user, like not even for me, but it's just like a bad user experience for most people. like casuals who may have put action in and then they go to watch the game or whatever and their stuff is canceled and they don't have a sweater or anything or stuff like that. To me, something is either peer-to-peer or it's not. And I mean, this honestly doesn't even work me up that bad because of how DraftKings has... un-peer-to-peerified the VIX thing. So it's not even peer-to-peer anymore. But it is funny how much that game has changed since it started, where the guy, I remember, who was leading the charge when they were fighting the DFS pick-em apps. I literally remember him tweeting out verbatim, DFS is meant to be peer-to-peer. I honestly wanted to go back and find the tweet. Do you think you have a screenshot of it? Tweeting stuff like that out and then to see what they've done to the game. This was just like, you have a child. They've disappointed you so many times and just one more nail in the coffin. That's my only personal grievance I'll share on the show. I
SPEAKER_02:assume we've aired our grievances with the un-PVP a thawing of pick six?
SPEAKER_00:We actually haven't. I could do a whole 45 minutes on the saga of pick six and
SPEAKER_02:get
SPEAKER_00:a lot off my chest, but I don't think anybody wants that.
SPEAKER_02:It could come in in a hot take or I don't know. If
SPEAKER_00:you're ever unavailable, that's what I'll do for a solo show.
SPEAKER_02:Yeah. Well, I mean, look, pick six was a big part of my life last year. Last year, yeah. And it was great. And I loved it. I really enjoyed a lot of the analysis and building of tools I did around PIX6. And a lot of that had to do with the peer-to-peer nature of it. And now it's obviously anything but that. And we won't do it. But I do think what's frustrating with something like this, and it goes back to that tweet that you referenced, and any attempt I've had to talk with my VIP or whatever and be like, the issue is that it's not a peer-to-peer game. It's like if you have minus 200 lines, If it was truly peer-to-peer and you're just taking a rake, then DraftKings would not care. The problem with now what they've done with increasing the floors, and that was really what made it a player versus house, because all of a sudden there was opportunities where you would just win at the minimum payout and everyone would jam in the same place. plays and you could win because there was no punishment for being duped. The problem now is you create all these offshoot problems of now you can only go over on certain props or now certain props have a payout boost. It makes literally no sense within the realm of peer-to-peer. Or this, you now void bets. It's just become like... where you tried to solve one problem, which was like recs were probably a losing too much. I think in my data, I think I shared this with you in my database, but like the rec ROI or like the not, if you weren't like in the top 20 of players in volume, at least for NFL or something, you were losing like 50%. It was like crazy. So one, the recs were losing too much, but two, when they won, they won really small and they complained. And like, I understand that is a problem, but it just feels like now we're here where you're avoiding entries. And like, it was a predictable road. I think you and I both set a time and you said to your rep, like, basically this is exactly what will happen. Like some form of this. And for them to be disingenuous on if it's peer-to-peer or not, or wanting to be peer-to-peer, it's just frustrating because you, one, engage with them in good faith around the discussion on the product, but now they will not basically engage in that discussion because it's become some kind of fake peer-to-peer and it doesn't help them to admit that.
SPEAKER_00:Yeah. I've had conversations with several people who work there in some respect. They've been open to hearing at least me out but nothing has changed, right? So like they've decided what course they want to go on with this. And I think you captured it perfectly. It was one, you know, they had a pivotal like fork in two roads or fork in the road and they decided to go into like the price pick light product. And yeah, so they're just facing all the problems that price picks faces, limiting, you know, pulling stuff off the board, voiding, one directional props, all this stuff. You can only put so much, you know, it's funny, like DraftKings, it'd be crazy. Like DraftKings DFS, salary cap DFS, it'd be insane if you could, you know, you can make 150 lineups. It'd be insane if you could only make, let's say like you enter your 76th lineup and they're like, no, no, no, you've had too much Patrick Mahomes. Like we need to put someone else on lineup. Like, Because it's a peer-to-peer game. You should be able to do whatever you want. But, you know, pick six isn't like that now. It's like, oh, you've hit your limit on this person. So it's just, it's all because, you know, they're trying to protect the soft liquidity. And I'm not necessarily in a position with the data I have. I don't have all the data they have to make that decision. What I do have, the data I do have, and I do collect a decent amount for this. They're not. Protecting the soft liquidity. They are just as worse off as they were before. Maybe they're getting less complaints. That's the data I don't have. But they're just as worse off as they were before. And it's just worse and lower liquidity for everybody else. So disappointing, but such is life.
SPEAKER_02:I loved the NFL picks.
SPEAKER_00:It was fun too. That's the thing that gets missed on it. It's like... a peer-to-peer game is just generally more fun than like prize picks. Like for, for someone who wants to like think a little bit for like a true casual, it probably doesn't matter. And so I get that perspective from them. Like, but like for someone like you or I, who wants to think, or even someone like cat, like there's plenty of people who don't win a DFS, but they like the game of it. Like it's more fun than just like making a, like betting on, the Lakers tonight or whatever, like it's, it's more fun. Um, so that's, that's another, that's just, you know, another disappointing part of it is just less fun. I
SPEAKER_02:agree. It really was the most fun I had, um, in gambling or, or betting in a long time and like a really long time because it, it, part of it was cause obviously it was, it was profitable or there was a lot of, there was a lot of square liquidity, um, obviously, but I, The thinking part of it in the building. It was all a lot of fun. Anyway, did we offer DraftKings if they wanted to? We'll give them lifetime sponsorship if they switch. If they switch pick six back, I don't know if I speak for you or not, but if you switch pick six back to normal...
SPEAKER_00:I'll sponsor. Yeah, yeah. You can just be the sponsor. I'll give a shout out every podcast. Yeah, you put you on the
SPEAKER_02:cover. You don't have to pass anything. Just switch Pick 6 back to what it used to be. And you can be the sole sponsor of the show. No money down.
SPEAKER_00:Correct.
SPEAKER_02:Okay. That's out there for whoever. Whoever from Pick 6 who definitely listens to this. Okay. Q&A?
SPEAKER_00:Yep, let's do it. I think we got some good ones this week.
SPEAKER_02:Okay. So the first one was a buzzer beater that just missed last week, but it's a good peer-to-peer-ish question transfer. So Lady Lado Potato asks, when betting into exchanges, what amount of available liquidity scares you enough into thinking twice about placing the bet? Of course, it depends on the league and bet type. mainline player prop, et cetera. But can you give us some examples of what would or would not cause hesitation to place the bet or get you instead to look to bet the other side on a normal book entirely? I can
SPEAKER_00:take a stab. Okay, you go ahead. So I think we've talked about this a little bit in prior podcasts. And I just want to, because I think we've gotten this type of question and I think it's a totally normal one. But what I do want to warn people of is in these games, it's never going to be this simple. It's not going to be enough to just look at the liquidity and determine, is it good or not? Especially as Profit and Novig and these companies have more and more API people using the APIs. And I've personally seen some of this. It's much easier to... effectively drip the liquidity in. Early on, if you're doing everything by hand, it may have made sense to put, if you want a fill of$2,000, to put the full$2,000 out and just help people take it. If you have API access, there's no reason to do that. What you should do is drip$200 in, let it be taken, drip another$200 because you, A, can see if you're getting filled at that price. And maybe if you get filled at that price, then maybe you can offer a worse price, right? So you can do it in chunks better. And you can basically codify all of this with the API, whereas it's not time intensive to be punching this all in. So I want to say that because I worry that people are getting the impression that it's like, As simple as look at the liquidity. If there's a lot of liquidity, then I don't bet it. But if there isn't, then I can bet it. I think anybody who's posting prices is smart enough to know that people know this as well. I think Alex Monahan and Odds Jam is onto this and is tweeting about this. So if they know about it, then certainly the counterparties on these sites also know about it. And they're always going to be trying to be one step ahead of you. Me, I just sort of think if you are not originating, for the most part, you just have to be extremely careful on these exchanges right now. Just because you don't have a true estimate of the price and anything that's posted there is posted there intentionally by some party. It's not posted by... You don't have a ton of big entities at this point on those sites from everything I can see. You have bettors who are posting stuff intentionally to try and get arbors, top-down bettors, et cetera, to take your action. I don't necessarily just want to talk about it as a dollar amount anymore.
SPEAKER_02:Yeah. My opinion is this used to be... More of a thing than it is now for the exact reasons SP discussed. And to be fair, I have data on this, at least some. And there definitely was a predictive effect of liquidity on a prop, basically, with the P&L of trading into that. prop so that historically was certainly the case but just like with any other game and I think this is this is like you either embrace the chaos or not and this is kind of like you'll find a wall with some people embedding where it's like okay well then what exactly do I do well you know it was like this okay so that means I get this dollar amount I should penny jump them or I should, you know, go bet the other side of the sports book. Well, maybe like it's useful information. Okay. So, but like, what is the exact dollar amount? It's like, bro, I don't, I don't know. You know, it's going to change people. You're going to, you're going to adjust and then they're going to adjust and then you're going to adjust and then they're going to adjust. And the thing is that that's like everything in betting follows that pattern. kind of path so but I do think what is important it doesn't mean like okay then nothing matters the liquidity matters certainly having higher liquidity just in a vacuum if you were to be like should I bet into like let's say you have a$100 bankroll so you're not constrained by Kelly or Betsa you know whatever like you're agnostic to to the sizing advantages of betting into a bigger order. That doesn't matter to you. So let's say there's two props that show value, and one of them has$10,000 in liquidity and the other has$10 in liquidity. Since you have a$100 bankroll, certainly the$10 one in a vacuum, I would say, is a better choice to bet into, as with anything. But does that mean if there's a$500 one and a$10,000 one, well, now like SP said, you could be dealing with somebody who's dripping orders in. Now we're getting into the financialization of betting. There's financial... trading techniques that will start to come into the market. And a lot of execution traders do just that, is you never show your full size on the bid or on the offer. You never execute all in one market order bet. If you're taking liquidity, you're going to do it in some type of VWAP-esque way. If you're posting, it's going to be dripped in and not posted all your whole order right on the offer, right on the bid. That just never really happens in financial markets anymore. And for just this case. But that doesn't mean we can take a few things away. And one thing that Lady Potato mentioned was depends on the league and the bet type. I'd like to make a specific bet. caveat for bet type and it goes with what you're saying about who are the people posting liquidity it's usually not i don't think on these exchanges big entities it's just a lot of like sharp individual people so what in my having met a lot of people who either originate or do well betting their own stuff most of them are doing well in player props Most of them are doing well in player props. So to me, it's a big difference between a mainline and a player prop, specifically on Novig and Profit. The player props, I think, are scarier to me. This is like a reverse. It's like the exact opposite. I think if anything, the mainlines are just going to be connected to Pinnacle or whatever, basically. But the player props on the exchanges, based on who I think the... liquidity providers are, are the most interesting to me because I think that's possibly the area where they're going to be disproportionately sharp. So that would be my only response to this question is like in a vacuum liquidity, Vino liquidity, like it has some effect, but it's less than it used to be. But the player prop thing, I still think based on the clientele is now the key distinction of these exchanges.
SPEAKER_00:All of those points make total sense to me. Like the liquidity, I guess the way I would think about liquidity, like big liquidity is still should be a red flag, but I guess small liquidity should not be a green flag is maybe the way to think about it. Yeah. So, yeah. Uh, to the next one. Sure. I can read this one. I'm curious your, your thoughts on this one. Um, it's from Wolf JB. Can you talk about offshore books in Florida? And he only uses hard rock exchanges in, in the pickums. Um, he wants to make sure his, uh, like his money's good. He's risk averse in general, but wants to make sure it has a, maybe a little bit reluctant to trust a site. He that's unregulated with his money. Um, And basically just wanting to have you talk about offshore and or how to evaluate offshore.
SPEAKER_02:Yeah. I think there's a lot of good... Honestly, there's a lot of good offshore. We can name them. One thing I like about doing this show with no sponsors is we just can name these books. I'm not saying that your money is 100% good with any of these, but these are the ones that I think... are pretty good that you can just sign up for without going through an agent and getting credit or whatever. BetOnline. I love BetOnline. Never had any problems. You'll probably get limited from PropBuilder on BetOnline if you're hammering SGP angle, but overall, for example, I can't bet on PropBuilder, but all of my normal golf stuff and whatever is just I can edit click it in never had a problem cashing out cashing in Bovada that's like a soft offshore book I've I there I still think that you would call them trustworthy but they're a little in my opinion they will sometimes not have a fair ruling in the customer's favor. And then you have Bookmaker, never had any problems with them. And then a newer one, Bet105, I've also had a very good experience with them. They actually, I think I've messaged you this, they hit us up. We'll say no sponsors, but we'll give you the shout out here. Bet105 used you. felt like that's another very good option. And you want to be, in my experience, there's no point in transacting with these offshore books in any way besides crypto. So if you're using crypto and you're depositing, withdrawing to these books using crypto, it's going to be really easy. You can get your crypto from those, at least from BetOnline Bookmaker, Bet105. Although DraftKings has done a better job of withdrawals for sure, but you get it really quick, you deposit really quick, and I've never had an issue with any of them. And I think most people in the community would recommend at least those three. And how do you make a determination of if it's legit or not? I think it's just reputation. for most people who are listening to this show who've done any type of like betting like not DFS 2.0 or whatever but betting they'll have heard of those books and they've been operating for dozens of years and them stiffing you is a bad business decision for them because they do a lot of business and they've been around for a long time. And they've had reputable people who've been semi-public figures in the gambling sphere associated with them. So again, any book can go under, but these aren't credit books that are run by your local bookies. These are multimillion,$100 million companies probably in certain cases. I bet Bovada is worth a decent amount of money. And yeah, I think to me, those are kind of the ones that jumped to my head as been around. I bet 105 hasn't been around as long as the other ones. But recently, I felt like they've done a good job. Yeah. I don't know. I love Offshores. They're great.
SPEAKER_00:Yeah, I think that seems like a good list. It is, I think, much easier now with Bitcoin than it once was. I remember my first gambling forays involved. I think it was called Sportsbook AG, if I remember
SPEAKER_02:correctly. Yeah,
SPEAKER_00:they might still be around. and some skeevy checks. It's a different game. I think the ones who have been around a long time, that's what I was going to... The only thing I was going to add is the longer they've been around, the probably less incentive they have to screw. The longer track record they have of building a reputation that's not screwing their customers over. I think your list was good.
SPEAKER_02:I'm sure I maybe missed one or two, but Yeah. Better to have a shorter list. Yeah, better to have a shorter list. Exactly. Exactly. All right. Let's talk about– look, some people in some states don't have a lot of options, and sometimes they turn to some sketchy stuff. But some of those sketchy things aren't the offshores. Two Natural asks, thoughts on player profit, funded sports betting, and quotes. Obviously their challenge is extremely predatory looking at the rules and sizing, but is it even possible with a good run? What the fuck are they doing with the accounts that actually make it through? So did you look, look into this at all?
SPEAKER_00:I have never heard of this. And I was, I was looking at this while I was, I was looking at this today. This is the most insane thing I've ever, I've ever heard of. So I
SPEAKER_02:had, so I think it's, it's, it's definitely bad, but my, My read on this is like one of my friends who's just so good at sniffing out if something's an end or not went looking into these and decided it's not worth it. And I think that some of it is counterparty risk and then obviously we can talk about it. But it's like you... I keep hearing funded. I get funded trading ads. Is that a thing? Is this a thing that's not in sports betting?
SPEAKER_00:Let me tee up my understanding because you may know more about it. I looked at a website for legitimately three minutes today, so maybe I don't have a good read. My understanding is basically you go into the site, you pay the site some amount of money. In the case I was looking at, you paid them$800. Then you make It's not totally clear to me. It's either 15, like on the site I'm looking at, you make like 15 to 60 bets. And if you hold like 20%, then they give you like$50,000 to bet and you keep 80% of what you make on that. And to be clear, I think where you went was actually where my head went first too, of like, It's not that these companies are predatory or anything. That's not where my head went at all. It was how stupid are these companies that they're going to get angle shot in some sort of way. Because my first thought was you're going to make someone, let's say maximum 60 bets. Again, it's hard for me to decipher what's going on in this one I was looking at. But you are going to have no idea about someone after 60 bets. And my thought was, I'll get me and eight buddies to make 60 bets. One of them is going to be up, hold 20% or whatever. And then I have 200K of their money that I get to hold 80% on and just fire. It's like a free roll. It seemed like a disaster. Now, I didn't actually do the math. It sounds like you had a friend or partner who did it to see if it can overcome the challenge fee, I guess, to pay. This is just one of the...
SPEAKER_02:Well, that's exactly what he... He was like, I'm not fucking wasting my time trying to hold 20%. I would just be ripping. I would just basically just be maxing out variance and then seeing how many accounts can get through.
SPEAKER_00:Yeah, right. If you have enough people, you're going to get some through, right?
SPEAKER_02:Yeah, and the thing is that this could never be legit. This could never be legit because you're either going to get, like you said, it's either going to be legit for two minutes, whatever you're saying. If the offer is able to be plus EV to the better, it's never going to be end up being good for you. And they probably will stiff you ultimately. So I think what this is, is like just some really, really sketchy, like scammy way of booking in my opinion. But I don't, there, there is these weird companies in trading that like would, That you would be like a trader, but then you had to like pay money to trade there or something. Like I can't try and remember the names of some
SPEAKER_00:of them. I didn't even catch that part at the beginning that you had to pay to, like when I read it originally, you had to pay. And I'm like, this is, they're going to be out of business in like 10 minutes. What are they doing? But still the, the, the sort of moat of making you pay. I'm looking like$800 to$1,400 depending on how much liquidity you want to hope that you can hold 20% over 60 bets or whatever. You're right. Effectively, what these companies are is they're booking that people can't do that and just hoping, praying that they don't have to give these people the actual 200K or whatever. Right. I
SPEAKER_02:think basically you'd get There'd be some kind of back off or void or whatever in terms of if you ever found a way to make this profitable. This is the other thing when I hear people come to me with angles. I'm way more interested in a FanDuel angle than I am in this angle because for something to be actually an angle, the person needs to pay. And the people running these companies don't have any money. This is not the type of legit company started by somebody who's really well thought out, understands sports betting, and is doing something that has a lot of longevity and professionalism and everything like that. So to me, regardless of if there's an angle here, there's not actually an angle. Because if there's an angle, they're just going to pull the units. Like
SPEAKER_00:I've been there. You don't want to run up a 200K balance or whatever with someone who has no money. And there's just no way. It's hilarious. If you're listening to this, you should go online. I'm looking and they have like testimonials and they are the most like fake, like AI generated, like headshots of people. No. Sarah L., who's a professional sports trader, says, the challenges are tough, but that's what makes them worth it. Passing the challenge was a huge milestone in my trading career. Wow. Sarah L., if you're listening, we'd love to have you on the pod. If you're out there somewhere, let us know.
SPEAKER_02:Yeah, we would really, yeah, we'd love to celebrate your milestone with you here. Yeah, definitely stay away from these places, like, Don't go near them with a 50-foot pole. My friend who was poking around is very risk-on, willing to get rolled, willing to get stiffed. Just a legitimately good gambler who just has a really nice overall understanding of risk and what it takes. And he's deemed these to be stupid and not worth it. So... I would shout out to Nico the Tico. Let's all follow his path and stay away from this. I appreciate that bring up. That brought me a lot of entertainment today, though. It is pretty. It just feels like it's an Instagram ad with some guy. It feels like it's legitimately the GP Academy video I made, but as a site.
UNKNOWN:Yeah.
SPEAKER_02:Throwback to that. This is the second part to this question. This is actually something I thought, this is an interesting one. Basically, too natural. He says he just turned or recently turned 21. He's gone to Vegas and he's severely underwhelmed. And he thinks casual bettors are... He mentions that MGM, Caesars, and Circa have extremely poor apps compared to FanDuel and DraftKings and extremely limited options for whatever, all this other stuff, which he mentions like parlays and exotics, whatever props, which even his non-sharp friends were really disappointed in, which that's actually very important. And then he says, personally, I'm from Cali. and I would much rather bet there than Nevada, which is wild. Shout out to the sweeps, I guess. Add that to the physical verification, no DFS, and I feel like Vegas is just inconvenient and not what it's built up to be, especially for a new generation of people used to mobile betting. Does this have weight? I'd say it certainly has weight. You nailed it. You nailed it. Las Vegas is not a sports betting town anymore, in my opinion. And it's just because of this. I think it's because they had such a legacy betting, gambling, whatever business, that they tried to halfway mobile sports betting. They went half in the physical location. That's something a lot of... the Vegas apps have, I think, is you have to be physically present in a certain casino to bet on the app. And it just turned into this, they halved it. They didn't really do online sports betting, but there's no real place for live sports betting, or at least they didn't figure out how to make that to differentiate that from mobile sports betting, which I, I mean, I think it's really hard to do because how do you, so they kind of just went kind of halfway and made this like really weird, inconvenient sports betting environment. And Vegas is great for a lot of stuff. But there's a reason we never talked about it on our, when we get asked the question of where would you live to be a full-time sports better? Like it definitely wouldn't be Vegas.
SPEAKER_00:I mean, the way I think about it is Vegas, and I have not been in quite some time, but Vegas is more of like a hospitality town. And the companies you're talking about that are like big in Vegas, like the MGM Caesars type of like those companies, yes, they have mobile betting apps, but really they are, again, hospitality companies, whereas FanDuel and DraftKings, some people might get mad at this, but they're tech companies. Like some people might get mad at that because they think their tech is bad or whatever, and they have to limit and everything. But at the end of the day, they're head and shoulders above the other competitors in terms of their tech they're offering, their user experience. And so that's the difference. It honestly reminds me of what we were talking about when we were talking about that 50 cent tax. It's like the MGM, Caesars, et cetera, are like, still showing Gone with the Wind and FanDuel, DraftKings are showing whatever, Avatar in 3D or whatever. It's just a better, more updated user experience, especially for young people. So I will say the thing that I think Vegas has going for it is it still probably has a lot of network effects as it pertains to betting. Yeah. Bet Bash is there for a reason. I think it's always going to be viewed as a hub for bettors in some respect. I don't know. Maybe this isn't... Tech companies are all over the country now. There's different tech hubs, but I think Silicon Valley, San Francisco still holds some network effect benefits. I think that will always sort of be true with Vegas. Maybe that's wrong, but I think definitely that is still true today. So there's aspects of that that are helpful. You probably can just get in touch with more gamblers there still probably than anywhere else in the country for the most part. Is that necessary or not in today's gambling world? Maybe not, but I do think it has that part going for it.
SPEAKER_02:Yeah, that's a good point. It's the spiritual home of, of gambling at least. And, uh, you need to have a place, right? Vegas is like St. Andrews for golf. You know, is it the absolute best course in the world? I mean, I don't know. I haven't played it, but my opinion from watching, maybe it's not the absolute best, but no doubt it holds a special, it holds its place in the game and, you know, people come back to it. Um, Okay, PoutineSteam, cool name. Want to really highlight the first part of this, which is love the podcast. Thank you. Thank you. Automatically, your question's going to be read. I'm a successful top-down live arb better who is having success using mostly AusJam. I've been testing automated bot betting scraping AusJam EV software cheaper than API. Uh-oh. Don't let Alex hear you. It's done over a thousand bets at positive supposed EV of 2.5% or higher. And I am at breakeven slash slight loss on these. Okay. So basically he's collecting data on historic bets, which is very smart. This is something you should be tracking. And want to use AI to search for patterns in the data to remove patterns. Bet types, betting times, whatever. Basically use AI for filtering. And is this a fool's errand of trying to reverse engineer a top-down edge instead of doing the hard work you do and create models? No, I don't think so. I don't think so at all. I think there's like this... I think that there's like... There's... this concept of like, you're either like an originator and you make, you build bottom up models based on data from the sports. And then maybe, yes, you do regress it to the market at the end, but the cornerstone of your model is sports data. And then there's like, I just use odds jamming, but the thing on the screen, but there's this whole middle kind of type, I think of like being a smart custom model. top-down better. And I think this is like, if you think about someone who's really successful with this, like Spanky, as far as being top-down, but doing it in a way that is really thoughtful and understanding the different effects, which you talk about, betting times, bet size, maybe a book being sharp here, not there, and you know, line movement now using AI to search for patterns. Um, uh, there's, well, this is interesting. I don't, I don't want to get too lost in a rabbit hole, but like maybe you'd be thinking of using a classification model from AI to basically tag bet or no bet. Um, I don't know. I, I, I don't think it's that outlandish to build. If you have, I think a thousand bets is nowhere near enough to train, uh, this model. So I think you'd probably want like hundreds of thousands of bets, but I don't think it's crazy to think that if you have the bets tagged properly with like interesting, um, data like time to post or, um, I don't know, distance from sharp midpoint or size of move or whatever, you do a really good job of building a big database. I actually don't think it's crazy to think there's a black box solution to this type of problem, but it's going to be way higher than a thousand bets. So that's why the most important thing you can do is start collecting your own bets. data or your own alert data or whatever, because this is something that I think is possible, but it's going to take a lot of the right thoughtful inputs. I don't know. Do you have any thoughts here?
SPEAKER_00:Yeah. I don't think it's a fool's errand either. I think this is definitely something that could work. What I would say I think was getting to some of what you were just saying is, the way I would go about it is not necessarily just with the number of bets you have and data you have. And even if I had a lot, I probably wouldn't just throw it blindly into either an AI or some sort of more black box machine learning algorithm. What I would personally try and do And this goes to like the overfitting discussion we've had a couple of times of like, because that's the biggest risk you're going to run if you do something like that without any sort of specification is you run the risk of, you know, whether it's the AI or the machine learning, like saying like, oh, we don't want to bet on MGM at 2 p.m. on baseball games that involve the twins. And that probably is not what you want to do. Um, you probably want to, uh, sort of yourself use intuition and, and, and, uh, common sense to construct variables or features that are actually going to be predictive in, in saying like, this is something I want to bet or how good, like if it's where I'd sort of think about it as if, you know, you have it flagged as like a 2% edge or whatever you can model, like what, like an adjustment to that is based on some other features, um, Whether that's actually going to return 1%, negative 2% or something like that using some other features. But I personally would try and probably be a little bit more hands-on on the feature selection and just use my domain knowledge of markets, what matters, what doesn't matter. Where do I think Odds Jam is overestimating things? That's more of how I would do it. But I think the general idea is certainly a good one. And I wouldn't... you know, the way the question ends are doing the hard work you do and create models. To me, this is hard work and create a model. So I wouldn't like sell yourself short and I wouldn't like over glamorize like a bottom up or model, like a origination approach. Again, I think we've talked about it. Like that gets, I feel like those people sometimes get like put on, on a pedestal or whatever and on like gambling Twitter, but I don't think that's appropriate at all. And in this case, like, I would stick with it. It seems like you're 75% of the way there.
SPEAKER_02:Yeah. I think top-down modeling is like a real thing. And this is, this is something like that. I've recently gotten a lot more exposure to with, with GP picks plus and literally doing this type of top-down modeling. But like, this is a, a skill just like bottom up and it's hard work and like build your bill. You have a thousand bet database that just you have right now. That's great. You know, like keep building that.
SPEAKER_00:Yeah. Because if, if you get the, the other part I was going to say is if you're building the features yourself, a thousand bets, well, it's still not a lot is more than like, basically if you are using intuition, you need fewer records to fit something usable. If you're solely just throwing it in and hoping for the best, you need more records. Right. Right.
SPEAKER_02:And look, I'm the, I'm the, as I've said, dipping my toes into the black box world. There's some, I have some, I'm have some positive feelings towards it, but at a thousand bets. No, you know, no, but you know, at a huge, this is something that I'm going to look into at some point. So I don't know, like if I will, it would make sense for me to even share my results or not, but this is at least if, if, It's a fool's errand if you think that me and SP are fools.
SPEAKER_00:It's an interesting point because I gave that sort of insane example or contrived example about twins on Tuesday or whatever. But in reality, this is true of anything. If you have enough data, you will pick up stuff that is meaningful, which you would never think to model. So I
SPEAKER_02:have an idea of what it could pick up. is it would have this nonlinear representation of edge, where it would be like 1% edge and 2% edge is okay, and then 3% and then 4% model is good. And then as it starts getting higher, it starts dipping. And I think the black box model would
SPEAKER_00:be able to capture that.
SPEAKER_02:So if it shows a 15% edge, the black box model is going to be like, Red alert, red alert, or at least you should test it on that. Because I think we can all in our own intuitive experience, at least our experience of sports betting and know like that's where it could get kind of wonky when you start showing these big edges.
SPEAKER_00:That would make sense to me. And I, yeah, I mean, you will capture stuff. I think you just, for every one thing you capture that is legitimate and In that stuff, if you're not careful, you also will capture stuff that's not legitimate. So you just have to be careful.
SPEAKER_02:And it's all about the features, just like you were saying, for doing even something that's more descriptive with a thousand bets. You still want to make sure all your features are as good as they can be. So just keep building that database is my... you're doing the hard work right now. Like keep doing that. That, that to me sounds like a really worthwhile project. And I wouldn't recommend you go to like leave this project right now and start bottom upping some. Yeah. This sounds like a good project. Good job.
SPEAKER_00:All right. Last one. JS23. Maybe not a question, but could you talk about adding to a position? For example, he had a bet at recorders of a unit at plus 116, moved to minus 125, added more, basically added more to it, moved again in his direction, added more to it. He did mention that at loss, so we appreciate that. We all knew that was true. He was humble bragging about all his CLV, but he did say at loss, so we appreciate that. Basically, when adding to a position, do you look at market, limits, sport, time before the event starts, or any other thoughts about how you add to a position?
SPEAKER_02:This is Bayesian. This is a perfect last question because you have your fair. I assume this is a bottom-up situation because I don't really understand how this would be a top-down spot where you're still betting it at minus 150 from plus 116, but I don't know. Maybe the steam just really steamed or something. Assuming this is a bottom-up situation, and I guess if you're betting... you know, at minus 150, let's say you have a fair of minus 200. And so first of all, maybe you're regressing to the market, maybe you're not. But let's just say for whatever reason you weren't doing any regressing to the market, you bet, you know, whatever at plus 116. And then that book moves to minus 125. So anytime an action is taken, I give that more weight. And this is why I'm a firm believer that like, Line movement is more predictive than stale de-vigging. But the sportsbook sees your bet and they decide, okay, we have this information. So we now have the information that JS23 bet in at plus 116. And we think that they're pretty smart and they're doing their own Bayesian updating. Now they've moved the line to minus 125. But we got to give them credit for knowing something. So something that we don't know, right? It's usually useful. So, okay. So now they've decided, okay, we know this, we got that information. We're putting all of this information together, new lines minus 125, and we've moved it. I'm now, after they've made a move and they've basically made their recalc of their fare, I'm changing my fare a little bit because they've, made an action in response to my bet or my information. Usually, I'm not re-betting on such big moves like this. Let's just assume your fair is like minus 200. I just think in practice, it's fine, right? Because it's minus 150 and your fair is minus 200. But in reality, oftentimes, the If they're moving, their number they move to is usually pretty competitive with your fare and there's not much value there because now you have to remember they have your info, they have whatever they use to set the line, whatever other bets have kind of come in. So once they move, depending on who it is, I'm giving them some respect. I'm probably not rebetting. at minus 125. I'd probably re-bet at plus 110 if it's just like an auto mover. But something tells me this is like not an auto mover if they're going to minus 125. So, yeah, that's my thoughts.
SPEAKER_00:It's hard to give an exact example without knowing like what you're betting, size of the market, all these things. But to maybe also try and wrap it back to a Bayesian way of thinking about it. Like, If you say the price is minus 200 and it's already moved considerably, you're effectively saying when you're betting after it's moved 60 cents that the open price was whatever, 110 cents off market. And that is... Possible, certainly in some markets. It's possible in any market, but it's likelier in some markets and less likely in other markets. If it's bet ESPN, some obscure prop, there's a higher likelihood that their open is substantially far off of what the true probability is. If it's a sharper book in a reasonably regularly priced type of thing, They probably have some... I'm not saying their openers are great in any book or whatever, but you have to ask yourself, do you think they missed by 25%, 30% or whatever your price is? Because you should use it as some sort of data point, like what their open was. Because likewise, they have some process to generate it. It may not be as good as yours, but it is some process. So... Yeah, I think it really just depends on– it also depends– I think of this as a similar discussion as how we've talked about the sort of fractional Kelly in the past of you scale that to how confident you are in your edge and what you're betting. And to me, it's like if I knew with 100% certainty that the fair price was minus 200, I would bet till– minus$199 or whatever. But if I've only been betting this for two weeks, I might have way more margin. If I've been betting this for five years and I'm quite confident in my prices, that's a different story. But that'd be how I would think about it.
SPEAKER_02:Yeah, I think that's a good caveat because I was just thinking there have been situations where I've certainly chased the line this far. But because I know for sure why they're wrong. And it's just some big mistake. So in that situation, I think my fair would probably have to be a little further than minus 120. But like you said, it's actually how confident you are in your fair. And if it's like, oh, this is an obvious mistake because I know exactly what they're doing wrong. And it's not like, oh, they don't know this player is good. It's more like a structural thing that they're doing wrong. They don't realize that the rules are different for this event or something or whatever. It's one of those where then, of course, I'm chasing the line and whatever if it's a huge mistake. But it has to, like SB said, you have to be confident in
SPEAKER_00:that. Right, right. It can't be saying like, oh, I actually think this guy's strikeout rate is 30% versus 25% like that. I'm going to keep– no, it has to be like this guy– like the coach actually came out of press conference and said he's pitching one inning and nobody knows about this. Right. Then I can keep doing this. Right. It has to be something like
SPEAKER_02:that. Exactly, exactly. That's our last question. So I think that's a good way to go out on some– vibes-based thinking around chasing a bet. Any last thoughts on what people should do next in their Bayesian journey, maybe? I think the Andrew Mack book is a good start.
SPEAKER_00:No, I don't really have too much. I would just think about what types of things you're betting and what potential holes or how you can incorporate this type of thinking into your process most simply. I wouldn't immediately try and overhaul what you're doing, but there's probably opportunities or gaps in your thinking where you can rely more on domain knowledge, priors, et cetera, to fill holes. And so that'd be my takeaway. But I thought today was good.
SPEAKER_02:Yeah, this was fun. If you have questions or comments, join the Discord community. goldenpants.com. Just click join and it's free. And then you can go to the podcast questions channel, drop some questions in there. If you compliment the show, we will definitely answer your question, but we actually basically answer every question every week. So if you have any questions for us or feedback, comments, whatever, that's the spot to go. Thanks everybody for listening and we will see you all next week.