Scaling, Good Data v. Good Modeling, and Why Math People Don't Trust Math w/ Sports Projections | Ep 55

Um , I think that your most recent articles have been really , really helpful to me , because you're writing about stuff that I I think that is .

You know , we talked to Mr Limited last week and it's like that next step , like we talk a lot about how , um , what are like the things that can kind of take you from maybe just starting to win to like scaling and having sustainable success .

Uh , and I wanted to hear , like your thoughts about why you decided to to tackle that specific type of content , that like kind of next level past , you know , learning how to win , thinking about next steps .

Um , you know , if I was going to have been on Twitter I don't know , maybe like five years or so with this account and , um , you know , never really made any sort of what I'd call content beyond just the tweets , and if I was going to do something , I would want to do so if that's more evergreen and , um , you know , more technical in nature .

I mean that that's , in reality , like the content I seek out , and there's just not a whole lot of it . And so , um , you know , so much of what comes across my timeline is very , um , I guess , not evergreen , in that it's it might help you in a short term .

Maybe it's picks , maybe it's , um , you know , various angles that are surely not going to last a long time , and so I've been gambling in some form or the other for close to a decade now , and so I just wanted to use my content in the post and whatnot as sort of a creative way to talk about some of the stuff that I think about when it comes to gambling

and wagering on sports .

I try not to do talk about any specific like game or event , because I think the value , like I'd love for someone to come back and like read something or listen to something three years from now .

And you know , of course , maybe we learned something that we didn't know back then and some things outdated , but to be like , oh yeah , like that's an interesting way to think about X , y , z , I've learned something , instead of like it being about the , you know , niners Lions game . You know that maybe we'll remember three years from now .

It's like , oh yeah , I remember they blew that lead . I don't really care about this , you know I don't . So I think , like a lot of the shows I respect are you know we were talking before this like Spanky's podcast . He just is very evergreen . It's not really pick spaced and I love that stuff . Mr Limited's doing stuff like that .

So I don't know , I feel like there's there's some more of that which I love because , like you said , it's the content you seek out , it's the content I seek out . So I don't know , I think it's , I think it has that value and has that longevity . That , I think , compounds that , that value .

I want to , you know , focus on things that are going to teach me things and skills and whatnot that are going to compound over time and be able to be useful in more ways than one , and I think that was a lot of what was in that article as well .

As you know , when , when I'm trying to figure out where to have a full time job , I know a lot of people starting out also do well , they're gambling .

You know you have a finite amount of time and you know , I think people are not necessarily always as thoughtful about where to where to spend that time and don't necessarily think of that time as a as an investment , long term potentially . Yeah , because there's a lot of , there's tons of different ways to make money in gambling .

I feel like 90% of the arguments on gambling Twitter is there are people , one person telling someone else that their way to make money , so there's a billion different ways to make it , and I think for me , as someone who wants to do this for a long time , who's done this for a relatively decent amount of time , I'm more and more try and focus on things that

are going to compound and allow me more opportunities in the future . So I guess I could just give a couple examples

. You know so much of what I see across the timeline .

You know today is like different examples where you're , you're more so trading time for money which is there's nothing wrong with that you know some examples of this is like any type of live betting , especially if it's , you know , like you're using you know whatever third party software or other outside software .

It's a little different if you're like building your own thing , I guess , because that you're probably developing some sort of scalable skills there or , you know , dependent on this score , whatever , like the . There's a lot of opportunity and there has been a lot of opportunity in the ecosystem .

You know the past couple of years , just as a lot of things are new and whatnot . But the reality is like those things are just really hard to grow in a way that you know you can continue to do for a long time and continue to branch out to other markets .

This very example , if you're you know your main strategy is I know all the rage these days is is a live arming .

If that's your main strategy , like there's a finite number of hours in the day , right , and so like that strategy very much requires you to be in front of a computer and you're creating your time for money and again , seems like you can make a lot of money off that . So I'm not knocking that or whatever .

It's just you know in doing that , for me at least , like I would feel like I'm not developing skills or anything that's going to really help me in the future . It's sort of just like a honestly like a job at that point where you're punching the clock and making money , which again nothing wrong with it .

But that's like for me I'm trying to make gambling something that I can make more from each year with less amount of work . So that's , that was sort of the impetus behind that article .

You know it's like there's there is some , there's some value , you know , to thinking about not just how much money you're making , but what . How much are you trading ? How much time are you trading ? Because I think we all trade time for money , but we'd like to trade less of it .

And when you're younger and you're starting out or you're first getting into gambling , trading time for money is is a huge win , because most people trade time and lose money . You know like it's it's trading time for money and gambling . You're already doing great . So it's like it's a great way to to see some , you know , see some shots .

Go in right and build up your bankroll . But I do stress to people like , start thinking long term , you know . You know by all means , live Arb and make a ton of money . I know people are making having like 40k months live arbing . That's great , that's awesome . Save that money and then maybe invest .

Invest some of it into learning some skills that are going to give you leverage right , like I think we both kind of think along those lines .

So , just to separate yourself and really have success , you need to do something well . Whether that's it can be , even that can be a variety of things . That can be , you know , data collection , it can be data modeling , it can be getting accounts .

I mean , there are people who make all their money just because they're really good networkers and you can constantly turn accounts . Yeah , that's not really specifically of interest . I mean , everybody needs to . At the end of the day , you need accounts to bet and win , so that's obviously important for everybody , but that's not necessarily like .

You know what my strong suits and what I want to spend my time on personally . But there's a zillion different things you can do there . But you need to have some skill . And yeah , I would just recommend anybody certainly always take the money when you can get it .

There's nothing that's going to make you more money than increasing your bankroll quickly , right For sure , so you can bet more on other things . So , absolutely take the money while you can get it .

But I echo what you said I think everybody should always be trying to develop , improve their new thing , because that's how you're really going to if it's something you want to sustain . If it's just like a side hustle , then you'll go find different side hustles . Take the money and take the easy money and run . I goes .

So if you bootstrap your bankroll through the stuff that doesn't scale , through trading time for money , and you learn how the sportsbooks work , you get comfortable with risking money or placing money on bets .

Those are things that aren't tangible skills , that help to just get a general level of comfort and understanding and meet people and you'll find you might find your skill later . But you know , take the money Like you said , like there's no nothing that makes you more money than just having a bigger bankroll so you can bet more money .

That's a huge skill and that's also why I say like it's helpful to be cheap as a better . If , like , you really do want to do this , try and be kind of cheap , because if you make a lot of withdrawals from your bankroll , your unit size starts going down and that does not help at all .

But when we talk about scaling like the second half , like OK , we've done that , we've done , we've traded our time for money , we have a bankroll , we're starting to get limited at DraftKings and we're getting limited at at you know whatever rec book . Now we're thinking , ok , well , I actually really like this , I want to scale . What's that stuff like ?

There are ways to get limited slower . You just released an episode that was very good on some of those ways , so I won't rehash all of that .

But I guess one one thing you certainly can do is if you are you know I'll give it exactly if your edge currently is , you know , subscribing to some discord and gives out a play and it's , it's good play , but 500 other people are betting it like that is not a way to have longevity .

You know , purely picking off off market numbers , that is not a way to have longevity .

If you are able to , you know , develop your own prices on things , on things that aren't that frequent , but they could be not that frequently or their , their liquid markets and more , more liquid markets , and you're able to bet into those and win , you know that's going to buy you more leeway If you're able to bet .

You know , one of the things that Some of the I feel like a lot of people on Twitter miss is a lot of people you know see high holds and like in certain markets and will , you know , say , bookmakers or scumbags or whatever . And sure , that's probably a case a lot of times . The reality is oftentimes like the most valuable opportunities .

Some of the most valuable opportunities I've ever had is betting into markets with extremely , yeah , high holds or higher holds , because you know , when a bookmaker puts a higher hold on a market , they're thinking that they're doing that because they think they need some sort of like more margin of safety on that market , and oftentimes it's it's it's like a false

sense of security . I mean , I think a lot of future , future pools are like this , you know . So for me it's because you're going to be able to Last much longer if you can bet into markets where their high holds , where that book is not the best price in market , like that .

That is something that is very good if you can win Betting into not not best best price in market , because if you're just thinking about , like from the sports perspective of what sorts of things they are looking at , you know that that that looks like you know the person who's always taking best price . It's not a desirable customer .

So , right , I think having some of those , those skills , is how you can Continue to expand and grow . I think the other thing I'll mention like for me that has been beneficial is if you actually build like a skill set and think about things you know , like betting , projects or endeavors , whatever more long term , you have more optionality in the future .

So some examples of this are you know , if you , if you had your own models or had ways to identify , for example , like correlation in sports betting , before pick-up maps came around , like the DFS pick-up maps , you have a huge leg up , whereas if you're dependent on again , like software or whatever , that's not something you know how to compute or figure out

Similar sort of thing with . It seems like more and more companies are releasing these peer-to-peer pick-up type games like draft teams , pick six and , I think , underdog and even prize picks . I think rolled something out in a couple of states recently . If you have framework and I played a lot of I call it DFS .

I guess I can't use that term anymore because now DFS has been hijacked by these pick-up sites , but the traditional DFS salary , cap DFS .

I play a lot of that , or I've played a lot of that in the past , and if you have framework to think about those games and betting and whatnot , and you just build processes and methods for analyzing and attacking those games , you're able to go after new games faster than the crowd , and that's where so much of the opportunity is .

I always think in my head it's better to be early than to be smart on anything gambling . So if you're able to be first to things , that's where some of the biggest wins for me has been .

But I want to go back to your own prices , actually . First question , because I think about this all the time . Let's say that you're originating a market and you're looking around and you see the best price at a book that you have , at whatever app book that you have liquidity at and you can bet it , you have second best price . You also have liquidity .

Are you actually in that moment , if you're very confident in the edge , skipping the first best price and going to the second best price ? Or are you only thinking about that more on sizing up , if the best price is at some book that you don't have access to , but you know that the trading team might check just to see how off market they are ?

Really I'm happy to take whatever the second best price . I don't think often I would forego that , whatever market best price , unless it's like a real outlier . I have reason to believe that book is specifically going to take offense to that action . For me it's really just like a bet size thing . I guess what I was getting at is .

I think that's helpful to the account still , because someone who's reliant on whatever on screen , by definition they're only really ever going to be betting the best market price . So they just will never have those opportunities to get bets into accounts that are not the best market price and those things bounce out .

So maybe on book A first bet I take it's the best price and I make another bet it's not . So it sort of bounces out can help a little bit . I've based in my experience .

But when you do find those situations where you can bet second best , third best , whatever it is price , knowing that you have an angle that works , that you have better data , whatever you have something figured out , it is really nice feeling because it takes a lot of the anxiety off of wondering if they're going to drop the hammer on you .

Like , of course , they still might If you have such a good spot . That's going to be obvious in a week why you won all this money . You still could get limited , but it is nice to find those spots .

And speaking of those spots , I think now the question I'm getting the most besides , how do I do my taxes , which I don't respond to those questions and inquiries . I'm not an accountant .

But the other question I get the most is how do you dip your toe into creating your own prices , to finding spots where you can bet the second best price because , like you said , if you're using the odd screens , you're by definition betting the best price , like 99% of the time .

Hopefully you have a couple of options to you in terms of books . But I would peruse the books and try and find , I guess , first finding the market I would go after . So I would try and find a market that ideally , is something that's offered with some sort of regular occurrence , and what I mean by that is it's frequently offered .

So a bad example of what I would not go after is I know the Super Bowl is coming up . I would not do this exercise for that and I'll explain why in a second . But I would look for something that is offered regularly and ideally I would look for something that is unique to one book or a few books . So I would look for a prop .

I should have clarified that . But I would definitely look for a small market , like if you're just starting out . I would look for as small as humanly possible . I would look for something where you're thinking who and like , who in that right mind would bet on this honestly ? And I know that's , you know , sort of counter to what I was saying .

But if we're just starting out , this is where I would start out . And so now you have a market only offered on one book , that's . You know there are prices on for every day and it's better if there's like a lot of prices .

So to give examples for this , like you know , like sports , where there's stuff happening every day or very frequently , like stuff like tennis , or you know there's golf basically every weekend of the year . You know stuff like that .

And then what I would do is , you know it's very dependent on what the prop is , but I would just , you know , try and think through how would I actually go about pricing this ? Like what do I think the book is looking at to price this ? What do I think I would do ? And so it's usually you know the process of pricing anything is .

You usually need some sort of collection of data . So if it's , you know , like a derivative market , like if you're looking at , I don't suggest , you know , going after a market like this , but an example if you're trying to price second half totals in the NFL .

You know you might need you know what the historical database of lines in the NFL , and then you know results and can do something along that .

If you're doing something like you know more not or not , as that is clear of a derivative , so more like I don't know something like more player prop base , you probably need some sort of you know player results database or something , or play by play data that you can roll up or whatever .

And so I think I think picking the market is the first step and then trying to come up with a price is a second step and it's going to probably be quite bad .

Yes , to start , and I think just starting in Excel or something you're comfortable with is even on , like legitimately , you can do it on pen and paper for like a specific example , and just think about the process and convert that to Excel , and where you can write that to code , as you can do , you grow , that's fine .

But I think coming up with a price and the reason it's good to pick a market that is offered frequently and it's good to pick a market where there's a lot of offerings is you're going to get more rapid feedback than if you , you know , do a super bowl of proper whatever .

So if you are coming up with , let's say , you're , you're trying to project how many double faults , for example , someone's going to have in tennis , like tennis , those you know there's going to be dozens of tennis players playing every day . You know you're going to be able to compare your prices to market and you're going to quickly figure out where I'm wrong .

If you're consistently whatever 50% higher than the market , you're probably missing something . If you are consistently low on a certain type of player , maybe there's certain surfaces or whatever that matter , like you're going to , you're going to quickly figure this out and you're going to be able to continue to tune it in and figure out what you're missing .

And so that would be , I guess , my overarching sort of how to or what I would suggest someone do .

I had just read Moneyball and started following like fan graphs and I was sure that I was going to be able to like price baseball money lines . Like I was just in Excel . I was like , obviously this is going to happen . Like I definitely didn't win , but I created something that spit a price out and that helped .

And I did get a lot of feedback and my feedback was like I probably can't win doing this , so I stopped , but and I continued to play poker . But I I think getting a price out also just shows you that you know it's not all magic . Like these books aren't just like magic . There's a set of steps they take to put prices out that you can take too .

So , yes , your first price might suck , but you're getting the reps in . But if I had maybe picked a better market , I might have found I could win and spit out a price that won the first time and that might have changed things drastically for me .

So going to pick a market first , I agree , is more important than trying to just go find a price , spend time thinking about the market . The question I have for you is should we be picking a market that we have some domain knowledge in ? Like know the sport pretty well ? Does that stop us from potentially massive pitfalls we might not be aware of ?

Or does it not really matter ?

I've certainly bet on sports that I know more and less about . I bet on sports and bet on a sport , maybe with watching very little of that sport , more math and data focus . Then you watch a sport and you see like whatever five pitfalls with what you assume you're doing . It doesn't make sense . I 100% think having domain knowledge is valuable .

The only thing I'd say on that , though , is the majority of sports people watch are the popular ones Football , basketball and those are also the most competitive ones . If I were to pick the perfect sport , it would be a sport you like and will watch that is less popular . I always think about it . It certainly helps .

For the people that are listening and love NBA , love NFL , is there something should we be maybe pushing them , like an NFL person , to go take a peek into , like Canadian football or a smaller market within an NFL game they specialize in ? I don't know . When betting opens up at halftime they start betting second half player props . You know what I'm saying .

Should we be niching down within the sport , going to a slightly different sport that might be more niche ? What do you think there ?

really small and , I think , really beatable for a lot of people . Like , if you go small enough , like I said , I don't suggest doing like going after markets that are very available across all books on odd screens . I mean that's a good way to know .

If it's not something you should go , it should very much not be on an odd screen , because if you can see that then so can everybody else , so can anybody who works at a sports book . Like that's not proprietary stuff . So that's what I was sort of getting at it the only one book offers it .

Because then if you're going after something that really only one book offers , which is again sort of like why I really , why I wrote the scaling article and why it's a lot of what I think about is like if you don't have a skill set to attack that , like that's just not going to be something that's attackable for you and like .

That's where the real opportunity is , because when you are , when only one book offers something , it's more , it's like one V one . At that point , yes , right , you're not . You're not competing into a market , not merely in the same way as you know if something's on 10 different books .

So I'm always looking for stuff like that , where this is the only book doing this . Now I only have to be like one person .

But the one I give kind of frequently because I also say , try and find something that only I , you know , this is something I try and tell people who are getting into bottom up , it's the prize picks fairway market and like I took the time to sit down and model out fairways and I had like a 50% edge , like it was insane and they were the only book that

offered it . It was me versus the guy that prize picks hired . That you know , like just it wasn't , it just wasn't a fair fight and what they would do is they would put it up , they would take it down , they would put it back up .

You know , one week it'd be down , one week it'd be back up and they just they're totally lost because they had no other book to go look at .

It was a true one V one and if you just out work or you know , just do it better than the other person , you only have to beat that one person , then there's nothing they can do in defense is they can't lean on an odd screen . They can't lean on another book to simply just price off when they're unsure .

So what they did with fairways is they just put it up and put it down and put it up , and then they for a while they took it down for like two months I feel like they must have thought they worked on it put it back up , got smoked , took it back down . Like is just like this hilarious back and forth .

So I mean , just on the that prize picks example , I always I think I tweeted about this before like I always think it's funny or not funny , but it doesn't make sense to me .

The people who are like going after , like spending their time and effort , all their time and effort , I guess I should say , like on the DFS apps , on the , the markets , that there is like widely available lines for like I'm not saying you can't beat like NBA and NFL and whatnot on prize picks and underdogs .

You obviously very much can , especially with so the the correlation stuff that's been greatly reduced . But like the real opportunity to me in those sites have been like they are offering lines on cricket and I think it's pronounced high lie and you know like legitimately are not offered anywhere and so they're like even more .

So you're really competing against like one guy with some probably like Excel model working at these companies trying to put up lines , and that's that sounds like what you were describing in for the fairway stuff . So that that's certainly one example

. I think the other , the first example that came to mind , is at points bet , which is obviously notorious as pretty severe , limiting , but so at least when they first came to the state I was in , when a legal state , they allowed me to bet for just a couple weeks , I think , and I think part of the reason . Well , it was quite short .

So if you bet , for example , on the Chiefs minus seven and they win by 40 , you get paid more than if they win by 10 . And there was some market up and they had really complicated stuff . They had stuff where it would be like win margin times total or something like that . It wasn't just linear .

And I remember there was a market up that was , I think it was win margin squared or something along those lines .

And so there was big opportunities betting on like bigger favorites . I think they kept , like they didn't allow you . If it was , I don't remember if it was like more than a 21 point favorite , which was sort of like a sign that if you're betting a set 21 , it's like really good because they sort of just virtually cut it . And so that was .

There's your question , like that's an example of a book that has a unique bet type that nobody , like they don't have anybody else to compare that pricing to , and so you could just make like huge errors in those instances . So I felt fine talking about that , because I don't even think that .

I don't even know if it's , I don't even know if points bets around anymore . I think they got acquired .

Like every sportsbook says and points book even said it but yeah , we get information from winning betters , we want winning , you know whatever . But yeah , I mean I don't .

It's almost like a points bet edge is worth , like not that's one where you can just share , because it's like might as well have people hop on and hop on and hit it for the one day they're allowed to bet there .

But yeah , I think that's a perfect example of what you need to be looking for and I think in it highlights another important concept in gambling , which is like the looking , the scouting concept .

If you talk to the the , you know the APs that go casino to casino and they're looking for maybe like slots or certain table games with certain conditions , or they don't even know what they're looking for , but they're just trying to find something new or find a good blackjack game with good rules .

There's a lot of time spent scouting for the right conditions for the right game and there's this , there's this incentive for us to feel like , oh , we need to immediately get in and start doing something , and sure like if it helps to have the bias towards action , but like there's so much value poking and prodding around a sportsbook just seeing where you should

devote your time , because you know , if you point 10 hours of time at the wrong thing , it's it's . You could make $0 . If you point 10 hours of time at the right thing , it could be like a six figure opportunity .

So the the payouts are so massive when it comes to the differences and opportunities that like it's worth the time to go look for the right spots Absolutely .

If I'm going to try and beat something , I want to make sure that , like I say , either either that opportunity in of itself is going to be worth it or it is going to put me in a better position to do other things in the future . So that's that's what I'm really looking to do when I , when I go after , when , after things .

And if that project didn't provide any actual monetary value to whatever the whatever the , whatever bets I was , you know , whatever the model was outputting , if it like didn't win or whatever , if the opportunity got shut down by the site and they didn't want to take bets , that's okay because I learned Python .

So , like , when you're evaluating an opportunity , like SP said there , it's like the second half is am I getting something from this that will compound for me later ? And if that's , you know some type of technical skill ?

I think the answer is , like always yes , and something that I hear a lot from I think it's from Colin Jones , who's a Blackjack Blackjack player . It's like have multiple bottom lines , like one of your bottom lines could be monetary but the other could be skills . Knowledge could even be having fun or freeing up time , like to think about it holistically .

I think helps with what you were talking about with scaling and making it fit appropriately , like into your life and thinking about it in terms of , like a time trading for money or leveraging skills , you know XYZ . So I think that second thing is evaluating are you getting something out of it that's not monetary ?

If yes , then it's probably there's massive floor to the project . You know you're going to win one way or another , and those are the . You know you're basically free rolling the situation .

That's again , that's why I , that's why I try and focus my time on projects where I'm going to even if they're not , you know , successful in gambling like it's going to set me up for something else , whether it's a different gambling project .

You know , maybe I have a chunk of code or lines of code or workbook or whatever like that can help me on a different project . Or , you know , I just learned something and can do something faster , you know , in the future .

So this is something you kind of pose to the community . You wrote about in a long form article . Do you want to just briefly walk the audience through the article and then what your thoughts were around the conclusions and why you wanted to pose that question to the community ?

so . I think the general thesis of the post was that there's really three ways to win in gambling . I've said it a couple of times that there's many , many ways , but at its core it boils down to three things .

You can either have better information than the market or your opponent , so that could be broad , so that could be you have a proprietary data source , or it could be something as simple as you know someone , the start quarterback's not playing , so any type of better information is one way to win . The second way is the interpretation of that .

Your interpretation of that information can be better . So that could be , again , different , that can take different forms . That could be you are a better , better modeler than the guys working at prize picks on their fairway model , or that could be . You know , again , it doesn't have to be all modeling based or technical based .

It could be this quarterback's out , you know , you think the market's overreacting to it or whatever . I guess I'm saying this to say like these things don't have to be technical . And then the third way I guess , which I didn't really cover in the article , is you can be more effective at using those first two data and the interpretation to actually get down .

And so this is just . This is more like the person who really is the good network or gets accounts , that's their sort of thing . So they don't necessarily , you know , focus on the first two , but they're really good at getting money down and working with other people .

So my area of interest and focus really is on their first two , and what sort of I was trying to hit on on the article is I think there's like a People tend to idolize people who they view as really smart in terms of modeling and data science and all the buzzwords these days about I feel like two thirds of people giving out picks these days have some sort

of artificial intelligence thing in their bio or something , and so people really gravitate towards that . But at least in my experience and in my opinion , that is much less valuable than having data that is unique and valuable and interesting . So the reality for me is that there is pretty quickly decreasing marginal returns on modeling .

I mean to give , so I'm in , but my day job is I'm an actuary , so I'm definitely not like the . I'm certainly not like a 10 out of 10 modeler , but it's , it's a big part of you know what I do , and so I definitely probably biased a little bit by that fact that I'm not .

Like if I was the world's best modeler I probably would be saying like , oh no , this is the most important .

So I think I could talk a little bit more about , like you know , some of the ways to define that data . Yeah , I'm not giving a , a whatever .

Like there's so much data that you're probably not even thinking is like accessible or gettable . But there's people out there Like what's the stop you have of sending someone to each PGA tour tournament taking a moisture reading from the ground ?

You know , like there's , there's data , there's like unlimited , honestly , upside of good data and I say this because I have this example it's like if you knew the game was literally rigged and that this team was going to win , then I don't care . You know how much a model , you know how good a modeler are . You just going to smash the .

You smash full bank roll in . This is just assuming this is actually like reliable information and you'd win . And that's the upside of a data advantage . And of course I'm not advocating for any fixing of games , but like as a thought exercise , right , the upside to good data is so much higher than the upside , like you said , like the cap .

The ceiling of good modeling is nowhere near the ceiling of good data or unique data .

My thinking now is maybe we just briefly talk about , like someone's listening to this they're like , okay , great , I didn't want to become a 10 to 10 modeler anyway , I want to go find good data . You know where do they start , what are they looking for ?

Most of the data edges for me has come from being able to scrape or compile data sources that are not necessarily easy to compile . I mean , there's millions of people out there who can scrape stuff , so I'm not unique in that . But to like a pen To apply , to apply it to markets .

So if you're technically , if you're listening to this and you're technically good , I would say you know like a lot of websites make it very hard to scrape or they try to and the harder it is generally like that's more like of a green light that tells me I need to do this because there's probably a lot of people quit along the way .

So what I mean by this is like there are so many markets these days on stuff .

I'll just give some examples Like first basket or you know you could do different teams first play tendency , or there's a ton of stuff and opportunity in some of the same game Parley type markets where you know those are priced very , I guess , uniformly , based on everything I've seen .

So they're not really taking into consideration that the wide receiver won and the quarterback have this interaction on team A and they had this interaction on team B , yeah , and like how those that's just one example , but like how teams specifically like interact or correlate or tendencies , and so , if you can , my thought here is trying to collect data that's not

in , like it certainly shouldn't be something that's available in a box score . So it could be stuff .

Like you know , these things have probably mostly tried up , not entirely depending on the market , but , like you know , some teams do different things on coin tosses or some teams , do you know , like different , different things like that , and so , like collecting databases like that it could be very manual , right ? Sure ?

But you said that's where I would look for like tendency team , tendency type stuff and trying to find patterns there that are not priced where price list of there is a pattern there or something like that .

It might be kind of smashing together a couple of different data sets that are a little easier , but if you , if you combine them with the appropriate question in mind , you can get actually a like kind of a thoughtful answer .

That might not be easily apparent Because first of all , people have to take the time to like properly create a database using a few different sources , which is already like a little bit more of a pain than I , at least I would have . I always originally think . And then it's like you want to .

If you have the right question in mind , usually you can combine them in a way where you can get to an answer easier with like pretty basic modeling . But you have to kind of set yourself up for success by knowing what you're trying to figure out .

I think the first basket thing is really interesting because you're thinking , yeah , first play tendencies , like is this a team that likes to shoot threes on the first play ?

It comes to mind as something that might be might be important as to like maybe a team that likes to take a two pointer and go for like a fast two pointer and the pain or something , and who's their center and how do they always put this guy out on the tip or whatever . But there's data there that you can maybe have to even hand map .

But if you're solving for that problem , eventually you're going to come up with a data set that you can use without scraping to try and figure that out . I think .

I would guess it's not , and so if you have access to that , like that seems like a good , like I feel very strong that you could make money off of that and having access to that might just be like you have whatever the NBA , like all 60 or I don't forget the NFL version of it , but it's like you just go back and literally start of every game .

Kind of the more annoying something is to do , the more of a green light it is , I agree . I agree with that wholeheartedly .

But if it's , if , like , you're searching for it online , like yeah , let's say it's on a like some style or something you wanted to put into your you know model or pricing , or it's just it's in itself , if you can't find it anywhere , like that is likely valuable and so , whether it's collected manually or scraped like it's hard to scrape or you have to just

a lot of work to collect it manually , like not saying no one else is doing it , but probably not many people are doing it and they're , therefore , it's probably like valuable In that sense .

I think that it was incredibly good at first , but I wanted to hear your thoughts on like what you think first the prospects are of this type of game . Like do you see this catching on or do you see this kind of fizzling ?

I actually had discussed this maybe a couple of years ago with a couple it was actually a lot of different podcasts and this thought has been the thought of this type of game has been floated for like years within the traditional DFS community as , like this is what's you know , sports betting or DFS is going to turn to .

The problem with the game as it is currently constructed is it's incredibly simple , and so when you are playing , when you have a peer to peer game , you like I'm speaking from the vantage point of someone who wants to win money at this .

So , like I'll comment on , like what I actually think of the viability , because the viability to me is really going to depend on how much , like recreational players like to play . That's pure day peer to peer game . That's the reason , you know , traditional DFS has been so successful for so long .

If you just have a bunch of sharp people in there , those ecosystems don't they quickly enter like a death spiral ? Yeah , so the game is just to me it's a little , it's too simple in that the strategy is it could be explained in . You know about two . Like I could explain it in about two sentences to someone . You know it's not really rocket science .

The idea is to get as many picks right and do it in a way that other people are not getting their picks right , and so I think most people who , who can gamble successfully in other arenas , can figure out sort of what , what the strategy is . I've tweeted a little bit about it .

I mean , it's no secret , like people generally love taking , like recreational people generally love taking overs . That's not a secret , but more and more so it's .

It's getting to this little bit more of an equilibrium , it seems , where you know there's enough sharp people in there taking the unders such that that's like you know weighing , like like evening it out , such that there's not not nearly as much opportunity as there was in the beginning .

I haven't done like a full you know sim or anything on it to see how profitable it is .

I still play , I still think it's it's good and beatable , but with the you know the break structure and just the simplicity of the game , I don't think it takes , it will take , very long for it to be become a game that's not really beatable or not super beatable or worth worth time . So I worry about it from that perspective .

So I mean the variability to me makes the game a little bit more interesting and it is simple , but it makes it more interesting than just straight up sports betting . But I worry that that is a bug and not a feature for most people and so that also does not have to be super optimistic on the game in general .

Like you said , you could go in and you click through the CSVs that show like the contest picks , which is something that everybody who's playing picks six , and I assume I don't know if underdog does this or not , but at least for pick six you can see everybody in your contest entries and certainly something that you need to be doing if you're playing these and

you'd be like , you'd be like the only person A taking unders and B , you're the only person correlating and it's like this is absolutely insane . You're just free rolling like a 40X payout on , like a pick six or like a 50X or whatever pick count you're doing . But I always thought in the beginning , of course , just like win whatever .

But I was thinking of the flip side . It's like the flip side of that is , when the rec wins or the overs hit and they all had the same like six picks , they're getting a pick six that pays like 9X .

And there's that one Twitter account that you like responded to and this Twitter account's like a comedy the DK guy and he's like , anytime like a big pick six comes out , he'll post it and then he'll never post like the ones that aren't good , and then he never talks about the rake .

And it's just like this clear , clear , like push almost now feels to me like this guy and Dracking is trying to like put lipstick on a pig like from the very get go , because the recs all hit together , because they've got the same stuff , so they're probably just so frustrated . They lose , lose , lose , lose , lose .

They finally win and they get nothing Like I can't imagine . There's just like a more frustrating feeling for a recreational gambler . So I'm kind of with you on the fact that the rec experience in these games , combined with the rake that's , let's face it , too high . They're doing nothing like to earn that rake .

It just feels like the rec experience is just like the experience is just losing , that's it .

actually asking for some feedback on my thoughts on some of it .

So I know that it's early , so I'm a little hesitant to be like overly critical of the game because I mean , like DFS , like traditional DFS had a lot of I mean a lot of the same phenomenon happens in DFS , like big part of daily fantasy for sports , where duplication is a big issue , is being able to submit unique lineups and not do what the most common

thing to do is , I guess , because it's very frequent there might be 100K prize up top and when I'm 30 , people tie and now you're not getting paid a whole lot . you want to win when you're the only one who wins . So , this is . It's really the same thing , and so it . I'm not saying it's like it's definitely over , it's just it actually becomes a little .

I think we're still in an interesting period , because I think , right now at least , we're at a point where you have the wrecks pouring in like all over is , and then you know , a complicating part of the game is they have been pricing really terribly . I don't know if you've been playing lately .

They've been pricing really really terribly like far too high on almost all their props lately . And so that's getting you know . In the beginning it was good like taking a 50 50 under .

Now because it's like weighted to like maybe I don't know they've been really bad , like maybe 65% of time going under , like just in general , because they're pricing all the lines way too high as compared to the market . It's just getting all these unders .

And so now you have sort of like a barbell where you have a ton of people taking all overs , you have a ton of people taking all unders and I mean I have not thought about the game a ton Right In terms of strategy and like not done a ton of work but me

neither . You know , the other day , I guess you know , on NFL weekend it was . It was a sort of an interesting one .

The odds were extremely high and I think it was a large part due to the fact that , like , basically , I think if you're playing for the game to end zero , zero or I know I can't end that- but , like a very low scoring game or a very high scoring game . I find it hard to believe there's like a ton of opportunity and either of those right now .

If I'm giving a little bit of what I would think the strategy is . I think you want to think of different game scripts .

Selling that because right now , based on what I'm seeing , it's just going to be very crowded at those Like if everybody goes under or everybody goes over , and I know it's not everybody , but if lots of people go over there , lots of people go under , like it's getting to the point where the benefit of being unique , I think , is potentially worth more than than

some of the correlation aspects . So that's not , again , not actual analysis or anything , but I'm just based on what I'm seeing . That would be . That's sort of like what I'm thinking about the game these days .

And after , like I would always check the entries like manually . I didn't put it into any databases or run any code on it , but I would just check through it . And then I saw people are doing like more correlated unders .

And then I basically came to that question that you're posing , which is , are we trying to find a mixed strategy or does that outweigh the benefit of correlation , or does it outweigh the benefit of these lines being so bad that , like the unders are going to have a 65% hit rate ?

And I thought to myself I was like I don't really feel like I'm going to be able to solve this in a fast enough time to make it worth it for me right now , and part of it had to do with that was like a little skeptical of the longevity of the product .

But I certainly think like that's a good situation if you live in a state where you're only on like the DFS apps and you want to do something that's worthwhile , like maybe go try and be one of the first people who really gets good at the game theory of these pick six games . There's a huge opportunity there . These games run every day .

It's peer versus peer , so you're as good as the difference between you and the average player in the field which could be stretched . You could put a lot of time in when they're not . So I think it's a good opportunity to do some really thoughtful analysis .

Certainly like I haven't really found the time to do it , but would you say like this is kind of a good spot for somebody who's in one of those states that might not be welcome at prize picks and underdog much anymore and wanted to find something to do 100% .

So , like in traditional DFS , you know , in the very early stages it was enough to just play honestly . It was enough to play people who weren't injured or sick , like that , is enough that first . And then you know it was people just played the quote unquote best plays , like the people who were projected for the best , most points .

That was good enough to win tournaments . And then it moved to oh actually these people are correlated , we should put them in our lineup together . And then that was good enough to win . And now DFS is at a point where you know it's all of that .

But then it's also what is everybody else in the field doing and how do I best put myself in a position to leverage that , like so , when you know what outcome is , when these certain outcomes happen , I will be , you know , at the top alone by myself . You know I can really stand apart . And so it's going through .

It's basically yeah , it's going through those same cycles . It's really the same game .

The thing that makes it simpler and , to me , less just , generally less interesting , is the fact that it's binary in a way that DFS is so like it would be potentially more interesting if , instead of like you win six legs , you lose six , whatever how many six legs , like the placement position is on .

It goes to almost like a points bet type of thing where if you win your prop by 30 yards that's better than winning your prop by 20 yards , so like that , and because you know , as it's currently constructed , there's just got like a lot of duplication and overlap and it just makes it simpler when it's binary like that .

So I certainly think there's opportunities in the in like that game theory aspect of the game .

Yeah , but I think it's yeah , it's like I guess going back to collecting like data sets and with you know if you collected , I mean , I think my basically most of the edge for me , I guess in the beginning it was , like you said , mostly just people not knowing what the hell was going on , but also like just collecting data on what people are doing I think

was valuable , like because even it's one thing to know people are going mostly overs but if you know what team or players or stat type , because there's there's very much a bias . It's very funny Actually .

There's certain stepped stat types that almost are exclusively taken over and there's certain stat types that all the the sharp people take the unders , and it's , it's very like that .

No matter how , like even if the probability is like whatever 70% , you just want to be on the opposite , because if that 90 , if the play that the 99% are in wins , like no matter what you're on , like , even if you win your other pick , like it's worthless , like you you want no money , right , you really win when other people are wrong .

And so if you know , you know what teams , what types of players , what types of stat types people want to bet on or not bet on , like that , that is valuable . So I definitely think there's opportunities still . I hope it . You know they can . They can continue to improve and tweak the game . Yeah .

They have , you know , max entry dollar amounts and is there point in time where maybe they just kind of inch those down for certain people Maybe , but then it doesn't necessarily even hurt the ecosystem that much . It could even be good .

So there is longevity and getting good at that , which is nice , which is the same as traditional DFS , and I always was somewhat jealous of the people who are really good at traditional DFS because they're never thinking about cover and getting limited and whatever . They're just doing the best they can to win the game .

And Sam's poker , you know it's , it's , it's , it's definitely , I'm sure , like , as you're someone who comes more from the background , it's kind of nice right To just be in there battling other people and trying to be the best , instead of worrying about getting getting the boot .

Games like that is , you know , like a true cash cow , because you are not constantly churning accounts , you're not constantly getting booted and have to figure that Like , if you can . I mean that's the equivalent of being very , very liquid markets .

In peer-to-peer , wrecks are good for you , whereas in major markets wrecks don't really make a difference one way or the other .

They're not giving you back an extra point or two because a bunch of recreationals are making a stupid bet .

You have so many people who are just doing it for fun there's nothing wrong with that , it's entertainment but their entertainment cost is going to the people who are trying to win .

I've done that just audio , not video , because we're giving SP some secrecy , an aura of mystery . I've just been taking notes and really honed in on this conversation . I had a great time . I learned a lot . I mean I can go back over my notes from this , from just recording this , and I have five or six ideas or thoughts that I might want to test out .

So thank you so much for coming on and giving us great , thoughtful content .

You can sign up for his content to be delivered via email or you can bookmark the website and definitely go check out the articles he's written and encourage him to keep throwing those out , because they're truly some of the most valuable long form stuff in the space .

So yeah , everybody , thanks for listening , thanks for caring what we have to say , because I have a lot of fun recording these and I will see you all on the next episode .