How to Not Lose on Prediction Markets & Kalshi Outage | Ep 126

So we just kind of sat there and we were just taking all the Scotty, all the Scotty, and you know, you have to make a decision. At one point, we were, yeah, probably net short, even more than 150. What's up everybody? GP and SP, flying, solo or duo or whatever. Disturbing news, it seems like people are actually listening to the podcast. This is horrible. I'm still pretending people don't listen to it, but yeah, appreciate everybody's comments and whatnot, but it's disturbing to find out people actually listen to this, isn't it? Well, you know, in the early days, it was a true struggle to get questions, and now our entire episode is effectively going to be questions. We had a topic today. We were going to talk about exchanges and market making and all these, you know, how we're thinking about these things. We got so many questions and so many about exchanges and taking versus making and all these different aspects of exchanges that we are planning on talking about, that we sort of reorganized it. So the first half, we're our first part here. We're going to, we're going to talk about exchanges and all the questions we got related to that. Then we'll hit the news and there's a lot of news you need to catch up on, and then we'll hit the rest of the questions. But yeah, I guess that's the downside and the upside of having more listeners is we are getting a lot more really good questions that we want to discuss. So yeah, I appreciate everybody sending those on. Yeah, this was a good, I've became very excited for this episode as the questions started rolling in because I mean, they're really good. Yeah, let's kick it off. So the first, like SP said, the first block is going to be exchange-focused questions. So we started this one from actually a friend of the show. We have to remember to refer to Ryan as a friend of the show. And he says, am I wrong in thinking exchanges are terrible for many listeners of this podcast. Small winners who can beat DFS 2.0 sites and rec books, but won't be able to compete first. The top 0.1% of betters who will now be their competition. I think this is a great question. I'll kick it to you first because of it. Yeah, so when I saw this question, I've actually talked with a couple other listeners or people I just chat with on Twitter who I think are sharp and have wanted other things who share a similar sentiment to this. So I almost feel like at least myself and maybe this podcast, we've done a little bit of a disservice to the scaring people off of the exchanges. I worry we've done that to some extent because I think there's sort of the way I think about this is we talk about exchanges a lot in this podcast and I think a lot of the time when we're talking about it, we are future casting like the end point of exchanges. But I am talking about it in that manner, but not necessarily. That's different than what the exchange ecosystem is today. So no, I don't think this is true today. I think people who've won on really anywhere can win today on the exchanges. Is that going to be true like five years from now? No, and we'll talk we have some questions and whatnot geared towards the future. But like today, I think it's just it's all it's new to like really everyone for the most part. I know like exchanges have been around, but like the most of the participants, it's new for them. So I think there's a lot of opportunities as we go through. We'll talk about like some of how how these people can can attack the exchanges like if you're coming at it like new or from DFS 2.0 or recreational bedding or top down or like I think we have questions on all of this stuff. But I definitely don't think people should just be like, oh, I'm just going to be facing the top 0.1% of betters that I don't want to dive in. I think that's a bad attitude to have. Yeah, I agree. I also think like the way we talk about or maybe the way I talk about it a little bit has like the connotation of you have to have some, you have to be automated, you have to use the API, you know, all of this stuff, which like isn't true and there's very successful. There's people who make, you know, more money than us not using the API, you know, uh, uh, Float comes to mind as somebody who has said he doesn't, you know, put orders in automatically and he's been very, very successful on on the exchanges. So, you know, I think the difference is I think you have to rethink a little bit of how you bet because I do think that there, you need to get into some passive, um, like posting liquidity habits. So, well, I don't think like, you can't beat it. I just think like definitely rewire it. So actually, I was just clicking around on, uh, call, call sheet on some stuff that's on, um, on Robinhood, but I think it's probably not super, uh, sharp. So there's, if you think of, if you look, so what, like good Robinhood app, this is an important piece of advice if you're trading on call sheet and see what's on there. So something that is on there usually for a bunch of stuff is futures. And there's a good amount of liquidity, but the spreads are still pretty wide and I have to imagine that if you were like to put passive orders on futures that you like and you can kind of, you know, figure out how to, how to price a future or how to find a spot where you might find value on a future probably easier than than a side. Um, you could get a lot of, you could get a lot of order flow there and you don't have to hold it necessarily even the whole season, you know, so you could start trading in and out of the futures market. Um, as one example where you will still get Robinhood flow, but you're not going to be up against the absolute best of the best. So I think it's just like figuring out ways to, um, figuring out ways to adapt a little bit and just realize like it's a new game, but they're still, you know, how do you capture some of that recreational flow and being a spot where you're capturing it, where you're not competing with like the best of the best. I think is a good first step and, you know, who knows? I mean, I know other people who've, you know, there's kind of these mentioned markets or whatever that also could be good. Like you, you know, they seem like a nightmare to market make. So that's probably a good sign of some, some spot where you could actually take the liquidity. So I think it's just like, there's a lot of opportunity, but it will require maybe like a overhaul in your strategy. Yeah, and I think I just want to make it like digestible for people if they are intimidated. Like, you can go on to an exchange and like, let's say you, you, you, you, you, you know, Ryan had mentioned beating recoups. Let's say you have something like, you're, you think you, you have an edge on some market, assuming one of these exchanges have it, you know, like profit has a lot of markets. No big has a decent amount of markets. Obviously, cal, she has fewer, but like assuming they have that market, like just, if a, you know, assume, assume fandal has minus 110 both sides, just, you know, make a, make a line for whatever, like small, small change to you, but like for minus 104 or something. Like that can't be that bad, right? Like it, it's, it might be bad, but it can't be that bad. And right, if you start out small, like you will, you will figure it out and you will get more comfortable on like, what I should be offering, how much I should be offering, like all of that stuff. I think people just get, like, scared at the idea and like, there's like an emotional even reaction to like, I'm getting, you know, like taken if I like post a price and someone takes it and I lose that like feels worse than just like losing to nameless faceless draft kings or something. But I think there's like a really easy on ramp for this to just like start playing with it. Like if you're, if you're making the markets, like you have full control over like what, like you will not get filled at a price worse than you offer, right? So like even if you just offer plus 110, and you're given the same line as fandal, like you probably won't get filled that often, you know, depending on what you're doing. And but you will like figure it out and you could slowly bring that down as you have confidence when what not. So I do, I think there's like a really easy on ramp for this where you're not like in the exchange and immediately you've lost your entire bankroll and you don't know what happened. Like I think it's very easy to like dabble in the in this. For sure. And then it's scale. I mean, scale is pretty nicely, which is a good reason to give it a shot, you know, you know, get in there. Makes it up. So I think, uh, yeah, I think think it like it like being passive, posting liquidity, you can get some really good prices. You might not always get your fill, but if you think that you want like you think this is a really good price and you post it, the worst thing that happens is you don't get filled. And if you find a price, you're like, well, look, I'd be really happy to get this price. I don't think I'll get it that much, but I'll be really happy to get it. Sometimes you get those. Sometimes you just get those and it's a good way to test out maybe some bottom-up stuff because you're like, look, I know this is my model is not great, but, you know, I would, I would take it at this price if, because it's a huge, you know, a big edge for my model. So I'll just toss it out there, you know, and you might get it. And then yeah, yes, be said, you just kind of work from there. Start there and then kind of spin it up. And you'll see more markets coming to the exchanges that aren't the sport-specific ones. Like I think, I guess we'll talk about this in the news and whatnot, but like I think Call of Shia and Polymarket, you'll see more and more markets that you would have seen on TFS, 2.0, some Esports, you know, some props, like all of this stuff. And yeah, so that could be quite, quite good again. So should we go to a bunch of shoes? Yeah, let's escalate this real quickly from going to the placing, like, you know, small orders and getting comfortable to a bunch of shoes question, which was, can you please talk about your tweet, read market-making, being short, 150K Scotty shares in terms of pros, cons, pitfalls of market-making. I thought that whole exchange could be worthwhile to expand upon. So I put this as a second question because I think this was like a broad question to talk about how we're both sort of approaching the exchanges right a little bit. And then we are getting into some like, we'll have some like adverse selection questions that I think we'll sort of touch on probably in this first answer. But I thought this is a good way to to teed up. So what do you want to say about, you know, being, being short 150K Scotty? Well, I think, I think, like I said, no, no one's an atheist from their short Scotty Shuffler. I stand by that comment. It's quite scary. Yeah, I mean, this, this was a tweet exchange. I don't know if like the questions about the whole exchange and like people posting like super off market numbers. I don't know if that's really worth getting into. I think that topics been beat. Okay. So let's skip that. Let's talk about just being short Scotty. Yeah, this was in reference to the Torch Championship. When we were, you know, Torch Championship was on Robin Hood. We had a price on Scotty that we posted just like we talked about the time before that we thought was really good. And we just kept getting filled. But like we were like, this is a great price. But the whole Robin Hood, the whole world just wanted to bet Scotty. And even if we dropped at a level, we still just kept getting filled. And so we just kind of sat there and we were just taking all this Scotty, all this Scotty. And you know, you have to make a decision. At one point, we were, yeah, probably net short even more than 150. And you know, that's a lot. And that's scary because it's Scotty Sheffler. And he's, we still think he's most likely person to win the tournament out of anyone in the field. But in that turn, in that instance, like we basically were like, I was pretty sure that our price was going to get jumped eventually. And we were probably going to be able to get short some other guys because we still held like the top level on some other popular players that I felt like would, you know, get some action. So we kind of just sat there and then yeah, a couple people eventually jumped us on Scotty, maybe a couple cents. And we didn't really chase it up. And we just kind of took that good price and worked with it before the event. I mean, we went into the tournament still pretty short Scotty. And, you know, during the, during the tournament, of course, like we were not shorting really anymore Scotty. So as the tournament progressed, you know, it was part of the strategy was, you know, taking that good. What we thought was a great fill was like, we'll try and make sure we're working that position down a little bit as the event goes on when we can at good prices. And, yeah, I think that's kind of market making. You know, I think that's in a nutshell, it's like sometimes you'll just get a huge position on a price that you love and it's a good price. And part of being able to take a lot of volume there is to also have a strategy of like working out of that efficiently or like hedging out that efficiently. So then you can be more aggressive in a spot when you're just getting piled into. And there's no signs of stopping. If you're like, well, I'll never be able to get out of this. You'll have to pull the plug early. Whereas, you know, we felt pretty comfortable that, yeah, we might not be able to unwind the whole thing, but we were gonna, we knew kind of what to do to keep, to make that, to keep taking volume there. And to have that volume be valuable like AKA, we're not hedging out like a terrible price to make it minus EV if we take additional fills. So, yeah, I think I think that was that's kind of it in a nutshell. And it's very scary to be short. Scott Shuffle is the other the other thing. So you're telling me you dealt a hedge to perfection and got, you know, got to zero. The delta is that's coming up in the in the news. I have, yeah, thoughts on that. All right. We'll get to that. But I think that that's a good way to tee up FLEPS question, which is the next one, which was how do you, how do you think about adverse selection with Kalshi? Do you think when you make an order, the taker will get at least x price, which is better elsewhere in the market? So it's not a big concern. So I think just, there's me jumping in now. I think what he's saying is like, let's say markets minus 110 both sides, you're offering minus 108. So that's like an incentive for people to come in and take your price. So it's like, maybe not necessarily an adverse selection play. It's just that you're offering a slightly better price than market. It's how I interpreted that part. And then the second part of the question, or do you think Kalshi trading or other market makers have the rebates so that they can pay less price and whatever you are posted, you need to be okay with getting filled all at once. So if you want to talk about this in the context of, you know, the Scotty Sheffler situation, that might be like a good or bad example, because when you were telling that story, I was curious, like, I'm guessing a lot of that, which is guessing is like, you don't have a ton of concern that that's like sharp money. Maybe you do. I felt recreational. Okay. So like, maybe that's that distinction is worth talking about. Like, because I think that to me, that's what this question overarching is getting at is like trying to your best to identify your counterparty when someone is taking your order and how do you how do you deal with that and the the adverse selection aspects of like making a market that aren't present. Obviously, when you take. Yeah. Well, yeah, I think call she's interested in flop. I think nailed it because I always, we calculate like what the taker would be seeing if they were just paying the regular taker fee. And I also, I feel like there is, I mean, I'm just going to make a guess here. So this is just a guess. I have no knowledge, but I do feel like the biggest live market maker, if it's sig, can just take with at least a big rebate. To potentially manage the book live and like, wipe size when they need to move or whatever it is, but I think that there's potentially some sharp taking that goes on live for sure. And then there's some sharp taking that goes on pregame from people who are, you know, people who just have have a good opinion. So we actually had something the other day where a prior guess to the podcast Maple had messaged me after we had lost like this, we had posted, you know, there's, we had posted, let's say, kind of, kind of a prop. And he knew we were, we were posting generally in this market. And we lost. And it was a big, we got a big fill. And I was like, well, that's, we're doomed. And then after when he goes, sorry, if that was you on XYZ. So that's one way to identify your counterpart is to have them text you. But the, the, the, the, the, the, just if it is like, I agree with fluff. It's like, I really don't think I'm calling you specifically that much taking is super sharp because of how big the takeer fee is and how you're almost always going to have better options as a sharp better. But I don't think that you can count out somebody who sharp, who doesn't bet with a lot of partners or whatever. And maybe, you know, the best price is on Fandall, but maybe the second best price is you on call she after the takeer fee. And if they can bet 20 grand on call she and, you know, two grand on their 10 percent Fandall, they might just bet, bet you a call for you at a better price. So I think like part of evaluating it is you only get so many big fills. I mean, fluff knows this, you know, so it's not too hard to like look at your big fills and try and figure it out. Oh, was this, you know, was sent in somebody released this? Was this ETR? Was this unabated? Was this an ARB to somebody? And we didn't realize it or or whatever it is and try and figure it out. But yeah, you're always like a very key part of figuring out like or a key part making money passively, like as a maker, is to like dial your size you show so that like the small recreational volume doesn't get completely wiped out by the couple of times like someone sharp comes and takes the max size. So it is very important to understand like, is there really room there for somebody sharp to come take it or not? Yeah, it's a great question. And, you know, certainly like we'll still get adversely selected from time to time. And, you know, it's just about about learning from it and dialing your size. And I think, you know, adjusting your size is is an important defense there. And yeah. Yeah, the way I was sort of thinking about it was when you, there's sort of like three camps broadly. There's like what I would call like the price shoppers that that like could be taking your order. There's recreational flow. And then there's like people with opinions. And it's pretty easy to have your counterparty be like the price shoppers. I think there's a pretty straightforward way, you know, to do that offer a price that's appealing to those people. And if you think what you're doing is is like better than what those people are doing, then you can in a very straightforward compete against those people. It's when you are like in the just to use again flat examples, like let's say market everywhere minus 110 both sides, you're offering minus 108. If you get filled there, that's not a price shopper, right? That's either someone with an opinion or it's a recreational better, right? And that's the hard part, right? Is when you're in that game of like, because it's like it could be those are two opposite sides of like who you most want and who you least want. And so I think as soon as you get into like that game where you're not competing, or you're not like trading versus top downers and all these people, like it becomes much more challenging to evaluate. Am I getting adverse selection? Because like again, if you're just like offering something that's less EV to the market, like I wouldn't even really consider that adverse selection, right? Like you know what's going to happen, right? It's when you you sort of have a price like you're effectively just asking like acting as like fandal or something. And then you are going to get adversely selected. So yeah, certainly the thing that I do like anytime I take a big fill, since you don't get a ton of big fill, this is like I'll go check. I'll just go look around what other books offer what. And if it's like every book is offering a worse price, and you know, if for example, we had somebody come take a big fill on somebody in golf and they were at a worse price at like every other book. And I actually thought that we were quoting too passively on this player. Like I thought it was like, yeah, I thought we were there was a really bad price. And that situation I was like, yeah, that's totally fun. But we took a big fill in like we took a big fill and we took maple's fill. And when I saw that come in, I was like, well, that's terrible. Like we're so screwed. Like and it was because it wasn't something we make the number. And it was only better the only better price was at Fandall. And I was like, yeah, we're doomed on that. Like that was certainly like we lost somebody there. So I think it's like it has every, you know, there's a lot of parts to it. But yeah, I mean, you know, at the end of the day, if you if you can hedge it out at a better price at another book, like it can't be that bad. So if the prices are better elsewhere and you actually have access to them, like, well, the worst thing that happens is maybe you better bet it out. Yeah, yeah, exactly. And that's why, you know, going back to the first point, like I think there's relatively easy on ramps for for these where you're not just like getting killed right off the bat. Right. I think that that leads into the next question pretty well, also from Bunchew. Playing off the adverse selection question, if you're posting prices, how do you think about defending your spot in the queue? If you keep getting action taken on one side against you. And so maybe I'll talk about like how I how I think about this. And I think you were alluding to some of it there. I think the first and foremost thing is like your confidence in your actual edge, what you're betting. So like if you are like a top down better or you know, just trying to market make like major markets or something, like you you should be like much in my opinion, like much less confident in in your edge and be like much more much more much quicker to move your prices, right? Then and not willing to like dig your heels in. Because like what what you'll see is like, you know, you'll get wiped out on one side and you'll be like, oh, but like the whole market, like I'm taking all these great prices to the market. And then you get wiped out and then the whole market moves after you got wiped out. And it's like, oh, actually, they're just declaring me out first, right? So I think if you don't have like an actual source of truth, you have to be much more careful in that. So like that would be the the first thing on like how hard I'm going to defend a price. Is it something like I'm actually modeling and I have my own, like I said, source of truth. I think the other part is really trying to identify your counterparty. And this is what I was just as much as you can. Obviously, like you don't get some like label of who it is or what they are or what anything. But you can make assumptions based on it. Like I think the example I was saying, like if you're posting something where you expect to get, you know, pop on odd screens or expect to get like, you know, like price shopping, like those are the betters who are going to fill your price. Like you shouldn't start panicking when you get those fills, right? Like that doesn't really make sense. Like you you posted it to get it. Again, if you're posting like closer to market values, it might seem counterintuitive, but I would actually be like more scared there because that again, it's more likely someone like actually having an opinion and not just like a top down better getting to it first. And then the last thing I was going to say here was just one thing you said is like, I really think one of the great things about these exchanges is like, at least now it's really easy to keep orders small and move like or repost or whatever. Like it's, it really is like very limited downsides to that besides time, at least currently, like time investment of maintaining that, like maintaining an order book and everything. But like you can get you can get to the like the best possible price fairly quickly and fairly cheaply, like that people are willing to take. So those would be like the three points I would have for for this one. Yeah, I'll come at this from from a call sheet standpoint because, you know, I think that's a little different. And I know a bunch is probably asking about this from a call. I know he's on call sheet like trading in good amount of volume, but the Q and call sheet is like very important because the tic sides are still big and, you know, hashtag keep the tic's big. You know, that's my big platform on this on this podcast. And I got a DM from somebody who works for a friction market that wasn't call sheet that said they agree and that they think the big tics are the way to go. And so it's just saying there's other people out there. It's not just me. But yeah, the Q is really, really big. And I think like let's walk through an example of a bunch you may be talking about. So you post two sides of main market. And let's say you're three cents wide. Or yeah, let's say, let's say the market is, let's say, let's just make up some numbers. Let's say the market's 45 the 53. So it's two cents wide, right? And you're at 45 and you're at like 52. So you're three cents wide. And you start getting a lot of 45 is your top of the Q. Well, it really depends how many people are at 53. And if you really think that you can quickly hop and get out of 54 is basically like, do I think there's two way orders coming in, you know, I might be taking a look at the 53s and seeing if they're like kind of getting whittled away at because basically what I want to, what I want to be able to do is if it's clear that I'm just going to get run through and the market's going to flip to the market's going to flip to like 44. I wish I wasn't doing this on an audio podcast in my head. 44, 54. I want to be able to quickly post the 54s and like take some off with still making a little bit of money. So I'm willing to take like a full massive one side of fill there, but I'm not going down with the ship. Like if it's, if I'm in there without a matte like opinion like SP said, then I want to be, I want to be like ready to go on the other side at the new level when it flips down basically is is how I would think of this because then you're still getting out, you know, even with fees out of small profit plus you're like managing your risk. And in that circumstance, the other good thing that happens is the market flips your direction and now you're sitting there on the other side at the front of the queue and you're collecting a three set or so like that's kind of what you're setting yourself up for. So you, you really have to think through like both opportunities and then basically play out how many times one thing happens or the other and you'll find like that being the first to protect yourself when the market goes against you and kind of post that that new level, it's going to allow you to do a lot more size and then capitalize on the time when the market goes in your direction, you capture three cents or four cents or whatever, you know, whatever you're quoting both sides. If that makes sense, I don't like it's it's kind of weird to explain over audio podcasts, but that kind of track. No, it definitely did. It's just, it's so interesting how because like when we're talking about this, like I didn't even really think through the dynamics that you will be talking about all these from like a calcium dynamic and me more from like the sweepstakes exchanges and it's so funny how different the considerations are just because of that tick size part. Right. It's just like because I'm over here talking about like, oh yeah, just, you know, it starts small and yeah, exactly. So it's just, yeah, I guess it's a good point to point out that, you know, like when we're talking about these, we they're they're very different for that reason of like what's possible, what's optimal and how you should like be trying to extract as much from these sites because the the tick size just changes the ecosystem entirely for makers and you know, it's not not not being as involved in calcium. That's that's interesting to hear about those aspects. Yeah, and the flip side is like we're we're starting to now finally get a little extra time and start building, you know, the stuff to get on to the to the sweeps and to the smaller tick size spots and we have to like rewrite our whole quoting logic, like it's totally different. Yeah, what we have for call she is not what we're going to use for those and that's fine. Like it's just different, but yeah, like reloading and iceberging on on the smaller tick sizes is certainly a strategy, especially since you don't want to like we talked about last week, like show to the sharp money tool. So like that's a whole nother consideration there too. The sharp money tool on call she would be hilarious. It's like there's a million dollars on this side. Yeah, there's also a million on the other side, but yeah, like it is it is a very interesting. It's cool. I mean I obviously have said I like the big text size, but there's it's just a different game and there's different strategies that people I think do really well on on the small ticks too. Yeah. Well, I mean, I think bigger tick size by definition just well, I guess it depends what you're doing. Like if you're fast, like I it emphasizes being fast, right? Like the bigger tick size certainly I think in general bigger tick size is probably better for like makers, right? And it's better for like originators. Yes, and it's worse, it's like worse for like I'm just thinking of like the consume like the and consume it, right? Like the it can only get so thin, right? Like how she can never get to minus 101 minus 101, right? Right. But you said like that for the day early, like I would actually push back on the it's better for consumers because you could still like bet you can basically like you'll have more liquidity out of bigger tick size and you'll have more early. And it'll be more like but you know, I don't know what the recreation of racing bets are. That's true. That's fair. Some do. I'll tell you, yeah, some do. Some do. But yeah, that is that is fair. I do think all these like the tick sizes will continue to change. I think I saw like a calcium and we can move on from tick sizes. I think they're changed. They're talking about changing their tick size, right? I did. Yeah, I saw that, you know, I have my opinion on that. I don't think it's necessarily the best thing. And we'll definitely see, but yeah, tick sizes is like the whole the whole game. And the other thing is like, yeah, yeah, it's so important. And yeah, now we now we say, we got EV Chaser PM. This was kind of like what we've talked about, but he says PM guidelines, nuance, order flow, good and bad times the bet using top down identifying manipulation or sharp action against the market. Yeah, I think like mainly we've we've covered at least some of the nuance, the order flow type stuff. Good and bad times to bet using top down is interesting. Do you have thoughts on on that? Yeah, so the I think we covered a lot of this. The one aspect that we didn't cover that I wanted to talk about here is like, I think the missing dimension of of like how people think about exchanges are like the tools out there. Like we talked about the sharp money tool or odd screens or whatever. Like I think exchanges forces you to consider a third dimension. So when you think of like an odd screen, I think of there having two dimensions. One, the price is like the first dimension. And then a good odd screen would have like the depth or the liquidity as well. And I think the third dimension like that is important. And if you can capitalize on this, I think you could have a lot of success in prediction march or exchanges would be like time like historical what has actually happened like a ticker or what like the equivalent of that. Because then you can see like what was actually traded. And I think that would help inform a lot. Like I think it's always like a joke of like 87% of the money, you know, like is on this team at this sports book. And the reason that's like a joke is you have no idea like what the you don't you don't know if that was one person like one whale. You don't know what price it was taken at all these things like I don't like if you had all of that, I do think that data is useful and you could do a lot with it. So I think if you have like some I think we've talked about it before of like understanding like movement of prices is like more important than just like what the price is right now and how it compares to other prices. So I think like this time and sort of like movement dynamic is more of like what I would lean into if I was a if I was trying to like top down or derive information like from the prediction markets right now. Yeah, that was the one thing I was like when I was trading someone was like oh what if you could only have one thing like to view the market what would it be basically like you could have like a chart like a graph of the stock you could have the level two which is like the order book. You could have your tape and I was like tape like the tape is basically it just shows you know this traded out this price for this many shares at this exchange and it was like it's like a no-brainer not even remotely close in my my opinion of like which of those is the most important. So I think that's a good point. I don't really have anything I think I think like I'm interested to see like what some of the actual good tools are that come out around prediction markets because there's a lot of like I respect someone put out a tweet saying like I'm making like a professional trading terminal for prediction markets like what would you want to see and I read a couple things in I'd like to see I don't know kind of lost track of that that account but yeah I think I'll be interested to see like there could be like a whole new wave of top down using some like tape time order flow type stuff. Yeah I think that's that's pretty cool. Yeah I mean I think that's that would be like you know the next evolution of the sharp money tool right like that can't be faked so easily like money like dollars it still can be right you'll get into like things like wash trading and like I don't know if these things like happen in in financial markets or they're illegal or what's going on with with all that but like that would be like the next layer I would think in in the tools is like the time aspect of it so um a couple more questions on on exchanges this one from from former guest doctor Plop how much worse does someone who someone have it who chooses to tackle prediction markets from the take side rather than the make side uh he he had some Spanish saying why not why not both I had to look that up Spanish is a little rusty. Low is your middle school language or high school language. I do do that's how that's how yeah that's I knew the yeah it's it's you knew what dose yeah yeah I could I thought poor K meant because but maybe that's if it's if it's together yeah okay all right so he's washed up um but yes I did take Spanish so don't remember much but why not both but have been wondering if many people are crushing it the take side so I do think most of the people I've personally spoken about and heard talk about this are all like on the on the make side for the most part I think part of that is like at least with Kalshi which maybe you could speak a little bit more it's like just a fee situation um that that's like a big difference um I do think and this goes all the way back to our first sort of question I do think there like there are opportunities on the take side like today like I don't want to again like talk about like everybody if you take a take sign on exchange you just like auto lose there's no way to win like the reality is like a lot of these markets are being quoted like several cents wide right like and they're not like they're they're not like the most they're not all just like main market stuff like there's there's plenty of markets that are really thin um that it's not like just with a snap of the finger like all these quants have come in to price these things like a lot of these are just dudes like me and you quoting our opinions and like there are a lot of people who are who are smarter than me I won't speak for you who are smarter than me who could beat me on things um so like I think I think there are there are a lot of opportunities like again if if these flourish into like ecosystems for years and years over time it'll attract more money more talent more effort and that will will change but it's like it's not like we just woke up one day and like every line is like the true god given line like like everybody know like true 50 50 and no edge can be generated like I don't think that's the case and a lot of these markets are quoted like that and I think that's because you know a lot of these companies are they want people to get to have they want to show tight spreads and they want also people to get filled so they have you know market makers with these incentives and whatnot to to quote markets maybe they don't even want but like the mentioned markets or any of these like just stuff that's like BS and won't always be quoted this way probably um like to me there's a lot of like effectively promotional money in these exchanges right now and I think that's where I would look if I was trying to take I would try and take this makes encounter intuitive but I would want to take versus market makers right now I wouldn't want to take first like a random person um because the market makers have different incentives than just making money right now like versus call sheet trading or no big trading yeah exactly yeah like I if like if I was going to go on call sheet and I wanted to to to take like it would be a huge benefit to me if I knew if like every side that call sheet trading was offering the price I knew which one it was like it's like these 12 markets like I would love to just go after those right like the things I'm scared about is like some random dude who knows something more than me who's not covering 30 markets and is not getting a rebate to keep the spread three cents wide like those people are the people I want to trade against right like so um that's that's how I would look at it is trying to find what's being covered by market makers and not just like people yeah yeah I actually I think the take side has well it's a you know obviously uh no big the lines are crazy tight now I don't know obviously there's also smart people shaping those lines and and whatnot but there's gonna be times where if you have an opinion the best price might just be on no big at both sides and you even if it's for small liquidity you can just take and you know do do fine um the other thing is you know obviously from a little talk about call sheet you're worried about the fees the takeer fees so you'll first want to you know calculate takeer fees and it's not like you basically it costs more to take closer to 50 percent so you want to kind of understand how the fees the math of the fees works and then you'll be like okay I can kind of figure out when it might just be a pure price play I think taking the mention markets you know all all of that is good but the other thing is you know how many times do you see stale lines on sports books or you know news come out and the market moves a lot like in those situations the traditional sports book situations where like the market might move a lot they still exist on the exchanges like you can still catch people and you know the thing is if it's like a you know uh really important piece of news I mean that fees doesn't matter at that point to move so big so and the size that's posted the good the good news is the size that's posted could be way more than at the rec book and you won't get your account limited for like betting uh you know a new spot or whatever so yeah I think there's definitely definitely spots to uh to take I mean it's not but I would say is it's a it's very situational you can't be like you can't overdo it with taking but um the times when you do take because of how much money is offered at some of these places like it could actually be quite good and just to maybe add one like to to to talk more about the time component that I was talking about like I would not want to like in general I think you're going to make more money on take orders the longer that take orders up like I would rather like just knowing nothing about the markets or whatever if you told me you can take an order that was posted five minutes ago or take an order that was posted two days ago I want the one that was two days ago um and so that's like another aspect of like where I think tools or whatnot could help because it could help identify like what's most likely to be stale or not being updated or whatnot whether that's market maker otherwise you know like in the injury example that's like an extreme version of it um but there's other you know types of things where what you want is like someone you where a market maker or someone who's just like left the price up there and something has changed right right it doesn't take much to change to get past the the takeer fee um absolutely yeah they're still opportunity uh all right let's take arbitrage abroad I think arbitrage abroad is a Canadian um I talked to him briefly in DM's so congratulations about the pledges and arbitrage abroad says how sophisticated do you think prediction markets speculators are how long before this market becomes more efficient um you know I don't really think very sophisticated at all I'll be completely honest uh it's early so it's hard to even be too sophisticated and I think like you know the people who may have some good market understanding lack like sports understanding and there's vice there's a flip side of that so right now and then you have I think like SP talked about as situations where the the exchanges are propping up a market and don't even aren't even gonna able to make money in it like I mentioned markets is is the easiest one to to use as an example but yeah I think like I don't know if you read Twitter I don't know I I think these Twitter accounts like aren't actually like trading but some of them are are yeah some of them are quite unsophisticated and I think like the the top of the leaderboards are usually mostly sports better is in a couple like legends like like Dahmer um you don't really have I still don't think you have like true um mega core trading desks here like what I would call like really sophisticated like that's not here maybe my answer no I 100% agree with that again that's why I think you know the water's warm right now um I think like there's there's like the LinkedIn prediction market influencers on Twitter who I talk about all these aspects but like you said like I'm guessing the vast vast majority of like the big players on obviously the sports ones but also like Kalshi are just like what I call either sports betters or advantage players they're not like actual like right traders they're just like people with better opinions than other people but that's like I don't think I think we're very early in terms of like sophistication around actual trading strategy like just look at you know like just look at the tools and like what's posted and what not like I don't think anybody like we're talking about all this because it's brand new about like trying to disguise orders and not give away positions and get you know the best price and attract the right counterparties and all of this like it's because it's all like new to this world like me and I think a lot of sports betters obviously you come from a little you know like a trading background so you definitely have like a more context on that but like I think all the the people who are participating in a serious way or most of the people who are participating in a serious way are just coming at it from like I'm just trying to bet good prices not like the delta hedge bro we haven't we haven't gone gamma inverted or whatever yet yeah and to be to also to be clear like I don't see us as sophisticated I spent my whole day trying to figure out how to make a button that will like save some command prompt entering that we're currently using and like that like that's where we're at I still don't have that button as a recording this podcast um so I think a lot of the people who are who are doing well are kind of just I mean you listen to Dahmer's risk of ruin he's kind of like I don't know I just kind of like read about it and then I kind of just like gamble on it huge because yeah just like people with opinions right like it's not like like I don't I don't I view maybe this is wrong maybe like a finance bro would tell me this is wrong and the winners in finance are just people's people with opinions but I think people like the strategies and the way to sort of maximize those opinions I think people are very very early on on that definitely definitely I mean like we talk I mean we'll talk about the call she outage but like one of the people who tweeted about it who like got I think the most like motion on the tweet was Mr. Peanut better who's as far as last time I checked like the number one public profile and profit this month and he was like yeah like unfuck because I don't use the APIs the person who's number one in public on a public profile in call she isn't using the API like he also is tweeting that he he was you know tilting and then he realized yeah he was like actually so I mean I guess the sophistication is yeah there is room to get a little bit more shout out to the stadium get it better obviously a sharp sharp guy in his college football very good episode of that the process recently but yeah that that was that was quite funny and we'll talk about the call she outage in a second but first we'll last question is from Evie Cactus who hung out with that betbash and is one of my favorite people on that betbash so he asks a similar related question will you be starting a betbash for PM prediction market dorks and there's no way I would do it because I think we'll accidentally get some of those like Twitter accounts that just drive me mad I like I can't risk that I don't know maybe like a sub that bash like everybody who's such a nerd that they like prediction markets you we can all go like count cards at L Cortez and like whoever gets like kicked out last has to make market make like mentioned markets for a whole month or something like that I don't know but no not what would our what would our sessions be we could have we could have fixing the education system with through gambling I could line some people up with that the sessions would be so funny being a hot dog vendor in the year 2025 and how you can take advantage of you know I we have to talk about your hot dog the your hot dog vendor thing is now getting serious like it has it has lore now on Twitter and the best reply was when you talk to when you were like you know the outage it really cost me because I'm a hot dog vendor and like I couldn't hedge off and then I'm playing him the Twitter handle but it's another guy met a betbash but he was like yeah like what you got to do is actually you got to get on the API and you have to spin up like etc etc etc and he says at least that's what several other hot dog vendors that I know did yeah but I'm just going to be a a rotating hot dog vendor for every Kyle Sheerated story until I'm just hoping one of these times they they reply like I I got to make I'm a little bit more believable I think so I just want to get them to reply to one of them or retweet it like that would be the best day of my 25th day like if they repeated being like yeah they should no retweet it like look at that series yeah we hold this hot dog vendor yeah that's why I need to make it a little bit more see what if you got put in the court case like as a as a as a to show why player props should be involved and they're like this tweet from this hot dog vendor I would show up in like I don't even know what a hot like an apron and like you know one of those paper hats to the court and just talk of that my hot you could literally save the whole sports betting world by doing that so I think that would be the really like best case outcome but let let's talk about let's talk about call sheet outage so on Saturday call she went down for hour hour and a half you couldn't get on to the website or the app and I'm trying to figure out how to like structure all of what happened but okay so you couldn't get on to the website or the app there wasn't much on twitter you you know the API was working and you could see the stuff on Robinhood I don't have money in a Robinhood account I just have one so I can see what's on there but like presumably at the time I thought Robinhood's trading I can see that there's some orders going through from the API so what the hell is going on and it took at least I don't know it took a while like people were kind of like tweeting out like what's going on mr. peanut butter tweeted something out that I think I forget what his first tweet was but it did start to get a lot of a lot of traction and eventually like somewhere in the call sheet discord they're like oh this you know it'll be okay but we still have trading up and it's like what the market's still up so at that point everyone's freaking out of course because it's like okay what trading is still up we have you know people have put orders out and presumably they're I mean obviously they're only getting filled if you know it's going against you and seemingly like everybody on Robinhood and who has you know API trading is totally fine so you're kind of screwed and there wasn't really like much response or telling you what was going on and yeah eventually like they it came back online there was a few tweets maybe from call sheet people about it being fixed but uh it was kind of baffling there was not a lot of communication and eventually they did refund they did refund people whose orders were out and we'll talk about that because as well because it's tbd on like what i'd love to hear your opinions on like how they refunded but but yeah i mean it was like it was it was a funny time to be on twitter i guess yeah there's some great great tweets i texted you that day that this is a good day on twitter just because of the humor um you know everybody knows in in peer-to-peer games that occasionally one of the uh the players loses their racket you know in tennis they just they they're not allowed to play with the racket that's just how peer-to-peer games work and the other side is that you know it's just big to accept them to know uh but no i mean like i i think the big thing is um they don't seem to have the the instincts around PR down yeah kalshi like i don't think they have like there's companies who just get PR like right off the like they get it like you can just tell in every one of their interactions they get it uh this is like the this a couple times now i the last one i remember is like where you know they they were like unhinged about like trying to uh explain kalshi trading to the public and like they just clearly didn't get how to communicate that well so i don't i think they need to evaluate like who's doing the PR over there um and maybe try a different strategy because that doesn't really seem to work out because like in reality like this stuff does happen to newer companies like all the time um and i think if if there was like a different PR approach like nobody would have cared at all um even not not that they wouldn't have cared at all but they would have been much more like reasonable i think the the way they've communicated in the past and like um you know they just have made themselves an easy target that anytime something is gonna go wrong there's gonna be a lot of people who attack them um you know just the way they've they've branded themselves and whatnot and and they're doing their PR um i am most interested in like the the uh the actual like payback aspect of this so i do you know exactly how this worked because it seemed i don't know i don't know but i have a theory of of of what of what happened so you are i read the the tweets or someone posed a question to us basically about this um and they had a tweet thread of kind of like their payback their their their orders that left out and then what they got paid back on so going through that i have a guess of what call she but how they paid back and it's like they but the problem is that there's one part of it that's like a little subjective so it would be good to hear like what they how they actually calved this but this is what i think happened based on the the tweet thread from uh from Garrett so we basically what happened was when call she went down you know people had passive orders up and i will say actually i'll take this moment to just um mention to people that there's a cancel at event start option for orders that you place passively on on call she and unless you're like really trying to ride that order up to the last second because like sometimes they say the event starts on at eight on call she and actually starts at eight oh seven and sometimes in that seven minutes you know there's a decent amount of volume like if that's really part of your strategy and like you're going to be there like locked in between eight and eight or seven then like i'm not going to fault you but just want to let everybody know there is um if you select it's like right where it's you know good to cancel is the standard order that you're going to put out there on call she you can change good to cancel and the drop down to cancel at event start i would recommend that if you're doing these like far in advance it's just like much safer and if i ever place something manual and it's you know if it's you know more than an hour or two before the event i'll just set it to uh cancel at event start so just reminder that's a good way to stay safe out there um but okay so you have your passive orders you roll into the outage and only the orders that are bad get taken so okay so you have a order that's that's bad to get taken you lose the bet what i think they're doing is they're only refunding first of all the orders where you lost so like if you got a bad fill but you end up winning the bet like you get to keep that money and so you're kind of free rolling a little bit but i think what they're doing is like when you lose the bet they're basically refunding you the fair market value of like the order you placed so when you when the order was was filled like not when the orders went down and not after now i don't know if you you've made it a different way but that was my understanding now how do they calculate that fair market value especially during the outage when spreads are kind of wide that's something i'd like to hear from callshy like i don't know i never saw them post like the the logic you know what up for their count year their formula um but yeah i think anytime this happens like you want to be like overly explanatory and like there should be you should be like yeah we're marking it to the last traded price or the price that you know you know whatever what like how do you calculate fair market value essentially in a spot where like the market moves and your order gets picked off right like let's say somebody scores the touchdown and you're like one cent above the market and all of a sudden they score touchdown and now you know you're like 20 cents off market and you get traded at well it can't be the price that you got traded at like that doesn't make much sense so there has to be like some understanding of like the value you lost basically is what i'm thinking it's like you go from this price to whatever you know where you know what you got traded at to like what the fair market you know basically like equity you lost so let's say you went from like the whatever you somehow you got like screwed for 20 cents basically they're giving you back that 20 cents this is my understanding and then if you're bet one then you just keep what one but how do they calculate like 20 instead of 15 instead of whatever that's where like they really need to come out and just be like this is how we're calculating it and if there's not a lot order flow like we're taking a mid and this is how we're taking a mid or we're using like a volume weighted average price like but like not just that we're doing it for the last 10 minutes or something like it has to be like really really clear because like in situations like these people always going to assume especially in the sportsfaring world that like you're trying to screw that and I've changed my mind a little bit about like I can't say that I don't think they should have like I still lean you got a halt trading but like I think they may be made a conscious decision there to be like we're going to stay up because we need to keep rather than happy and we're going to refund everybody after because it was our fuck up and we're going to do right but we have to keep it up and we'll fix it quickly as long as you think we can fix it quickly. You know it's hard for me to say you shouldn't halt but like I can see the logic there but in that in that logic in that strategy the thing you need to do is make sure that like you go above and beyond to make it right and to make people feel like you made it right because you don't want to lose good will net goodwill if you keep it open so you know Robinhood says happy and you do right by the customers and you come say hey we're really sorry this you know we fucked up better PR like all of this I could see this being an and I can see the defense of the decision to keep the exchange open or you wash the trades but that comes to like are you going to wash the Robinhood users trades like will they get pissed but what I can't get behind is like this plus a nebulous explanation around calculating fair market value and you know a pretty poor PR strategy. Yeah that's that sort of goes to my point of like they just I don't think they have a good communication strategy like through this whole thing and just other examples in general but like it was not clear at all what they were doing I wasn't affected but like it was it was clear to me that they didn't like like I think other companies or went not in the space would be very like transparent about like what's happening because similar things have happened and this certainly wasn't it it is really weird like how they like to me not having thought through this a time it seems like you should I understand the point about keeping it open if you want to keep it open to me you should just reimburse based on the value when the people lust access to trade like you shouldn't do this convoluted like from time open to when it actually got matched because you're just going to those people just like you can't use the prices it was trade that like you said you're just those are all adverse selection like prices right like so that you can't use that and I don't even think those people really should be on the hook for anything right like if they were if they weren't unable to trade like they should basically get their money back from that moment to me like that to me is what makes sense um obviously that would probably cost more for kalshi because I'm guessing you know you had mentioned like it's certainly probably some some people did get lucky but like overwhelmingly what got taken the of the APEI users were like off market prices and good bets for that right so I don't to me like no matter how you even if you calculate the fair value in like a truly fair way which I am a little skeptical they did if they're like using their own prices and what was traded um it seems like they should just get the price when access was lost no yeah I think I think that that is one way or you trade you do it you peg it to the price like you peg it to like the best price like five minutes before they were trade happen or something to like get away from pegging it to the price when the market completely shifted and they were just sitting on an island and got and got taken you know but but yeah I mean honestly what should happen is you should be able to like win that change goes down like cancel all open orders um probably I mean I understand like your APEI users will get like their batch canceled but they'll just refresh so like you should yeah it's whatever this happens like this is not the first time in exchanges you know gone down and not the last time and whatever but I just think like yeah the first time it happens like I think I agree with you like you have to be overly like generous to the people that screwed over and then basically what you do is you say okay obviously like this is bad and we have to have you know maybe like we have to have a better plan for the next time this happens so maybe we whitelist like the market makers who use the like solely the APEI and then that way like take that whitelist off and then if the exchange goes down you just kill all open orders for like non-whitelisted customers and you'll probably end up basically doing a good enough job there protecting people um but yeah you certainly don't want to be seen as like skimping people on the payback like this it's okay like everybody is aware of people make mistakes like I know you know sometimes on twitter it doesn't seem like it but most people are pretty reasonable and you understand like it's hard it's like really hard to build the like a functioning betting exchange it's not easy so you know obviously I understand that it goes down but yeah this first time like you really want to be overly generous and just improve um and at a minimum be transparent about how you're how you're calcing the rebates and you know it just doesn't seem like that was the case yeah I think I think that's probably all all we got in that one if we have other prediction market news in that there's been a lot of talk this week about about uh different entities whether sportsbooks or otherwise choosing to hedge their exposure on prediction markets I think the the thing that kicked this off was underdog saying they're going to they have a desire to hedge their exposure on prediction markets and we probably don't have to spend all too long on this one because I think there's been good discussion on twitter and whatnot but like maybe I'm um but like there's a very easy way for uh for risk you know sportsbooks to hedge or get off positions um what does that they can they can move they have this magical thing called the price where they can move it and uh they will get action on different sides I mean I've seen different arguments of that like that's easier said than done and I've seen discussion both ways but I think you know the the point that I I found the most compelling um and so the discussions I've been following was like you want to keep that money in your own ecosystem if you can right like you know working in insurance and whatnot like obviously re-insurance and whatnot is a big big deal um but those are for like true catastrophic losses like there's no such thing as like really a hurricane in the sports vetting world for the most part like you can you can like move prices and attract enough liquidity like it is possible to do much easier than you know like for an insurance company or something um where like there's no actual way to to get off risk besides like um selling it to someone else so I don't know like I at least for the big players like this seems unnecessary and there's better ways to reduce risk whether that's um you know giving giving better prices to other VIPs or or big players or whatever I've seen discussed um but keeping the money in the ecosystem is almost certainly what you want to do you don't want to just like send that money elsewhere yeah I I agree I understand like where some of the the people who are defending this as a as a take are coming from because in you know in I would say like yeah like in the insurance space and finance space like there's definitely times people like layoff need to layoff big amounts of risk to other firms or companies in the space and there's some like um negotiated trades that that go down and what not just to make sure basically you don't go under um I don't feel like a sports book will be in that type of position but uh and also if they do like like you said like how efficient is hedging on PMs I don't know like like what are you going to get if you're your underdog really like let's just forget the takeer fee right even if we take out the takeer fee if you're underdog for a click on a main market you're going to get somewhere from between depending on the sport like 200,000 to like two million if you're underdog you can't you know is that really like going to be that's the end of the world or like then you have to go take a few levels but at that point you're getting such a bad price like you've really kind of gone now on the other side yeah I I don't really see this I really see this but it's good to like one thing I'll say like as somebody who's certainly smaller than underdog but like sometimes stuck with physicians I really like the function of prediction markets and being able to like get out the other side for a pretty cheap price as you know when you're passive or whatever I think having a couple more PMs in the the ecosystem that you can reliably um kind of get some risk off on isn't a bad thing I just think it's better for like you know bigger debtors not necessarily like massive companies like I think it will still just be good for betters to have that opportunity to get to to lay off some risk at a reasonable price at a couple other places as far as like it extending to underdogs or higher like draft kings and vandals of the world yeah it doesn't yeah I think you got to get it keep it on platform layout you know you have a really easy way to to our most things and if we're talking about SGPs I mean I mean come on like that's going to be like you you get stuck with like you get stuck because somebody it like let's say you're underdog and you have that like correlated golf angle up so you get stuck with like everybody played the overs of the British open or you're really going to be able to go to like a prediction market and like catch out of that like I just don't I don't see it yeah I agree with all of that um next news item we had was we probably don't even if there's another NCA basketball sports betting scandal these these these are happening like every month right so just just know what happened yeah we can we can move on um two other items we wanted to talk about in news um first let's talk about uh poly markets um 15 minute bitcoin up down markets um I tweeted about this the only reason I wanted to to talk about this is like uh it impacted your hot dog stand yes now I'm I'm insulated from Bitcoin phrases when I'm accepting uh Bitcoin for my hot dog stand and do buy yes uh no but like to me this is just going to keep trending in this direction of um this is how you do I casino in prediction markets like this is how it'll be there I don't think there I would be like I think these companies are smart enough to like not just like put a slot machine up there it'll look like a slot machine but it'll be like you know the three the the first decimal point the second decimal point and the third decimal point of the of Bitcoin's price and in US dollars or whatever and you'll you'll spin that and if you get them all right you'll you'll get a payout and you'll be predicting though like what what they're actually going to be and it'll be an event contract and it'll just keep getting more and more gamified and shorter time frame and more ridiculous to the argument that this like has any sort of actual um value and i don't know it'll be interesting if like like i think calcium is probably afraid to push this these boundaries um polymarket seems less afraid other companies who are going to try and overtake those are going to be even less afraid right and so i think we're going to see some really ridiculous stuff um from like third tier prediction markets in this like straight up gambling space in the next year or so do you worry about it having a negative impact on sports was your i was actually thinking about this like if i'm if i was draft kings and the next bullet point you know draft kings is getting into prediction markets but let's say you're draft kings and you wanted to like kill prediction markets like the nine this is a nine real level plot would be to like yeah like fund some company to do this to like do slot machine prediction markets and be like and then you would get you know like some US senator like just being horrified at this and bringing it all down so yeah like i think if you you want these prediction markets to succeed and be like federally licensed and sports to be allowed on them i think this would like definitely be a negative thing i think a lot of the prediction markets are in this tough situation where you you don't want to be the one who's not pushing the boundaries right of like these ridiculous markets and whatnot because you know people like them but you also don't like you don't want to be the one that caused it to all crash down probably either um and so it's like a little bit of a game of chicken and that's why i think like these third tier operators or whatever are going to be the ones who say like we have nothing to lose let's do this and um yeah we call she's the the no vigs the the poly markets are probably going to like not look fondly upon that but yeah that that will be i think ultimate like eventually it's just going to get to peak ridiculousness right like we're we're not even there yeah yeah yeah no no we have a long way to like if we'll get there to a point where it's like tot like that that's when the rubber will meet the road and i think you know decisions will have to be made on like how how the us actually wants to treat this but that's to me that's what's going to be the driver of get like of getting to peak ridiculousness is like some you know they have started to be the stories of like you know i lost my house playing bitcoin slots on right whatever the new exchanges right and that that's when that's when this will start yeah yeah i mean you know it is basically any market that comes up like this makes me nervous but if it's sports i'm like you know basically i think prop markets is interesting i do feel like there's just like maybe a different like it shouldn't matter it's the same as as betting actual sport event but i feel like almost you're like really like you do parles too you're just really throwing it in the face that you're your sports book so i was always like a little nervous around around that but yeah i mean you know this is we tell you have candy crush for money if you think parles are bad man right right well candy crush for money if you play peer to peer without where i've never played candy crush it's a one-player game something like that i honestly don't know that's how much of a boomer i said that game's probably not even popular anymore it's just always the game i think of when i think of like gamification like the mic yeah exactly yes so yeah i don't know it's it's yeah we'll see hang on i mean you know obviously we're all written for them in the courts and whatnot i would not be if i'm like the lawyer who asks like go up in court i'm probably like come on you know like uh now i have to go make it a case for this you know but you know we'll see uh mba starts yesterday if you're listening to this so you know good luck to everybody and that's i wrote in the the doc is this news do we care so i guess like we'll just leave it at that but mba betting starts so to Calvin and hobo and all of our mba betting friends good luck and then the big news that popped right before we started the record is draft kings actually finally acquired railbird so they will be going into the prediction markets the one thing i noticed when i was reading through the releases they didn't explicitly mention sports um you know fandal kind of took a similar approach so obviously like they have a lot to lose and there's been the the pushback of state sag if you do you know offer these sports contracts you'll lose your state game license for call sheet they're like okay i literally could care less but for your draft kings and fandal like that's extremely significant so they are playing it super careful but yeah they mentioned basically all the stuff besides sports in there in their release and i imagine that when they first roll out they will be doing no sports that that that is my guess um and you know i think you share a similar sentiment yeah i don't i don't think they'll do sports i think you know what will be most interesting to me is if they and it probably depends on like how long but like to me if they're able to do sports today like kalshi i don't know with your opinion to me like kalshi would be like dead like they would never they wouldn't have like a chance i don't think in my opinion they haven't built up like nearly a big enough lead i think i think they need a lot longer like to build like we talk about about this podcast but like if you're not a sports player you probably have like no idea what they are um or you're not in like one of those i think even many people in those those states that don't have sports betting but like yeah i just think they would have a tough time um competing with fandal draft kings if like the floodgates were just like opened right now absolutely it'll be interesting to see like how long they actually have to to hold that lead if if draft kings is um draft kings and fandal are on the sidelines at least with sports for now i bet they're built in their their apps though for sports like on the pm rail and that's a thing like it's gonna be good it's just gonna look so much better so much faster right because they know sports like they have they've built many sports products right so they have they have people who really understand the space like unlike like i know how she's improving quickly in this regard and trying to get like more sports centric and whatnot um but like say what you want about like like for example draft kings pick six say what you want about like the game and the the fundamentals but they basically like rolled out a product on par with like price picks like in terms of like oh yeah uh user experience like media like yeah like immediately right like so yeah um and i think that like their brand recognition is also like even in those states that don't have legal gambling like way more people in calisthenics. They're probably still showing where we live they show ads occasionally right and it's not legal here so like um yeah like way more way way way way more people know fandal and draft kings in california and then then no kalshi right for sure um so yeah for kalshi's sake i think they probably want this legal gray definitely area to continue for like at least a couple more years yeah it's it's literally draft kings in fandal all over you it's like we've seen this exact movie before they just needed a little longer that legal gray area and then when they have the recognition that decay and fandal got at the gfs boom then you can then you can kind of you know stand on your own um okay the the what i would call like the more random mixed bag q and a the non prediction either though the first question i had something to do with exchanges but like it this is going to be like more than normal q a day so yep yeah this first one yeah this is this is an interesting one uh it's good to come right off the news so we'll kick it to you first we have i a s 0 801 asking if you could only place bets on either exchanges or sportsbooks for the rest of your life or the rest of time which would it be and why to be clear if all exchanges go under you can no longer bet and if that's a very easy answer if you add sweepstakes slash pph to the loser would your answer be the same it's a very well structured question yeah so for me i don't i don't think this is particularly close in either scenario my answer would be sportsbooks assuming you define a sportsbook as like broadly player versus house type situation like because i don't know if you would consider like when i was thinking about this there's like dfs 2.0 like those types of games sportsbooks i would consider them sportsbooks um like yeah so that's that's that's i was thinking but not dfs 3.0 no that would be like i would actually put that like in an exchange i guess like category it's different obviously but like we'll keep it there yeah we'll just keep that out or whatever the the way i was thinking about this is like exchanges might be very very big um but in my opinion like the leading sports wagering products are always going to be like there's there's there's always going to be sports wagering products which are simple gamified for recreational people that can be exploited like i'm of that opinion that that will always exist like i don't think the majority of people will ever it will all like there there won't be opportunities for like gamified player versus house simple stuff like i think if anything things will trade trend simpler more gamified what not and i just i don't view the exchanges as the ones who are going to get there with with that stuff personally um you know i just look at like what i think like the price picks and the like those companies like i always think there's going to be like a new version of like something like that where it's like real simple it attracts casuals and so like that's always where i would rather spend my time personally like i don't think if these exchanges are like these huge liquid markets 10 years for now like i'm honest with myself like that's not where i want to be like i probably i i almost certainly won't be able to win there like so and so like from that vantage point that i would seem like a bad option the other thing is like i don't know what you would put the probability that like these exchanges cannot offer sports but to me it's like considerably higher than zero like i don't i don't know exactly where i have it but it's it's definitely not under 10 percent for me um so that also makes it sort of an easy answer for me even with um in the other part of this question that i was just wanting to just like adding sweepstakes pph like to me those are not long term things like a sweepstakes exist like 15 years so now i will be amazed the problem with that is if you add them like we've say for sure going down before it changes no matter what yeah right and i just i think the world the the country is going to trend towards like hopefully more um uniform like gambling laws and so like i think pph and sweepstakes and all this like i think again it's going to reach like a boiling point where there's going to have to be some sort of overhaul or whatnot um so those don't really move the needle for me um so for all those reasons i would i would go under like the versus house model yeah sadly i have to agree um yeah the first the first version you're like okay for sure sports bucks then you the second version i do like this question i think it's the parameters are very clear and you get a second crack at it i just don't think the second the extra the cherry on top is really really worth it with sweepstakes and pph i will say like for the next year it's not even close but it's the flip for me like then if it was the next two years then maybe the next basically until Trump's out of office i'm clearly picking exchanges over sports books and um you know that's where we're investing most of our time uh long-term it's really hard to it's really hard to well like the written like we'll talk about like risk of ruin like you have a legitimate risk of ruin with picking the exchange option where with sports books it's really hard to see a future where you're not able to bet on sports books whether offshore or onshore whatever like i just i don't know doesn't feel possible basically um so yeah i i would whereas yeah i mean the probability exchanges came off for sports i don't think i have it under 10 either so like scary scary enough like i definitely don't have that under 10 percent i think that'd be crazy to have it under 10 percent so yeah i mean just from a risk of ruin standpoint long-long term i would have to fix sports works as well so where is that big big big big fans of sports book love them uh next one from two chains i'm an originate not the real change baby you don't know yeah baby yeah all right i'm an originator a small market and a band and band multiple times limited on bet online and trying to figure out account long jeopardy i've heard you mentioned the shipper podcast for account long jeopardy tips but after listening to it my impression is that his advice is geared for top-down betting you know shipper would be you know disappointed i'll do that i'm sure um and my misunderstandings specifically when he says your pnl should never cross zero i i actually have a slightly positive ev market that i could use to make a portfolio but that statement on pnl has my head spinning so i i was hoping to go back and try and find the exact like what i know they're talking about today okay did you find it i didn't look i have time today shipper says your pnl should never just cross zero okay that's what i thought he either needs moon past zero or like be negative okay so yeah to that's that's where i was going to go with this i think what what he was trying to communicate was exactly that like you want to adopt a high variant strategy where you're looked at as a loser at all times until you're not right right but you don't want to like like it's much better to be you know negative negative negative negative up 50k then like you're just never going to slowly inch your way up to whatever number like i just made up a number but what you're you can't inch your way you can always variance your way up to a higher number than you can inch your way up to that number so i think that was the uh the suggestion there was like don't like when you cross zero cross it meaningfully um and so i think that that's what that comment was i don't know if you had anything else uh oh the other thing i guess on this question it was about the top down bed no i think you know shipper primarily was uh you know shaming the top down bed is not a podcast so i don't think it's specific to the top down top down bed at all i think it was it really i think was applicable to everyone i think there was a lot of tips what whatever you're doing about um basically having like a mixed strategy for lack i we don't need to repeat everything but having a mixed strategy for lack of better term um to to make it more confusing for a trader like when reviewing your account but i think it's applicable for all betters yeah yeah i think the key is the mixed strategy that that's the that was my big takeaway from from shipper and if i were to apply it to two changes two chains situation um basically you got your your small market that you're winning right that you're originator and that's good that's a good start because you know it's not going to be obvious stuff that's off the screen um the problem is it's a small market so then you also say you have another slightly positive EV market we don't know if that's necessarily a big market or a small market that would be make a big difference but basically what you're going to be wanting to do here in the shipper strategy is not just have bets from the small market create basically a portfolio where you have your small market stuff that wins at whatever percent you have your slightly positive EV stuff that you can kind of parlay with it and then some negative you know or close to break even stuff in other sports that you're going to either parlay with the other stuff that you're going to bet you know straight and you want your portfolio say you're i think the the the the big takeaway of the shipper podcast is say you're good stuff wins at 10 percent put into portfolio that looks so much more square in the portfolio wins at five percent in the last you know yeah five x as long and make more money i think it is just of it so if you think of it from that standpoint you know it's not it's not really about top down at all you know how you get your good stuff it's better to be bottom up because it's just like less likely to get attention drawn to you if you bet this you know if you're top down and you're betting the thing that's on the top of the odds GM screen you know a thousand other people just bet it so you know you're already dodging that if you bottom up so actually with the shipper strategy i'd say you're in a better position being bottom up and it's just all about the hard part is basically finding the rest of that portfolio that keeps you keeps you alive but also keeps you winning that i think is the the big skill behind shipper strategy speaking of the man from the shipper what are some bets where you believed at the time you had a massive edge but in in hindsight you truly had no edge at all i love this question because i always put this on i don't know if you've noticed i always put this on like our interview questions yeah i try to and we never get to it because we always felt time with other stuff yeah well we just get busy with other stuff but i usually skip this one but i think these are great stories of like where you're wrong so i do you have i i i came up with two today i have two also so one i've kind of talked about recently like a little bit was the writer cup um i do think looking back that so i thought i had a really big edge on on us uh and there you know if i'm thinking about times where you've made like really really big bets like the Jake Paul fight um and then another one i'm going to talk about but uh you know this was just kind of in line with that sizing wise and what not and and we lost and you know i mean i would not re bet the numbers i was i was betting like some of the early numbers i got maybe worth a flyer at a smaller size but like some of the worst numbers that i still bet like i think were negative ev and i thought i had a huge huge edge on it um and that was the first time i really was like in one of those spots um no sorry the second time was the the the next one i'm going to talk about but that that was like probably the best example of that for me i've talked about it a lot so i don't want to like hammer into the ground but another interesting example was the Patrick Mahomes uh this is Steph Curry and Clay Thompson playing a 12 whole golf match against Patrick Mahomes and Travis Kelsey and they were playing like they were playing a scramble but you didn't know what like this the type was until like not too long before the tournament and we were actually like scramble was number one on our rankings i was like okay if it's a scramble like it's the best for us and this is all based on vibes because basically the scramble is each player hits and they pick the best shots so and that i was like well in that matchup it's like okay Steph's clearly the best golfer he's like played on a corn fairy tour event i'll be it you know it didn't make the cut but he's really really good and then I had Clay like probably as the worst again these are like guesses and then i had like Mahomes and Kelsey as like okay they're okay but the gap between like like Steph and like whoever the best was of Mahomes and Kelsey was like the gap between like Scotty Shuffler and like the worst player on the corn fairy tour like it's just this enormous gap so i was like okay so this is like and it came out like minus one fifty and i'm just like this is a shipper would say a rort and it honestly like wasn't really that close like Mahomes and Kelsey just crushed them and i assume like i was just super super wrong they set the course up kind of easy like i didn't think it really was best suited for like Steph but i was just like oh my god like maybe like maybe that was just absolutely horrendous bet like maybe having like the best guy isn't really worth that much in a scramble or something like that so but then Bryson De Shembo put out his youtube video with Steph and i watched it and it honestly Steph probably gained like gained to a PJ tour player on that video and i was like hmm okay i'm re re-evaluating that i don't know where i finally land on that because this is like two years later but um yeah i'm always evaluating the huge spots that i lose like and i will to the end of time i'm never sure if they were good or bad uh do you i don't know i think it's easy to assume they're bad because you just lost some money uh and when i was talking about the rider a couple and bet the process like rufus was like well you know it might not have been bad and it's true but it's so hard to imagine it's not bad i don't know it's uh yeah yeah it's no one understands that it's like like we like we i don't think we can ever understand variance really like truly right so my three i think in my three i know they're all bad so this is that's good uh i i i could be certain they're all bad uh so your your your examples were more forgiving i'm gonna really make myself seem dumb on these three um so first one i think i've talked about before was um i remember when i was betting tennis derivatives like i didn't understand like the importance of uh serve like dominance and how that changes like spread so i just remember being like you know i i i think it was john is there or something like the guy who's super tall and just like uh is is like uh yeah yeah yeah serve with him and i'm like oh he's like he's like minus 210 like those people you know on average win like by three and a half games or whatever and then i'd watch it and like you can't move right and it's just back and forth but like the other guy can't return a server drive so that those words that was an incident today where i'm like wow the market does not does not realize how much john is there's about to dominate and uh he didn't dominate so that was one um the other one too that i've never i don't think i spoke about in this podcast um df s so dfs is actually like super common um because like in sports betting you make a bet you know the price you have a price everything's locked in you're good in dfs regularly you know i'm mostly when i play more seriously mostly playing like smaller field stuff or like you even down to head to heads three man smaller field stuff and you have to make guesses on whether what your entering is going to be good or not um and so you know during covid there's a lot of people who there's nothing else to gamble on and you'd occasionally like run into like someone post across the whole lobby you know 10k head to head 5k head to head 2k head to head and you've never seen the user name before or something in some sport you play and you obviously don't know like what that person is all about you know i remember trying to like google um they're they're draft kings user name to try and like find out who this person was to try and like see if i should should take the games or not but there've been many times where you know you flip over and you realize you're you're like literally flipping a coin at best for you know 20k or whatever and head to heads but that's just how go like that's just part of the game there so i don't necessarily feel so bad about that um but certainly there's times where you think someone is making like huge mistakes in dfs too we're like oh my god he's he played him how could he play him he's that's a horrible play and then you play him enough and you you realize what that person is understanding that you're not so that definitely happens too um but my favorite one the one that's gonna make me seem the dumbest is i remember for a super bowl um there was a market it was something like will this punter on Kansas City i think it was have a a punt be down within like the ten yard line or something ten twenty yard line something like that and i uh i was doing a lot of super bowl props that year i'd priced that one out i did all all this work tried to adjust for like you know punter specific stuff game specific stuff all these things and it was like an enormous etch i'm like oh there's no way he gets one inside the twenty or ten yard line or whatever like this is huge i gotta like start getting down as much as i can so i bet like all the easy places to bet and then i was i'm like starting to think about how i can get more down i'm like wait he i i've only modeled he's gonna put i modeled like what's the probability one punt goes in the uh goes goes within like the ten yard line and then i like plugged in like you know like what an expected number of punts was it it was like negative eight percent or something and i'm like oh okay well i'm glad i didn't get any more on this but uh that was a uh another time where i'm like wow this is this is big and then uh good and the thing is like i knew like because this this is a market that wasn't on any other book and stuff so i'm like i'm probably doing saying wrong but i'm like looking i'm like no other book had had this like maybe they're really screwing saying up and no it was it was me so that was that that was the that they're an example i uh those both both spark two and i mean one one impoker was just like the fs like i remember playing this guy heads up now i was like wow this guy's fucking terrible he's like limping the button like way too much all this stuff i'm just like playing this guy and i'm not winning uh and i plan like again like next day and not through too well and i go to like my friend and he was like yeah that's like the second right guy like in the world i was like oh i thought it was terrible so i guess i like that happens in dfs that happens in dfs where you're like this guy you i've never played and then the guy like you know so yeah so that i was one of the the the punting example it reminded me of uh bruh was a brison at oakmont so like oakmont i was like redoing some modeling stuff and i had to like build the course out from scratch because there wasn't like historic strokes game or shot data on on the course because it's like they only play it for the us open and in last time i played it was like 2016 so like i have to build out what i think the course will be and somewhere in like that logic i got accidentally doubled the amount of t-shirts of player hit so i have brison as like just the absolute goat like because he's you know the best off the tee and i'm like oh wow they're underrating brison here this is crazy like so i'm like going out and my part is like feels like a little high i'm like man i don't know but like you know it's a major course it's really long so you know in all of this and um yeah i we've had some brison early i'll say that and uh that's where benefits to have a part yeah it would be like this doesn't make any sense still that brison early um didn't didn't go as hard as we could realize i doubled the number of t-shirts on the course cut it down well let's just say it was less of an edge to to no edge there so that was a that was a learning experience kind of like the punt we all have uh all have stories like that i i'm sure i have many more i like those stories so i'll just not also try and think it's fine and then when you set the years out like they sparked one we should if we do it with gas like i'm sure we'll remember more too yeah exactly exactly so we'll do more of those because they're they're fun stories so uh next question in the failed modeling projects episode you both seem to agree that certain sports like NASCAR are very modellable but that NASCAR markets are too small for the effort to be worth it can you elaborate more on the market size pieces question how do you judge the size of the getting a given betting market so for me i think it's three variables um that we talked about before it's and i think most people only think of one of the variables the most obvious one is limits right i think everybody thinks of that but i think the other two things you should consider are the number of markets and the frequency of events yes and i think those those are under discuss so for something like NASCAR uh i think the limits are like low relative obviously the other sports like NFL NBA and they'll be like that in of itself isn't the the ruling out for me the things that would rule it out because like if something happens like like tennis for example or esports like the limits could be quite quite low but there's so many tennis matches there's so many esports games right so like that that's how i would think about it NASCAR it being like once a week and the limits being relatively low and the market like with meaning like there's there's some matchups there's obviously like top 10 top five all that stuff but there's not like tons of markets um there those would be the reasons for me yeah that that's well said i don't i can't really add honestly anything to that that's exactly what i was saying so we can hop on to trade gorilla this is a good one uh if you got wiped out this weekend and had to start over again monday with a 1k bankroll where would you start and how would you turn it into a meaningful bankroll again um i thought a lot about this because at first i was like well i would probably go take a road trip and do some like deposit bonus um you know matching right like at 1k like that's clearly the best bet from like kelly stand point but then i was like well i don't know if i have any more like spots really and if i'm staying through this question i am not just starting over we i've got wiped out it's not like oh i'm restarting and having deposits on any sports books so i think if i really had to start over with with with a thousand bucks i would honestly do it still i'd probably try and trade like a let's say like delta neutral i would trade a very like capital efficient strategy on call sheet or poly market i would go to the one of the pms but that's only because i feel like i've done a lot of work on there recently um and i feel like i could turn my bankroll over uh low risk kind of quickly in certain spots there um and yeah i would i would probably probably go there the problem it this here's why it's a problem because you're it's really hard you're not going to be able to work with like partner actually yeah that's the question okay i assume i can't play in credit with anybody too but like if you have 1k and you want to work with a partner it's like not worth their time to put them that's for you on wreck books too so you know i think i would have to go somewhere where i could be clicking and you know have experience so i would still go to the pm route yeah so i mean there's a million different directions like you could answer this question like because you like if it was so i i brought down a couple different scenarios if it was me personally like right now if i got wiped out like i have a job i would probably not bet until i have enough money lame worth it again right right so i i mean i would just like if it was me personally i would like i wouldn't bet with a a thousand dollar bankroll like at this point um just with the time aspect of it just wouldn't be worth it now i know that's not the question you you actually were asking so if i if i you know i couldn't add any money from other sources or whatever i had a thousand dollars and my goal was to to grow it as quickly as possible what i would probably do would be to try and live bet like get some yeah some type of live betting software and just live bet it because i think the thing you said in your answer that i would echo the most is if you have a small bankroll you want to turn it over as many times as possible right like and as fast as possible like if you have a thousand dollar bankroll your whole bankroll should be like in play you should be doing things where like your whole bankroll should be like in play basically every day there should be enough spots where you can do that like in a i'm not saying just like throw it all on one thing i'm saying but i'm saying like like with live right you can actually turn it over more like you can very easily bet more than a thousand dollars in a day if you're doing live right even if your bankroll is a thousand because you could turn it over several times so i think that's that would be the biggest thing if you have a small bankroll and like obviously live is like time intensive but if you have a thousand dollars like you should you should be more willing to trade time for money right if you have a million dollars you should be less willing to trade time for like your time should be more valuable so that that's probably what i would do is go live and try and run it up to enough where then i would move to like some and more the exchange stuff yeah and like on top of the live capital efficiency like on exchanges you'll have like position netting so you could trade you know like you could have you could take a thousand theoretically have like a hundred thousand traded on the game as long as like your net position was you know below a thousand not that that's super easy to do but you know that that helps okay let's go to Ben's question this is about you know how do you guys think about realizing profit from your bankroll i when do you pull from your bankroll to pay your bills or to whatever i think i'm really i'm curious i'm curious to hear your thoughts on this um i guess i could read the whole question so do you take a salary if you're in a down period you have a skip paying yourself uh has this changed uh when bedding's gone from a side to main gig or as your bankroll has grown um i i can kick it off uh yeah i mean since this is my full-time job you know i obviously have to take from the bankroll with at least semi-regularity we try and do it quarterly there's times when i've definitely kept money in the bankroll as like i try and keep money in the bankroll unless absolutely necessary uh especially now when i feel like we are really like using our whole bankroll um in a slower time maybe i would be more willing to pull but yeah i mean i'm let's just say this i'm never pulling money out of my bankroll just like put it in a in the s&p i'm never doing that if i'm pulling it out it's to pay expenses or you know use it for life stuff like i'm i'm i'm never taking money out of the bankroll just to like have it in a cd or like whatever so the bankroll is the investment vehicle um and what i'm pulling out i'm pulling it out if if i needed this yeah so i mean for me again this is different because i have a a a job right so like i don't have to necessarily pay myself um the same way having that income um i mean like honestly i don't we sort of talked about this before like i don't really even think of a bankroll the way most people do like it's sort of amorphous when you have a job right like do i count my uh you know earnings next year in my bankroll i don't know sort of probably um right so like for me it's it's much um for me like it's it's more like i want to keep as much in my like hypothetical bankroll as long as i can use it all right like i think you don't want to pull stuff out of your bankroll if you're still using it all like if you can avoid that obviously like take care of your life and family and whatnot but like you only want to i deal you would only pull out of your bankroll if you um if you're not using it right like so i think anybody who's who can use it like it's best to try and keep it in there right it'll allow you to bet more and make more in the future um i think for people who have have jobs here like other sources of income um i don't know this is like a really hard topic because i like i don't personally like segregate my money out that way right like i sort of view my bankroll as my net worth and like my net worth is like like i said do i count like what i'm gonna make in the future in that um because we've talked about like kelly and all this stuff but if you told me i'm gonna you know get a million dollars at the end of the year i probably shouldn't bet kelly based on like what i have available to me today right like i should be more aggressive obviously so um yeah yeah it's like is the bankroll in the room with us right now i think it it exists you know if people have a job maybe that's a a stranger wrongtake but that's sort of how i think yeah it's it's it's it's it's certainly it's it's a really interesting question um but i think you go down to like one of those things where you're never gonna have the perfect formula for you and you kind of just got to you know do what you think is right and certainly like we've leaned we've leaned on the side of keeping money in the company um basically at at all costs except for like ones that would disrupt our our normal life um okay Louis question I thought was really good uh and we not that the other ones weren't good um but i really like this question he said uh he said explain your opinions on regressing model output to the market post generating numbers versus using the market as input to the model directly pre generating numbers so basically like i think the question is um if you only have market numbers uh or you're you're actually sorry i'm actually misread what he was was asking because i was taking it from like a market-making standpoint but uh using uh basically like regressing your model to the market versus using model or market numbers as a parameter to the model okay so i think that regressing your model to the market isn't like one step so you want to know like a who is the market but this is i think of it so you come out depending on how how good you think your model is i kind of judge it by like would i bet into x book at x time and that doesn't mean i'll never regress to that book um like just because maybe i would consider betting you know close to close to post on maybe one or two golf matchups out of sharp book it doesn't mean that i won't regress to them as well uh but you know if i'm always gonna be betting my edges at post into draft kings you know or whatever like would i regress to them probably not but i think like regressing to the market you start out you want to be early you're going to be weighted more towards your numbers than the market right because you assume hopefully that like the opening number will move towards your number if that's not happening that's probably pretty concerning not i mean sometimes it won't happen but if that's not generally the trend that's pretty concerning so i think like how i think of regressing it's like i like you know generate your model and have your model be the main input right away and then basically like a time decay function like you decay towards the market over time uh and how you do it or who you peg to that's going to be sport specific market specific so i don't want to get too far into that whereas using it as an input i actually don't really like using it as a parameter in modeling because i think like a lot of the value comes from finding stuff the market isn't accounting for so i like to like build with no market input and then use basically like prices to see if it gets better so like example of this is if you're back testing um you would let's say you're back testing your model again other people will probably disagree and they they might be totally right but in my opinion like your model shouldn't use a market number as a parameter so what that means is like if you're doing like an ml thing like it's not going to be like uh you know record against the spread and you know that crisp money line and weather you know it's going to be weather in record against the spread and XYZ all the stuff whatever you put into your model without the crisp thing and then you generate something and then you run your back test or SP hates back test so maybe you run your forward test and then you see how well did it do and then you see how well did your model do go a 50% you 50% Chris or 50% you 50% Chris close 50% you 50% Chris open all of that that's i think in my opinion how you want to be factoring um the market in because if you're including the market as a parameter i think it's very nebulous like what's good you almost like black box your model because you don't know exactly what information the markets including or not including so if you add a new feature like are you double counting like you don't really know so i find like doing it after and like improving your model and then testing to see how much you need to regress it is the way to go um but you know i'm hoping to other opinions yeah so i was thinking about this one this this morning and my analogy is you know i don't know if you're a big baker or or even cook but a lot of times when you're cooking you know they they always tell you to use like the unsalted butter or the unsalted broth like when you're making something and the reason they say that is so they like you can control you know like how much how much sodium or whatever is in the final product like a chef won't use cinnamon sugar they'll use cinnamon sugar right so i think that that is why that is like the analogy of why to do it after the fact and not just put the the line into the model as a as a as a parameter whether it's you know a predictor or whatever um you generally like we were saying like you will lose control and you will lose interpretability are the biggest things um like you don't it's much harder to understand what's driving your predictions when when you have the the market number in there what i will say is there are times where i think it is appropriate and better to put it in the model but they're they're specific so i think the times i would put it directly in the model would be if i'm trying to capture like some sort of interaction between the line and a different predictor that i wouldn't be able to do after the fact so like a really stupid example made up example is let's say i think rook the games with rookie quarterbacks have wider tails of like final score margin you know home team like if a rookie is playing you know at ten point margin is more likely than you know all of else equal like these blowouts because there's more uncertainty or something um if you just like have some model about you know the dispersion of scores and then you like you can't capture that interaction unless you put it in right like you put in the line with the the rookie cubic flag or whatever and put those in a model together right and so if you do the whole model come up with some probability and then blend to some line you're missing that signal of like the interaction if that makes sense um so if you're trying to identify like some systemic bias unaccounted factor or an interaction effect i do think it makes sense to do something like this um but otherwise you you pose the risk of losing control and flexibility the other things i just want to mention if you use the line directly in the model that that you introduce you know problem potential problems is like what line are you actually putting in so like usually you have like an opener or a close or something in between but you're let's say let's say you have let's say it's a Wednesday right and you're you can either put like openers or closes into your model but it's it's Wednesday let's assume NFL like and you're trying to predict what the price is should be like what are you putting in are you putting in the opener are you putting in the close like so that that's one problem is just like the actual line the other problem is even if it's like the same line all the time like you're always putting in openers like there's non stationary the lines are not stationary like so like a line from this is a problem with all predictors but like ideally when you're you're building any model you want your predictors to have a constant relationship with your outcome variable over time right you don't want there to be different relationships and so like it markets get more efficient or like there's more signal in lines like that's also going to cause problems that's again to be clear like that's true of any predictor but that's why you want to use like to me like more foundational stuff in the model so I would say like nine times out of 10 probably more I'm doing it after the fact for those reasons but I do think there's times and places like it's really like you know all that stupid trend content of like whatever 26 like the way you do good trend content like the way you actually find trends in my opinion is like taking the line and putting it in a model with another factor and seeing if there's signal there like that's how you actually do that and there you are putting it in like the model right to some extent so but beyond that I know I knew that you're going to that you're going to find figure out like where it would be worthwhile I knew it I love it wow alpha just leaked right there I didn't even think of that we always saved the good ones for the the exactly that's that's the only way we can really keep the lights on here speaking I mean you know all we have another model really a question this was about one one of your your articles from from your website sportsprojections.com I believe it is Lincoln sp twitter bio it's from sharps research so actually because you know the article he's referring to I'll let you read it because he he references an article in the tweet and it's hard to yeah yeah so I mean the just that question was I think the article he's referring to is I wrote some some article around like non linear correlation and how people can think about it and it was like geared towards like parlay type stuff and angles and how you how you can think about SGPs and stuff better hopefully I don't write often by the way you're selling out the the article that's site like there's big something like there's like four articles on their full tier or so or that I remember around they're really good I appreciate it so I did I did go look at look at it the article so that's what the article is about and so the question was like effectively why would someone want to use a linear model over something like spine or gbt for regression or classification than using pdp and ice plots aside from his first model like um I don't know I don't even know if this is pronounced I'm not I think this is like a fast task rapn okay I don't know it's pronounced rapid or rapn and the need for l2 loss or a tiny data set I'd love to know any possible use cases where someone wouldn't want to even use a linear model so like I think the last question is the gist of it because you know we're we're two hours and 10 minutes and I'm not going to get into splines and gbts and what not but the reason I put this one right after the question we just answered was I think it's the same thing is like in sports betting sport to me sports betting is not a data science problem at all like I said it before like sports betting is not like a Kaggle competition for anybody who's familiar with that that's like where you're trying to like make the best predictive model like I think a lot of people who come from that background think you know sports betting is like I got to optimize log loss or mean squared or briar score like that's what I'm trying to do to me that's not at all like what I'm trying to do when I'm sports betting what I'm trying to do is develop models tools ways of thinking that identify misprice lines and that sounds the same but it's not the same to me because like I would much rather have a model that is worse performing on every statistical measure you know whether it's mean squared error or briar whatever like I'd much rather have a worse performing model where I know what the issues are like where I know when to bet and when not to bet and I'm not saying like I never use stuff more complicated than like linear regression certainly some problems like I do think you need non-linear methods but every time you introduce every layer of complexity you add to this is my belief and maybe this is just because I'm not good at statistics but my belief is every layer you add to to a to make a model more complex you introduce the risk of overfitting and reduce the interpretability of the model and for me I don't need to price everything right like I'm not market make I mean I get sort of am but like these days but like really I'm just trying to make a couple bets right like relative to the whole what's offered right so I just need to be right in certain cases and I need to avoid being wrong like big and so overfitting is like a much bigger risk to me than underfitting because if it's underfit and I can interpret that like I just don't bet it right there's no downside when I'm overfit I'm betting stuff that's bad and that's that that's like the worst possible thing so to me I'm always going to err on the side of more bias basically less variance so like I will always because I don't want to be overfit so that's like the the bias variance tradeoff is like the summary of it so I won't get into like that specific models or whatnot that we throw now but they're all like more complicated versions of predictive models for anybody listening but like you just introduce more ability to overfit and more ability to fit noise and that's that's what will really kill you in sports betting not passing on bets yeah that that's what we said I'll probably just add um I guess motivationally like some of the absolute biggest crushers on the rigging side that I've met and talked to are doing like either just linear regression or maybe like logistic regression you know it's not necessarily um all gonna be like your gradient boost I mean we'll just we'll bucket those all as like AIML or whatever like it's actually very rare still in my opinion for the best of the best to be like solely using uh I guess what I would refer to as ML although then you get into linear regression is machine learning um but like I still think it's very very rare and even in big markets like it's probably better to make sure your data pipeline now having a good data pipeline everything clean making good features all of that still applies to linear regression and like making sure you're doing it in a smart way because you're gonna have small data problems so working with that and playing defensively like sp said I think I think is the key and you want to you just want to be able to know what your model is doing and I've struggled with that when I've dipped my toes into the the ML world is to figure out what it's doing I usually have to like like artificially change around like one parameter at a time and just like keep doing a bunch of different stuff and not that that isn't something that you should be doing or maybe I probably just socket machine learning but like like that's I had to sit there and just change all these different parameters because you have interaction features to it and it's just like just to get a sense of like what it was kind of thinking like it took forever whereas in uh in like a regression or more simple model like it just tells you and it's super helpful so even if even if it's not gonna you know be at the close if you know where it's a little bit weak and you know it gives you a good baseline like usually that's that makes a lot of money still in the right person's hands yeah right if there are if there are three screws in one nail and you have a hammer you can still find the nail right in hammer it as long as you know like yeah what you're looking for if you just have like some if you have a blindfold on and you have no idea what tool is in your hand like yeah you might hit the wrong thing you make them so I don't that's really good that sounds like the intro to like uh data science book like the first chapter you know I've just read maybe I've just read you know I've been exposed to so many of it maybe I didn't make no you understand for your book um for your book linear regression is still cool or something um yeah yeah again I think we've talked about this a bit but there there's certainly there's certainly the draw to do the cool new thing and I've kind of gone down more rabbit holes than I probably should have chasing uh what I've seen deemed like cooler more advanced machine learning techniques and I'll say like they most of the time they didn't really yield yield anything um could be a skill issue for sure yeah I was gonna say maybe we could suck at it so could be a skill issue um okay you want to do your Bayesian updates yeah I have a quick one I'll go quick um mine was I actually have a sports ability to take so I'm gonna be wanting to be a sports gambling sports gambling podcasts are telling NFL coaches how to do their job um so mine mine is I've I've actually always thought this and even more so today so I've always thought that the the depth of the holder on field goals should be variable and like more so than ever today the kickers can kick further and there's been like maybe as a recency yes there's been like it seems like there's been the most block kicks like of any season so like if you're kicking a 20 yarder that should never be blocked like just send the guy a little 10 yards back or whatever and you know people might be like oh but then the snapper like come on like these get their only job is to snap has one like right they he could figure out like a seven yard versus 10 yards so that was that was my take I like that take a look at my update on on what what what teams should be doing it shipper was on me once because I I just gave it take a non-natural update so that's what shipper that was a that's what he does he just gave it take instead of an update so my my I'm back on my music update so that's I basically have two themes one's college football the other's music so this one's music um I was kind of like okay with the state of music there's some artists I like but I realized like music today kind of sucks compared to music like 15 years ago and it's because like rock has completely exited the like mainstream music scene so I think the difference between now and let's say like 2005 is that in 2005 you had pop you had rap and you had rock and they're all like you had like the art of monkeys you had blank one any two you had these good rock fans that were like somewhat popular in public and then you had the offshoots that like more your singer songwriter is like a Jack Johnson a James Blunt a Ben Harper who were like acoustic you could call him acoustic rock acoustic pop whatever and it kind of had this nice blend and he blended over into pop and there's this nice mix now it just feels like it's just pop and basically like hip hop in some rap and it just feels like there's like this whole middle section of music that's now just kind of missing and I'm just frustrated that you know I think we grew up in a good time maybe this is how everybody sounds now I realize how people yeah I was that's it yeah but I think I think it took you this long to realize like this is officially like this is how you know you're getting old like no matter I've always told myself no matter how old I'm I get I'm never gonna say yeah this this this music is bad because that's just like what happens to me is like I like the rap music now and there's a lot of the pop music I like now so I'm not even saying the music now is bad I just feel like there is this vacuum left by like rock not being in the mainstream anymore so that's my business you know I would I would keep it that way did not be labeled as the uh the old timer just just just just narrow the take down I always I always try and make my take they take that job time a little bit that's my style well um yeah okay so this one you know we knew this was going to be a long one but thanks everybody for the the questions they were awesome um it actually changed the whole structure of the episode because of how good and how many questions there were so that's pretty cool and thanks everybody for listening we will see you all on the next episode