Andrew Mack: Creating Edge in Options Trading & Sports Betting | The Outlier Podcast

0:02 What's up everyone? Welcome back to the Allire podcast. I have my new friend or soon to be friend hopefully, Andrew Mack. I came into you actually through your recent work with you and Sinclair who I've had on the podcast a bunch of times. Really enjoy his work. So seeing the mix of what your angle is with markets, sports, betting, and then Euan's mix in that book was actually really really fun to read. Looking forward to chatting with you. Yeah, thanks for having me on, Eric. It's a pleasure to be here.

0:32 >> Beautiful. So, we were just talking a little bit about the kind of beginning prospects of trading, right? And how a lot of people get into trading. You came from a slightly different angle because you span two worlds, at least from our lens. Most of my audience will be, you know, stock market, options, equity, stuff like that. and you span two very different worlds with sports betting and then just general equity markets. So if you think about the two of them and you were asked to describe how are they the same, how are they the different, what do you come up with?

1:15 Well, that's a very um long- winded answer to probably answer the entire thing, but let's start with the basics. trading is betting. Um the foundation of all of these things is positive expectancy, right? and that element which necessarily turns to topics of edge, the Kelly criterion, how much should you be risking um to to maximize your your growth without ensuring your future ruin and then um risk management all that kind of stuff. There are other ways you can look at it too. You can look at it in terms of the types of places that you find an edge which would be you know what we could call distortions of various types.

2:04 Now obviously the distortions between sports and financial markets will be different but thematically they'll be very similar in terms of the kind of places that you might want to look. So there there's a tremendous amount of crossover. The biggest difference is um trading is you know it's a continuous betting environment right obviously you have certain pauses like the end of the trading day or whatever but throughout a trading day or throughout a trading week you know there or or a year or whatever time frame you want to look at it's more of a continuous bet whereas every sports bet for the most part has a you know deceited end at the end of the game or the end of the quarter or whatever And I guess the the strongest comparison would be live betting a sports game versus the market because that's where you're going to find that things are really really similar. Um other than that um the convexity element in financial markets is different too because we have fixed payoffs in betting on sports. So that's another sort of um small difference that ends up making a very large difference in terms of sizing and then you start talking about volatility.

3:25 You start talking about Greeks this and that. So I mean that's kind of a long- winded rambling answer that's all over the place, but I feel like I got most of the important points there. There's something you talked about a while ago with respect to like Bradley Terry models and using something to like a basic starting point but then trying to find a good market to create this edge that you're talking about whether it's looking for distortions and one of them you were talking about was attacking smaller markets like the Icelandic women's basketball which is >> hilarious be I was just in Iceland and it's actually one of my favorite places.

4:07 I love it there. But I did not see a lot of basketball, believe it or not. It wasn't uh what wasn't everywhere. So, I find that correlation actually fascinating because in markets it's almost similar to what you were talking about people that are trying to just bet on the NBA or trying to bet on the NFL, really big dense markets. So if you were to think about that specific correlary, where can retail traders in equities markets find the Icelandic women's basketball equivalent, a space that might not be quite as saturated and a place to cut their teeth?

4:51 Now, it's a really good point because everybody wants to be known as the person that's betting on the NFL spread or the person that's betting on the NBA spread because there's there's this I I would say it's fake, but there's this this element of like prestige around it like, "Oh, no. I'm not a I don't do any of that uh Mickey Mouse stuff, you know? I'm a I'm an NFL spread better." And it's like, you know, you're either making money or you're not, man, right?

5:15 like um so a lot of the the smaller sports get ignored uh because there's no prestige factor and again I I use the term loosely and the second thing is that the betters aren't that interested in these sports and so they think well I don't want to watch Icelandic women's basketball for example I don't even know where I could find that like what what ESPN channel would that be on um but they're missing the point which is that you shouldn't be betting on stuff that you necessarily want to you should be betting on stuff that you have an edge, right? Like that's the whole point. If you're trying to do this to make money, if you're just trying to entertain yourself, I mean, that's a totally different conversation. And so, you know, likewise in the markets, there are an awful lot of people um and unfortunately, it's it's sort of promoted this this way by unscrupulous individuals at times. But, you know, people get steered towards you ES futures. they get steered towards S&P zero DTE and and there's this idea that there's a prestige factor with index trading, right? And you know maybe um again hopefully you're trying to impress you know yourself, your accountant, your bankroll growth, right? Like that's those are the things that matter to you.

6:31 So so yeah, it's it it is a good idea to you know take a look off the beaten path a little bit. What is the equivalent of Icelandic women's basketball in financial markets? Now, there's there's a question. Um, the first thing that jumps to mind is there are a lot of uh opportunities in single stocks where they're they're not getting as many eyeballs. Um, the liquidity maybe isn't quite as deep and you're going to find some some ability to capitalize on on volatility distortions. Now it's not exactly a free lunch because usually when you have spreads and so the bid ask spread will be a bit larger. In fact in many of the cases probably 80% of the cases something that you find in some of these smaller stocks the bit as spread has been widened to the point that it actually takes away you know any positive expectation you have on the trade. And so 80% of the stuff that you find is it's like it's there, but you can't really it's encased in glass. You can't really get at it because the market makers have they're not dumb.

7:44 They've widened the spread because they understand there's some some increased risk here. But you know that little 20% edge, 20% of edge cases, uh you'll find something. And that by the way, that's actually true in sports betting as well. So for player props and things like that, it's very typical for the sports book to increase the spread like to increase the the total vig that's being paid. So they're not um the opportunity is also the challenge. Like the reason that you can model uh a player prop really well is because other people aren't putting as much time into it and the sports books certainly aren't as well. But what they do to counteract that is they say, "Okay, if we widen this vig out to 8% or something, right, then we we've got a lot of wiggle room here to sort of protect ourselves." And you know, there's a huge headwind for the better to overcome when when they're betting into these markets. So, but yeah, I mean, there's lots of things in single stocks. Um, especially when the index, by the way, isn't moving a lot. Like when people are getting frustrated with the lack of realized volatility in the S&P or in the Q or whatever. Um, not a bad idea to start looking at single stocks, segment them by uh sector, segment them by sub- sector, and sort of say, okay, what's, you know, what's moving here? What what looks like there there is some inflows into this stuff?

9:10 It's interesting because I think there's a saying that actually gets contorted a little bit which is you know liquidity is king liquidity is primary got to look for liquidity everything be liquid and the message always makes sense right you prefer more liquidity but I've always found an interesting juxiposition if you go on a place like SSRN you'll find a lot of market effects driven by information and discreetness and coverage and how many analysts happen to cover cover a stock and what kind of lead into momentum factor that can have and blah blah blah blah blah.

9:48 So if we think about this idea of liquidity and how it's very frequent where we're really trying to maximize liquidity, what are your thoughts on striking that balance though to your point where there are instances where if it's too liquid, it might be there but you can't get it but then or if it's too illquid it might be there but you can't get it but then if it's too liquid it just might not even be there. How do you go about like finding those opportunities?

10:22 >> Well, you rais you rais an important point because like you said there there's no there's no perfect solution. When things are really liquid, the spreads get really small, which is great, right? Something that Ewan has said a couple of times, right, is that reducing your costs has an infinite sharp ratio like impact on your trading. So obviously less that you can pay in terms of pure cost is good. So we should be trying to find things that are liquid enough that are the spreads that we're we're attempting to to beat aren't uh prohibitive. Uh but highly liquid stuff is highly efficient stuff too, right? and and so now you're looking for um edge where where you have the most money and the most eyeballs on it and and that presents a new challenge. And then on the other hand, right, things need to be liquid enough that you can actually trade them. I mean, there's lots of stuff that has edge that you you won't even get filled on because so so it's more like a hypothetical mirage edge, but um so like you said, you're trying to find that sweet spot. Um, I don't know if I have any like cut and dried rules or or like principles in terms of like this is how you find the the best mix. I mean, essentially what I would suggest is that whatever effect it is that you're trying to target, you're looking for what looks like positive expectancy edge and then conditional on there being edge. Then you're going to look into the logistics like can I actually get filled? is this stuff trading um things like that. I would I would run it in that sequence. Um but I don't I don't have any like uh like I said cut and dried rules on on the kind of a thing other than the importance of start with the edge and then filter down for what actually looks like you could trade it.

12:20 >> Now there are a lot of them that you know something would look good but you can't get filled. Right. >> Right. Yeah. Now, the concept of edge is always super funny because everybody talks about it. You know, you're supposed to have it. It's almost like a buzzword, right? Everybody just says, "Get an edge, then go make money." And obviously, the the game of trading is it's wrapped up in literally in that one word and trying to create it or maintain it. So if you were thinking about different skillled traders starting with beginners, what's a reasonable place to start looking for edge? And for those listening, the giant caveat here is for newer traders, this is under the premise that they have already guided their expectations. And probably what we're going to talk about is some sort of grindy slash not hyper sexy way of doubling your account every 6 months, right? Like we're talking about the actual skill of trading, which typically doesn't look quite like that. So all that aside, for a beginner, if you were to give some ideas on where they can start looking for edge, what kind of market effects or profit mechanisms might be worth their time in air quotes, what might be good places to start?

13:50 Well, what's worth your time is a slightly different question because that's sort of conditional on how large your bankroll is and and how much you need to get down on a bet or a trade to appreciably move the needle, right? Um, if you have $5 million, what you're going to be looking at is probably fairly different from, you know, if you have $50,000, right? like those are two totally >> I think different scenarios in terms of uh the what you need to actually move the needle but to get to your question uh more specifically to take one step backwards from edge before edge it's it's effects and you use that word it's the right word you're looking for market effects and so you'd start you'd start trying to build up your sort of mental repertoire of different market effects um effects like the VRP in options trading, effects like the equity risk premium, effects like momentum, um effects like mean reversion as a general principle, and then you can dig into, you know, are there subsections of the market that show better or worse results for these types of things. Is there a way to make some tweaks to the the concept of momentum to identify things a little bit better? Is there a way to roughly time the VRP so that we can get some sense of when it's better to be on a short ball trade and maybe when it's a bad idea to be on a short ball trade. And then that can lead to a little bit closer to the types of stuff that I do, which is the inverse of that, which is there a better or worse time to be on a long ball trade, right?

15:40 Because, you know, in that counterparty dynamic, when it's a bad day to be a short ball seller, it's a good day to be a long ball player. And so, you know, and a lot of that has to do with trying to understand the people that are in the market, your counterparties, the other operators, what they're trying to do, how they're trying to do it, and get some appreciation for how these things sort of interact with each other. And if you have a some idea of how that is all sort of all these different moving parts are coming together, it will give you ideas for where you might find effect.

16:16 And so that can lead to things like the timing effects like the uh you know OPEX week, end of the month effect, um all of these sorts of things. Like that's kind of how you start breaking these things down so that you can you can start creating lists of things to investigate. Um I actually think that you would start with kind of general effects and then you would uh take a look at the counterparties, the other people in the market and the kinds of things they're trying to do so you can understand the pinch points. you understand where these people might uh buy or sell in a panicked state at prices that they might not want to. And again, another way to sort of look at where the pinch points might be so that you can find these effects and then you can test them to see if they have edge and then proceeded on until you're putting money down on these things. If you were to pick one of those, whichever makes you happy, and you were to begin the process of testing for edge, what are the broad muscle movements? And a a quick shout out actually here to your guys's book, um, Retail Option Strikes, it it talks about this, but figured for just a a general overview, um, for those that are listening, what what might be a good way to start that process?

17:41 Well, that situation you'd find yourself in is you have some kind of an effect and it looks legit. At that point, like we talked about in the book, you can do some Monte Carlo simulation to try to uh get a sense of the effect. From there, you might think about doing a walk forward test. Um most most people would tell you that the right answer at this particular juncture is to do a back test, right? To get in there and do a back test. Um yeah, you can do a back test. Back test is back testing is fine as long as you understand that it's telling you what has has worked in the past. Um if your effect is a risk premia, that's a good idea because that's going to help you really quantify the edge. And since it is a risk premia, it's not something that is likely to be ironed out of the market. And so for that reason, you know, we can use a back test and glean a little bit more information from it. If you think that it's more of a distortion or an inefficiency, a back test is probably going to give you a false sense of confidence because just because something has worked previously doesn't mean it's going to work into the future. And you know, something that I've noticed recently is that the the market on close uh for zero DT options, it's kind of changed a little bit the last 6 months, right? Um it looks to me like market makers have a much better handle on the market on close now than they did a year ago. And and so for example, you could have back tested some kind of a market on close strategy where you're you're looking for cheap options to buy and you know getting like a a very very quick trade in at the end of the day. And if you were trying it right now, your results probably wouldn't be what they were a year ago. Um so that's why I kind of like the walk forward test idea better because it it gives you a better sense of at least how things are are roughly doing now. And you know, you can argue about uh including transaction costs and things like that, but um the le the level of of detail you want to do in something like that, but I think generally the walk forward test is a a better way to go.

20:02 The other thing I would add to that, it's something that I commented on on Twitter, is that you you also want to try and understand the environment that whatever this this effect you're trying to harvest works the best in. And so um this can be like really simple like almost dumb sounding things like is the market above the 200 day simple moving average like so if you're trying to uh be short volatility you're trying to sell straddles sell strangles iron condors whatever it's good to know like hey does this work better or worse when we're above the 200 day SMA does this work better or worse when we're in backdradation or contango and hopefully viewers ers know the answer to that one.

20:44 Um, does it work better or worse when the VIX is above or below, you know, its 200 uh day average? Does it work better or worse when the VIX is above or below 21 or 16 or 25, right? Because that can help you understand when it might be a bad idea to do the thing that you're thinking about doing when when they're and it can it can kind of give you some ideas about how to improve the strategy moving forward as well. So, I would I would kind of add that into the walk forward test is like it's good to track some of these environmental conditions just to see if anything moves the needle a little bit.

21:23 >> How do you think about striking a balance between seeing something, being relatively confident it's there versus continually testing over and over and over and over and over again? and essentially just being a [ __ ] and not putting a trade on. Like I think a lot of traders get stuck in that. So how do you think about validating something enough so that then you can move to production?

21:57 >> It's a good point. There is, you know, it's a very very sophisticated form of almost procrastination to to just continually be back testing and never actually doing anything. Um, you know, the is where back tests go to die, right? Like when you when you actually contact the order book, that's when you're going to find out what's real and what's not real. Uh and and like I said, while while there are things we can learn from back testing and walk forward tests and Monte Carlo sims that are very important, at some point, you know, you do have to fire on these things and collect the data and sort of see how well it's lining up with, you know, whatever you thought you had learned prior to to executing on your strategy.

22:44 Um it's it's really important. There are there are people that you see out there that are going around in circles uh testing either two kinds two categories broadly. Some people are testing a hundred different things and never actually doing anything and other people are retesting the same thing a h 100 different ways and not doing anything. And I mean at some point it is a hands-on activity, right? Like you do have to actually uh put the trades on and get the feedback from the market with you know the cost uh in there, the market impact or response in there all of that stuff. Very very important.

23:25 >> Did I answer that question or or did I miss a part >> partially? So then the the main part going forward then will be after they do some of their back testing forward testing money Carlos Sims how do you bucket the stuff in your observation that's like I have to see this in order for it to be viable so that you know that you can move forward with it versus having complete certainty. So the way I think about it personally is if I'm testing something, there are certain features that have to be there in order for this thing to be viable based on whatever the effect is that I'm trying to trade. And I'm trying to think about or I'm trying to explore how you think about finding that balance between data collection, coming up with what things really matter in order to move it versus we're just testing now for the sake of it.

24:17 >> So there are a couple of things. Um, I think people probably would expect to hear something about statistical significance, p values, um, maybe maybe some basian baze factor stuff like like things that are going to try and quote unquote prove to you that you've got something real. And I would suggest that you're not looking for proof, right? If if it's like complete, first of all, I I've got a book talking about why P values uh don't mean necessarily what you think they mean and and they're sort of give you a false sense of confidence.

24:55 You are looking for an effect that appears to be strong, which you can get hopefully from the Monte Carlos sim part of your process. So hopefully you've already got that at least. You're saying the effect looks like it's a thing. Now, we're just going to find out if we can actually make money from this thing. And um so with regards to the statistical significance, you are you do want some data. Hopefully that walk forward test is going to give you some data. Uh but then I'm going to start uh firing on it. What I want to see are things that are are telling me that the payoff ratio part of the Kelly criterion looks pretty solid. So, generally speaking, you know, I want a positive payoff ratio. Um, I'm prepared to accept a lower win rate for that with what I do. But what I don't like, like, personally for me, what I don't like to see is something that is reliant heavily on the win rate.

26:00 Uh, something that I carried with me from sports. Uh sports is a little bit different because the payoffs are fixed with options with especially with long options like directionally. Um you have the convexity element, right? And so hopefully you're able to actually capture some of that convexity. Your payoff ratio should be healthy on these trades and you're prepared to accept a lower win rate in order to to collect that. Um so generally that's what I'm looking for.

26:33 Um, it's possible to do all kinds of little tweaks and things that will actually truncate the convexity and take a positively skewed strategy and turn it into a negatively skewed strategy, for example, uh, with options. Um, so I like to try and focus on the payoff ratio and the stability of it. Um what I don't like to see is something like I said that that heavily depends on the win rate because win rates can be very environment dependent as uh short volatility players found out in March and April. And that's that's kind of the main thing that I'm I'm looking for.

27:14 There probably there are a couple other things too. Um, but that's I mean with what I do, which obviously I mean that's what I know the most about that that's the main thing that I'm looking for. >> Got it. Yeah. the the win rate is a personal pet peeve of mine as well, just for retail options traders specifically, because I I also think it becomes a mental disease where you just you get used to most of your trades winning and even if you look at the end of it and then it's not really that great of a result because there's so many individual samples of winning, it can still feel and mislead people pretty heavily. So, I I fully resonate with uh with that component. Now, if you run the strategy and let's say you test it, things look good, you start firing on it, it's live, there's this new inject of path and just what happens to happen.

28:17 One of the things I struggle sometimes to explain to people is when you move something and you start using it, those first iterations of whatever you get, it's very difficult to tell a lot from it. You obviously need to get something that's a little more representative of whatever the effect is that you're trying to trade against the strategy. How do you think about when you move something it's live and how do you monitor it based on maybe like first and second phase where first phase it's really low sample size. What do you look for in that window? And then let's say maybe if you continue to keep it live and then the sample size grows, what do you look for in that window for your feedback loop?

29:08 Well, you probably want to combine um the total sample size as it continues to grow as well as some kind of a rolling set of metrics that's sort of showing you almost like the trend of your results so that hopefully if there's something starting to diverge, you can see that before you go down with the ship. Um, and that's something that I picked up from sports when I figured out that uh a sports season is not uniform across the entire thing that that you can kind of break it into different segments based on how much data is available and >> uh whether we're past the trade deadline and there are things that >> that sort of change um what works the best depending on kind of what segment of the season you're in. And I figured that out by looking at my results and noticing that certain strategies did really well early season and then they started to diverge from what I would have expected u after the trade deadline, things like that. And so you want to be paying close attention to divergences between kind of like the rolling metrics of what's been happening over the last month versus um the last year, things like that.

30:24 And one of the the follow-on difficulties I I have with people sometimes just in conversations, this is specifically for those I would consider at like the late beginner, maybe you know early intermediate stage, whatever that means, but just general context is sometimes they'll move a strategy, they'll have a couple results and they'll say, "Oh man, I'm I'm two for two. This is this is great." or they'll say like, "I'm 0 for two. This is useless. This isn't working."

30:57 What's your response to like that mental framework? Cuz obviously it is laughable. >> Yeah. I mean, that it's insane to to to come to those conclusions based on, you know, two spins, right? Like you need help if that's the kind of the where you're at. You're you're you're holding the stick way too tight. Um, it's impossible to know anything from from two spins of the roulette wheel or or two hands at the poker table. Like, there's just no information in that at all. Um, and what you get from that, you see this a lot, you see the strategy hopping, right? Where if something gets hot for, you know, three or four or five trades, then, oh, I'm a genius and I better I should I should start betting 40% of my bankroll on this, right? And and if something loses three times in a row, it's like, you know, it's the JPM collar trade out to get me or it's the or it's the market makers or it's uh and it's it's just a um overweing small uh observations way too heavily. Um impossible to know anything from that, right? And that hopefully that's why you've done the work of of trying to find some kind of an effect and taken a a really good look at it before you did this so that you know, okay, we're going to have to do this a bunch of times, right? Hundreds of trades in a in a trading career, thousands of trades, tens of thousands of trades depending on your time frame.

32:29 Uh so so two trades really, you know, shouldn't make or break you either way and and shouldn't affect you psychologically or emotionally either. If if it if it is, you know, uh maybe you don't understand the effect well enough and it feels like magic. Maybe you're betting way too much and that's why you're overreacting to every single thing that you're seeing. Like you're probably right on the train tracks for risk of ruin territory if if that type of thing is happening to you while you're doing this stuff. Uh because one one one two three four losses in a row uh four wins in a row, it shouldn't affect you uh too much, right? It really shouldn't.

33:10 It's and there there's really two follow-up directions I want to go. The the first one though is to talk a little bit about the psychological component of things because we were talking about this in the beginning and I think a lot about the beginner trader because I like I remember being there and I remember like all of the things that you're hoping to do and most of them are just just complete pipe dreams. But you mention an important thing where as you're building this strategy, you need to make sure that you're able to give it enough runway to see like is this thing viable? And that's something that I've developed a process around. But I'm curious from your lens as you're in the scoping stage of a strategy before you choose to take it live. Do you have certain metrics already built in your mind where it's my average should fall within these levels?

34:13 I would need this many of observations in order to accurately feed this back into the system and derive information from it. How do you go about that? Yeah, you definitely want to have some idea of the um less exciting metrics like your you know your average draw down or your maximum draw down. These are things that you're not going to know with certainty before you start, but you can you can simulate win rate and wi with some with some variance around it and you have an estimated payoff ratio with some variance around it. You can simulate that and you can simulate what the draw downs might look like. Um, you can sort those into uh quantiles. You can do all kinds of things to get an idea of like if this goes bad, how bad could it go?

35:09 um and and what are kind of the the rough guide rails in terms of where this thing should perform even with some randomness mixed in. Um definitely you want to have some idea of that and I know that um you know astute viewers might notice that referencing win rate and payoff ratio in this way um would be the wrong implementation of the Kelly criterion because the Kelly criterion in financial markets normally we're using mean return and variance so it actually looks closer to a sharp ratio but just conceptually right like this is one of the ways that you can give yourself an idea of like the kind of draw down uh or the kind of um turbulence you might expect, right? And and then also the the type of strategy you have. If it's a high win rate, low payoff ratio strategy, that's going to have different draw down qualities than a low win rate high payoff ratio strategy. So all of those things would kind of go into consideration, but you should be aware of them. Not because you're going to be able to calculate them uh with any certainty, but because you want some idea of what you might experience along the way so that as you're you're getting the live results, you're getting the feedback from the market, you don't get freaked out when you have a bad week or you have a bad month or whatever. Okay.

36:35 >> Now, when you're implementing something like the Kelly criterion, I assume you're using some sort of fractional variation of it. How do you implement it with a broad portfolio though that might be running x number of strategies? So maybe the easiest way to to answer this is if if you could give us an idea of like what a standard portfolio might look like. And again, you don't need specifics, but I'm thinking like number of strategies, number of trades, and then that'll help map how you use the Kelly criterion um in that mix.

37:17 >> Well, you and I talked about this in our book. Um there's a a small section on this about uh correlation between strateies and how that that ties into the mix with using the Kelly criterion. Um, we also talk about fractional uh, Kelly as well because it's a very bad idea to use full Kelly generally speaking because it's always an estimated number and we don't know what it will be moving forward most of the time, which means that you run the risk of uh, betting over Kelly, which people familiar with the math knows that that's the path to blowing up. And so the the idea is to have some estimate of the Kelly criterion number but to always be you know a healthy amount below the full amount. Um so you end up with quarter Kelly um or or something around that.

38:16 In terms of how many strategies running um I have a number of strategies but most the reason that I have a number of strategies is because the environment is changing and you know strategies don't work all the time right and so there there is a time when it makes a lot of sense to be long a straddle but most of the time being long a straddle is a losing bet and so you've got a long volatility long straddle uh strategy in your your you know your bag of tricks, your tool, but it's not the thing that you're going to be using all the time when when we're in a lowvall environment. Um you probably want to be able to participate in the market generally going up the equity risk premium. So, you know, you'll have a strategy for that. That could be momentum, that can be uh looking for breakouts in single stocks, it could be a number of things. And so, um, it's good to have a number of different strategies, but I guess what I would recommend is that try to try to find things that are complimentary like this so that you are versatile and and adaptive so that when the market environment changes and and the environment we're talking about can be the volatility environment, it can be the general, you know, equities market. uh you know, you you can look at that, slice it and dice it a number of different ways, but the idea is to have complimentary strategies that allow you to continue to find trades with positive expectancy regardless of where we're at in the market. Um so it's not really about running them all together for me at least. It's it's about recognizing the environment you're in and then using the right tool for the job. And is this part of the reason why you like to stay both in equity markets and then also sports betting? One of my curiosities is what what prevents you from like highly specializing in one or the other.

40:27 Um, well, I I told a little bit about this story on another podcast recently, but I put a lot of time into the trading side of things when the sports markets were going quite squirly because of the COVID 19 pandemic environment because all the the players were getting pulled into COVID isolation protocol like five minutes before game time and it was really messing things up for some of my playerbased stuff. Um, at this at this point it um it would probably make sense to just specialize rather exclusively on financial markets.

41:06 Um, but I've got all the infrastructure built, all the models, all the databases for sports, so it's not really taking a ton of time anymore for me to do that stuff. And so, um, the the compromise that I've made is that I I'm only betting on regular season now and I don't bet on the playoffs and I take the summers off and I don't bet baseball anymore. And that's kind of how I've slowly put together some semblance of work life balance. Um, but the liquidity, like how much you can get down in the financial markets is obviously orders of magnitude larger than sports. And so it makes the most sense for me to to keep putting the majority of my focus on that. Um, and in truth, I've thought about cutting a few sports from the portfolio. I've thought about cutting basketball and hockey just because the amounts that you can get down are not are not significant to me anymore. Um, so if I was going to do anything in terms of changes, I would probably stick to betting NFL and and trading.

42:11 um something I'm considering over the summer. >> Interesting. Yeah, I was thinking about it and like I could see it a bunch of different ways. Like there could be just you know an interest element as well. Obviously there there I imagine there's a subcomponent of that. But I also find at least for me the market is plenty. It's it's it's actually a lot in terms of different ways to get involved and constantly tweaking. So it makes sense on how you strike a balance there cuz that's I was sitting here thinking like man that would be really challenging to trade at a high level and then also be you know sports batting at a very high level but you also have a lot of experience there when >> mostly it's it's just the benefit of the you know the systems that I have right it doesn't take a ton to you know to refresh the data run the models and uh give me things to look for where it started to become a little bit of a grind is that you know with trading you know if you're not trading um if you're trading uh you know main stuff right you a Friday you're done you've got you know a weekend and then you go back to it with sports betting I mean it's it's around it's around the clock right in terms of like every single day there's going to be games on and so I think that uh that football still fits nicely in because there's there's good liquid that you can get down on on spreads and uh things like that. And um but also, you know, you've got Monday, Thursday, and Sunday. And so it it's a little bit more structured that way. So, like I said, it's something I've been thinking about is maybe just uh maybe just retiring the hockey stuff and the basketball stuff.

44:00 I guess we'll see. >> Yeah. when you're looking at these two different worlds, I'm very curious. It was funny before you were talking about some of the evolutions you see in like zero DTE, which I I trade those pretty regularly among a slew of other stuff, but how do you assess the rate of environmental and structural change between sports betting and equity markets?

44:38 the rate of change >> structural or yeah so I'm thinking for example what you're talking about there with the market makers and zero DTE at the end of the day getting a little more privy and tightening up the way that they're running their end of the day activities. Do you see that same kind of reactive component in sports betting where you're constantly seeing these small changes happening in the spreads that are provided or in the way that you're analyzing that stuff or is it kind of set and it doesn't evolve quite as quickly? I don't have really a deep sports betting background as I'm sure you could tell.

45:16 >> Um I wouldn't say that it's set. I mean one thing about sports is that you get a lot of uh rule changes from year to year >> interesting across the various leagues and usually those rule changes make they make some material uh changes to the total to the scoring environment in some way right I mean an old example would be you know when the NHL changed the size of the goalies pads uh regulations uh or when they you know that they changed the rules about two-line passes which made breakaway passes more viable. And then you've got uh offensive and defensive pass interference in the NFL. Then you've got the change in the uh shot clock in the NBA when we went from uh to a smaller like a shortened shot clock which which materially changed the total the average total on a basketball game. Um, then there was one, I think Barry Horus a few years ago mentioned something about the uh the base sizes in baseball uh and people were were reacting to that. So there are these there all kinds of uh and then of course you've got the the uh the baseballs themselves, right, which there's been a bunch of um interesting analysis analysis done into uh you know changing the manufacturing process of the baseballs and changing the scoring environment and >> man that goes deep >> all these kind of things. So you have things like that that are changing all the time.

46:43 >> Um and so it's important to react to that. that it's important to, you know, to be kind of have your eyes open and constantly be on the lookout for these things that might move the needle in some way. Um, I guess the other thing is that there are no risk premium in sports, right? It's it's all inefficiencies, distortions, or else you're you're losing money. And so so that means that when these things uh become part of the market, they they people are reacting to them uh people that are doing a good job at this are reacting to them pretty quickly. Um I don't know if it's faster or slower than financial markets.

47:28 It would be hard to say. I would I would think that it would be slower just because there's more money involved, more >> more eyeballs on the financial markets, but um you know, you you see a lot of um so I guess that would be probably the safest default answer, but um you know, you see the markets react pretty strongly and pretty quickly to things in sports though too. So >> yeah, it's just an interesting general curiosity trying to understand more of like the difference between the worlds.

47:59 It leads to a interesting market example. When the Trump tariffs came out, what did you think about the markets general volatility leading into that event and then out of that event? Well, I had um I had two of the best months of my trading career during that time. Um so I thought the volatility was wonderful but um what I recall more acutely about that time was that there seemed to be a very strong confidence and I would suggest overconfidence that um that the presidency was going to be like sort of an upon only presidency that the stock market was going to absolutely rip.

48:56 And there were a much there was a more minority voice that was sort of quietly whispering that, you know, some of the papers that were being circulated and things about how they were prepared to let equities fall a little bit and maybe that would do something to the rates they would pay on government debt and things like that. And so to me it looked like the fragility was to the downside. It looked like the fragility was exposure to long volatility.

49:24 Um, now that alone isn't a reason to take a trade, but that's kind of when I start looking for for other signs, looking for other uh things so that I could get positioned to take advantage of that. And and I got a little bit lucky because we did get we did get the volatility. And so, um, to me, it's much more interesting to look at when one side of the boat is overloaded, you know, when it when there's an offbalance in terms of confidence that this thing can only go up or it can only go down or this is a shortall environment and you start seeing uh people crowd into the shortall trade. Um it's really useful to try to think about where the potential fragility is in what's happening just not because it increases the chance of the you know that thing occurring but because that that's where the big payoff value is if you're ready for it. Um that's that's sort of my general take on that stuff. So >> yeah very very wild time. Um and and interestingly enough uh we saw a lot of disbelief when the first uh when we started seeing a first big realized vault to the downside. You saw a lot of initial disbelief and which was you know ended up being a good sign because we got some really explosive days.

50:51 >> How were you generally positioned when you were trading it? Uh I was long puts >> and then were you just trading them in the primary indices? Do you focus in one or do you spread between multiple indices? Did you think small caps were going to be more susceptible or uh when when I generally think there's a long ball opportunity, I'm looking at the indices. So, I'm looking at the Q's. I'm looking at SPY. I'm looking at SP X index ES. Um, that's kind of where where my attention is focused when I think that we're we're, you know, shaping up for a really nice long volatility environment. And then, of course, you're you're looking at the the the VIX complex too. You're looking at uh the term structure for the for VIX futures.

51:45 You're looking at the VIX. You're looking at VIX. um all things that can help sort of give an indication um if that fragility that you think might be on the horizon uh could actually be there, right? Like where are we in terms of that complex as well? And if you're looking at that, then it's handy to, you know, be targeting the indices, right? >> And how do you choose the actual structure for your trade? Do you try to spread across expirations to have some nearfront convexity and gamma and have some longerterm stuff in case it takes a little longer or are you like primarily trading it around the catalyst itself of the tariff announcements? How do you think about that?

52:28 >> Well, yeah, a little bit probably a little bit more lean towards the catalysts themselves, but I also like to pay attention to where we're getting the volatility from. So I'm looking at the intraday overnight ratios and and I'm trying to see like is is this uh being pushed like after hours like you know or is it intraday like where are we getting the vault and are we seeing any changes in that ratio because sometimes sometimes a uh big moves get dampened in either direction because most of the move is happening outside of the you know regular trading hours. And so it's it's really good to see where is that realized volatility coming from in terms of the intraday overnight ratios. Um and depending on what you see you might be trying to position yourself for you know a long straddle that you hold overnight for example uh which normally wouldn't necessarily be the best idea in the world but there are situations and we did see some during that time frame when it was a great thing. Um, so yeah, generally speaking, pretty shortterm time frames like, you know, intraday catalyst stuff, some overnight stuff, weekly stuff, things like that. Uh, I'm not usually looking uh too much further out for my main strategies longer than about a week.

53:56 Um there are some, you know, slower moving momentum stuff where I'm holding things for longer than that, but uh but the bulk of my focus is on, I guess, what would be be considered pretty shortterm time frames. >> Yeah. Yeah, it makes sense, especially with the the type of positions that you're talking about. A couple other questions just out of sheer curiosity. Um, when you're looking at something like the Trump tariffs, there's obviously a breakdown in something like the efficient market hypothesis. We already know that the strong form doesn't exist, but we do know that markets are fairly efficient.

54:42 They're not like, you know, obnoxiously inefficient. Yet, it's really interesting to find instances like the initial Trump tariff announcements where it does seem like the entire market almost got caught completely offguard. What do you make of those kind of instances when there's clearly a well whether you want to call it an efficient inefficiency or not but there's clearly a a big discrepancy between positioning and what information is known and then what happens >> yeah I don't know if I would I would call tariff announcements an inefficiency I would probably call that a catalyst right because the truth is that >> [ __ ] happens right I you know, a war breaks out or you know, a tsunami takes out a, you know, a supply chain that's critical to the world in some, you know, like these things happen. And while I'm sure that there are ways to try to get a handle on that um you know with forecasting and this and that there are lots of catalysts around the world um political environmental whatever that they they just pop up and we get we get um unappreciated volatility unexpected volatility.

55:58 So I don't I don't know if I would call that you know an inefficiency per se. Um, >> just quick >> what's interesting about those moments to me. >> Oh yeah. Okay. >> Yeah. No, I just had a quick question on that because I I agree with you in terms of like the catalyst, but the I guess the specific component I'm really curious about is that catalyst is fairly unique in that it was well known, right?

56:25 Like we knew when they were being announced, we knew the date. We had some pre uh prelude into like what might be announced. So to me it's like it surprised me to see the market positioned as it was coming into that kind of event obviously mispricing it. So that just to be clear like that's the specific part of that curve that I'm so like I'm curious on your take of like how the market positioned coming into that.

56:57 Well, like I said, I think uh I think what was really interesting about that was the overconfidence about what that was likely to mean and and that this this administration was going to be up only or or whatever. I mean there was a lot of confidence about what what this was going to mean extrapolated into the future and it produced it that confidence produced some very uh unbalanced positioning in one in one general direction and there's a lot of permutations in terms of how it was traded but it created potential fragility to the downside for realized bonds and it didn't increase the probability of getting downside vault because you know who knows what could have happened in alternative scenarios.

57:44 But I think what was interesting about it was just that the effect ended up being a lot stronger than it otherwise would. Not um not because of of an inefficiency, but just because the the positioning was so unbalanced. And so that's kind of what I was trying to say before about like looking at where the uh the fragility is because that's where the biggest potential payoff is when when a surprise happens. And it's not that you think you have a high probability of the surprise.

58:17 It's that the payoff for this surprise is starting to be get into the territory where uh this is becoming positive expectancy to bet on this even though it has a low chance of actually occurring. Um, and I think that's what you saw there. Just just a lot of of very confident uh positioning that that left left an opening. And when Vault picks up and when VIX picks up and when Vivix picks up and the term structure starts to lift, um, it has a lot of cascading effects in the market. And I'm sure that there are levels of detail to that that you know I don't even fully understand at this point but um there is a a very large snowball effect especially to realize all to the downside and and what we saw was uh was sort of that that all coming to fruition at once.

59:09 >> Yeah. Yeah. Interesting. Well, awesome. I had uh one last question slash just general topic for you. If you think about your time frame of trading in general, combining, let's say, both the sports betting and then the the market side of stuff, how has your overall approach and process evolved in general? You don't have to be super hyper specific, but curious, what are like the the main things you noticed as like key changes to what you do over time?

59:44 Uh well, with sports, the the first real beach head that I had was uh with player props and modeling player props. And so I got really into the modeling element and then started constructing, you know, full games out of uh estimated minutes played and rosters and lineups and things like that. As I moved to sort of the next levels, I realized that there were stronger effects, but you had to completely revamp the approach so that it was no longer about taking the players, combining them, you know, adjusting and accounting for uh co-variance and correlation between the players based on the the given lineups. There was there was a number of completely different approaches that were more market-based approaches that um that were really really useful and a lot of those market-based approaches were things that I was able to take into the finance world and use effectively.

60:53 Obviously, we needed to tweak them a little bit uh because it's not exactly the same, but there were many things from the market um the market strategies, the market models in sports that I was able to transpose, you might say, into the financial markets. And and what's interesting is that in options trading, I found a number of things that are actually quite useful for sports betting. Um, looking back on it now, um, >> I don't want to get into that too too much, but the the amount of crossover that I appreciate now, uh, between sports and options trading specifically is enormous. Um, probably more than um, I ever thought I would see.

61:37 And I guess that um I mean my approach really is like a sports betting based approach to trading which is kind of hard to explain and something that I'd prefer not to explain because it's working really well but I guess that my approach changing over time probably more market rather than playerbased in sports and then just appreciating the amount of transferability and using everything that I learned in sports in the financial markets to the maximum that I can. Um, you know, because I I didn't totally see that connection when I first got into to trading. And the the longer that I've been in it, the more I appreciate how similar they are and how useful the tools are in the opposite world, you might say.

62:31 >> Last question. How would you characterize the concept of implied vault between sports betting and markets? Obviously, um not a direct corlary, but imagine there's some artifact. Yeah. So, um, I actually had a tweet about this recently, but if you, if you take the NFL, uh, take any game and look at the point spread, there's a main a main line for that point spread, and you could think of that like the at the money strike.

63:04 >> Mhm. >> Now, every time that I bring this up, there's somebody that replies, but they're not exactly the same because options are have convexity and sports have a a fixed payoff. And it's like, yeah, we get that, right? Like the so sports, if you look at, yeah, if you look at a sports >> uh spread market, it's more like a binary option because the payouts are binary, right? Okay. But the concept of the strikes is the important piece. So if you look at the main game spread line, that is the at the money strike.

63:37 And then as you go further out either way, you have alternate lines. You could buy points to reduce the spread for a favorite. you could sell points to increase the spread for a favorite. And it changes the payout that you would get uh in the betting market. And that's just like the strikes that you see or at least very similar to the strikes that you would see on an options chain where you know you're going deeper into the money or deeper out of the money. Same concept. And you'll notice that when you look at the implied volatility service, that the implied volatility, which we can consider a proxy for the price that you're paying, the V that you're paying for these strikes, it increases as you get further away from the at the money strike. Um, and you know, there's a smile, sometimes it's a smirk, sometimes it's relatively flat uh on on one side, but that's the general concept that you're going to pay a higher ball as you get further away from the at the money strike. Well, in sports betting, if you look at you don't have an implied volatility, but if you look at the total vig for each uh line, for each strike, what you're going to see is that if you get further away from the the main game line, you pay a higher uh total vig as you get further away from the at the money strike. And so, it's the same concept, right? The sports book is saying if you want this, you know, far in the money or far out of the money spread on this game, we're going to make the, you know, the implied volatility, the the implied volatility or the vig in the sports book case larger to account for either higher probability of payoff or more uncertainty, etc. And so if you look at it that way, you realize that market maker and a sports book are doing very similar things. And you can find a lot of utility in thinking about it that way, even though they're not exactly the same.

65:39 >> That's a I mean, it's a great way to look at it. The the thing that really made me think about that anyways was when you were talking about during COVID, right? like 5 minutes before the game a player's getting pulled for something else and it's like damn like that that's a that's a big inject of all that is very difficult to to model. I mean obviously you can build one over time for something like that. But I thought the relationship there is really really interesting. But thanks for uh thank you for explaining a little bit more about it because it does help. Um Andrew had you for a little while and really really enjoy chatting with you. I would love to have you on another time.

66:17 I know it's Friday. I'm sure you got stuff to do as much as I can. I could go on forever cuz this stuff is so freaking fascinating to me, but I realize people also have lives to live. So, I'd love to, you know, keep in touch and have you on another time. But in the meantime, folks, they can find you on X. I can throw a link to your uh profile below. anywhere else that you'd like to send people to check your stuff out?

66:46 >> I think that's it. I mean, I've got a link in the bio for uh the Amazon books and uh various YouTube videos and uh Spotify, uh podcast, all that stuff. You can find it all there. So, that's probably the best way. >> Beautiful. So, everybody, you know where to go look to find more about Andrew. Thanks for hanging out and then we'll see you guys on the next episode.