Hello, and welcome to the Double Pivot, the world's most agreeable soccer analytics podcast. Astute listeners will notice that I am not Michael Kelly, I am Michael Goodman. Kelly is on assignment today. We are recording this on a day that has been somewhat momentous for New York City politics, and he is quite busy in Brooklyn.
You will hear this at some point shortly after that, and I am happy to say, I am not alone. We are doing an interesting podcast today with an interview, which is only kind of about soccer. So, without further ado, let me introduce Jonathan Jaffe, who is a poker coach and podcast host at GTO Lab, and also an incredibly accomplished high-stakes poker player in his own right. Jonathan, thank you for joining us.
Yeah, thanks, Mike, looking forward to this. Yeah, so I think I want to start with two things. One is I want to explain to all of you out there why we are doing this podcast, and then I will literally have Jonathan explain what the letters GTO mean. Listeners know that like many moons and several careers ago, I played poker seriously and professionally.
And then I basically stopped for 10 years, maybe more. And it's been recently that I started paying attention to it again. Obviously, I think during COVID, it sort of became popular in a way, moved back somewhat into the zeitgeist. It's become much more noticeable that poker is sort of a thing.
And I started paying attention, learning, playing after literally having not thought about it for 10 years. And there is some stuff that is extremely different. And I mean, I think that that's what GTO Lab is a big part of. But it struck me while doing this that the experience would not have been altogether different from if you like went into a coma 10 years ago, woke up, turned on a soccer match, and everybody was talking about expected goals.
And so what I wanted to do was talk with Jonathan a little bit about that, because I know, Jonathan, your poker career spans this entire era, right? So when you started, it was before things like solvers and all this stuff we're going to talk about existed. And now you are in a post-solver world. Is that a good way to sort of think about it?
Yeah, very accurate. When I got into the game, talent was a lot more important, or let's say innate talent. And today, hard work, being studious is probably a lot more important. So what is GTO?
Like, what is it? Yeah, that's a great question, because even though the definition is readily available as game theory optimal, and yeah, you know, game theory is this branch of mathematics that John von Neumann created. I could be wrong about that. Right, I believe it.
I think most of our listenership will at least in theory be aware of what game theory is, even if they don't necessarily know details or, say, practical application as we're going to talk about here a little bit. And that's interesting, because, yeah, I think when I got started in poker, you know, a little over 20 years ago, that may not have been the case so much. But it seems like the world is kind of moving in that direction. I think about money ball as a big shifting point when Michael Lewis kind of popularized the use of statistics and kind of hunting down value from this kind of game theory lens that, you know, Billy Bean used for the Oakland athletics.
And then when they kind of dredged up even deeper, Michael Lewis ends up writing a later book saying, hey, without realizing it, I kind of copied Amos Tversky and Daniel Kahneman, this economist psychologist pair from Israel that, you know, I want to say in the 70s. That's Thinking Fast and Slow, right? Kahneman writes Thinking Fast and Slow, right? Yes, that's Kahneman's work on his own after Tversky's death.
And Daniel won the Nobel Prize for that, I want to say, the work on system one and system two thinking, which I'd highly encourage anyone to check that out, very cool stuff. But yeah, there's a lot to kind of thinking probabilistically in poker. And when you think about why there are thousands of professional poker players, and, you know, I don't know the statistics on this, but probably only a few dozen players making a real living at chess. And, you know, if you compare the income, let's say, of the 500th best chess player in the world to the 500th best poker player, it's very much in the poker player's favor.
And I don't think it has anything, certainly not to do with intellect. I bet IQ-wise, the chess players are crushing the poker players. But it's the fact that, by and large, the world does not think probabilistically. So these illusions and mirages emerge in poker where a player can think that they're a lot better than they are and or can win on any given day, which is enticing for a recreational to step in and just say, hey, you know what?
I've got a legitimate chance to win this tournament where if I step on the pitch, like I can't even be a contributor in, you know, a college level soccer game. And there's no way I'm just going to luck my way into anything but in poker you can. And so that kind of intoxicating feeling really makes our world much wider. And yeah, recently, this game theory optimal understanding is kind of trickling out a little bit more and bleeding into sports for sure and definitely finance as well.
So in the last decade or so, there have been, there are now answers, right? To like what is game theory optimal when it comes to playing poker. I mean, I think poker, like a number of other games, has been increasingly... I mean, I know the term is solvers.
I don't know if solved is the right word, but there are like definitive answers to questions of what is the right decision to make in various circumstances. Yeah, I'm a believer that that extends to literally everything in life, that there's a true perfect answer, but for most areas, the world is too messy that we will never get to that answer, so it's basically irrelevant. But essentially, I think in sports, we're constantly both exploiting out of purposefulness because we know something about our defender and what they, you know, might habitually be doing. And then we're also not really exploiting, but we are working within our own weaknesses, be it, you know, our physical limitations or, you know, we only have a few seconds to think on our feet, and this is the best, you know, solution we can come up with with the current problem.
Yeah, so the soccer side of things, sort of over the last 10 years, what we've done is developed expected goals, right? And this is... Expected goals is a measure of the expected value of every shot that gets taken. It is not precise, but it is accurate.
So on any given shot, right, when you take all of the factors into account, you're getting the average value of that shot. And at a high level, at an aggregate level with the data, that does a better job of predicting player performance, of predicting team performance, than goal subs, right? So if you know a team's expected goals for and expected goals against, you'll do a better job of predicting their future goals than if you just know their goal score, right? That's sort of the...
That's the pillar of what's happened in the last 10 years. And the reason it's happened, in large part, and the reason it's proliferated, is that... Nah, I guess I'm old. We're creeping towards 20 years now.
Measuring data, measuring soccer games, has become better, right? So literally, there are companies now that do a... There's a lot of manual work of logging everything that happens on a soccer field, and then you code that into event data, and then you use the event data to build a model that shows you the expected value of each shot. And it's great.
I mean, it gives you tons of information you didn't have before. Markets are tighter than they have ever been on the gambling side because this information gets incorporated. I mean, I know that both on the betting syndicate side and on the house side, like there are groups that do their own data logging and build their own models. And like, you know, there's an arms race about the precision of the data, that kind of stuff.
But the thing that always sort of like, the parallel that I always sort of feel, right, is on any given shot, you have an XG value. But that value is an average. And so that specific shot could be higher or lower expected value than what the average is. And we don't, we're not able really to quantify that.
We just know that in aggregate it works. And it seems to me that like game theory, game theoretic optimal poker is similar about decision-making points where it's like, you can look at and learn and study what the best decision is in a given spot in, you know, solver world where everybody's a solver. And then you have to move away from that in sort of like messy real world when you're playing with other people. And, you know, it seems like in both cases, what we're doing is like starting the conversation from a better place, like starting the conversation about what we should be thinking about, what's going right, what's going wrong from a better baseline than we could have 10 years ago.
I like that as long as we qualify better is for more accurate because there is a counter-argument to be said that in some sense, when we see everything through an expected value lens, you know, like the expected goals, it starts to tear down a lot of the storylines that we all grew up with where we like to say, oh Actually an open mathematical question that has been resolved in both directions by various papers. it has depended on what subset of penalties you look at for the data, what conclusion you come to. So it's, yeah, it is not an easy answer when you start talking about these things. And that's before you get to the point that like in soccer, now, currently, what we have expected goals, which gives you this really nice EV sort of value to start with when you talk about these things.
There are problems with... problems is the wrong word. There are challenges with data collection and how you like... I think maybe this is the best way to put it.
We have a really hard time in soccer figuring out which of the stuff that happens on the field matters to creating good shots versus bad shots. We can look at the shots and tell you, hey, this shot was good, this shot was bad, based on positioning, based on the type of body part you're taking it with, based on defensive... all this stuff. But we have a very hard time looking at all the stuff, all the passing that happens in midfield, all of the ball movement, all of the interchanging, and being able to provably show which stuff matters to creating shots.
And like, you know, when I look at GTO stuff, when I look at solver stuff, and I see, you know, that they are very precise about the EV of various hands preflop. But like, the EV of those hands depends heavily on how you play them on later decision points and later streets down the line. It seems to me, especially if you were just learning the game, to be a very similar sort of thing, where you can look at, like, an endpoint, but if you're having trouble linking it up with the process to get there, there's, like, a lot of room for challenge and interpretation and discussion and debate. Yeah.
In my opinion, there's a bit of a psychological phenomena at play that underpins a lot of this, which is to say that people who think probabilistically, by and large, have a little bit of a chip on their shoulder because it's like we've been living in a world run by people who, in some cases, we just kind of want to... I'll just be blunt. We're just like, they're all morons. What are they?
How are they thinking like this? This is ridiculous. This is insane. It's kind of like, I don't know, I'm agnostic.
So whenever a presidential debate comes up and they all have to basically swear their allegiance to God, and we know that most of them aren't even religious, I'm just like, very frustrated by that. It feels ridiculous. And that's kind of how sports and announcing and all that has worked for so long. But that chip that we all kind of carry, it leads to a very strong bias, in my opinion, which is there's this bias for answers and people who think in terms of stats in this manner.
They're so excited as soon as a new statistic comes up that seems to capture so much that oftentimes there is a flaw in the data collection. And there's a very missing component to all this. And we can go eight years holding on to a stat with, you know, a biblical fervor and then find out, oh, it was a really flawed stat. That really did not capture things.
And maybe being a dunce and just kind of watching with your eyes was even better than that stat and what it captured because it was so flawed. My favorite example is like pitch framing, right? Like for years in baseball, like people like decided that catchers' defensive contributions to the game were not important. And you just stick a guy back there that could hit.
And then it turned out after a decade that like the way a catcher was framing pitches from the pitcher really did matter to like a significant degree. If you were paying attention to this stuff. Can you imagine some of that, you know, if we go back on a bunch of the Hall of Famers who had good and bad pitch framers, how different their statistics would look with different matchups. I mean, a lot of these guys back in the day, they played for one team.
They might've, you know, pitched 80% of their innings to one catcher and he might've been, you know, a hell of a pitch framer. You know, curve balls just above the dirt are getting called for strikes. And Bert Blylevens, you know, borderline Hall of Famer. Who knows?
Um, and like, we don't have some of like, I think a lot of people who work in soccer analytics are sort of like, we're sort of like blooded in the wars of baseball analytics, right? Like we came up during the analytics revolution in baseball. A lot of us are American, which makes for an added sort of culture clash. When you're talking about like people getting hired to teams and trying to implement this stuff.
And like, how do you do it smoothly? Like, I don't know, 20 years ago, people who like did very basic math in poker were sort of derided in a similar way by people who had like, come up in an age when like, you know, dealing out 300 hands to yourself was how you figured out math, right? It's kind of a nerd jock division as well. That has kind of, you know, there's a coolness to, you know, just doing things with your eyes and by feel.
And there's, there's a little bit of a stigma to, oh, you just kind of, you know, you figured it out with the computer. Right. What was always interesting to me was that baseball's always had numbers, right? Like it's just, you were doing different numbers.
Whereas in soccer, and I think this is a little bit more like poker. It really was legitimately no numbers to smart numbers. And a lot of what happened at sort of like the nascent stages of soccer analytics was people who'd been, who'd come through based the baseball world being like, okay, we can start counting things now from the beginning. Let's count things that matter, right?
Let's figure out what matters and count them rather than letting like stuff that we know doesn't really matter, like get ingrained in the game, like RBX, right? So, you know, there's a, there's a lot of stuff where you just sort of go down the list. Shots kind of important, not as important as expected goals. Shots on target.
Well, they sort of falling between shots and expected goals, and they're not giving you much that like one or the other doesn't give you better. Like does possession matter? How do we measure possession? Well, we measure it by share of passes, but we put it up there as like a percentage as if it's time.
Like that's a problem. And, you know, so I think that like, what's really interesting to me when I think about sort of poker is that there has been this drive where it really did go from like, nobody knew any math to like, There was a lot of basic math that made you very much better than everybody who knew no math, to like now there's like much more robust math that gives you answers that you then have to incorporate in how you're playing. Yeah, I was going to say so much of it is just embedded in your strategy. You know, not to go too deep on the poker, but there's...
Go as much as you want and I will translate as necessary. Okay. At the early stages in a no limit Texas Hold'em hand, you've got your pre-flop play where you just have your own two cards and you're trying to decide do I enter the pot or not. Between that and then the first stage when three community cards come up, the flop, it's a lot more static.
It's more defaulted strategy. And if you watch the difference between a top level pro and a mid-level pro, you're not going to see as much difference. You might see, you know, a fair amount of difference with a very new recreational player, but it's really learnable very quickly. And where you really have to turn on your thinking cap and start doing, you know, some critical thinking are the latter stages of the poker hand, the turn, the fourth card and the river, the fifth card.
And that's where you kind of have to put things together and be a good critical thinker. But early on, you can learn so much of this without ever thinking statistically, just kind of by being a copycat, by just setting up in a way that looks really similar. I don't know with soccer, but I, you know, know with football, for example, you could just, you know, you could copy a playbook and say, hey, look, I can watch how they're running their plays, but that doesn't mean you're going to play like them. Yeah.
So soccer, it's interesting. Now that we have expected goals, we can do a much better job of sort of seeing who is defending well and who is defending poorly and who is attacking well and who is attacking poorly. But there is a lot of tactical variation in terms of how you can create those shots. You know, you can, you can create, you know, 1.5 XG in a match by taking five really good shots or by taking 15 average shots.
Like, and, and, and you can attempt to limit it in the same way, right? Like you can attempt to limit possessions so that your opponents don't get very many shots while risking possibly giving up good shots. Or on the flip side, you can be very conservatively and be like, you know, you take pot shots from 35 yards all you want, you know, but you're not going to get good looks. Right.
And, and unlike something like basketball, where I think there is increasingly a solution, right? Like that there is like, you know, one, We kind of like a new set of problems to learning poker than like when I learned poker, where like, you couldn't get as far without theory, without like a grounding and thinking about the why’s and the wherefore’s and the how’s. Whereas now it seems like you can get a lot further down the path without that knowledge. Yeah, no, I would definitely agree with that.
It's, you know, there's just, you kind of had to do it by feel. And this innate wisdom was kind of subconsciously you're just absorbing all this stuff and how good you were at synthesizing that and creating strategy was how good you were going to be a poker 20 years ago and previous to that. Plus, I mean, there was still something to be said for network, just like in any industry, if you could, you know, protege, you know, under someone who, you know, took you under their wing and showed you what they know, you can, you know, get a lot more knowledge really quickly. But even then they didn't have this language to kind of bestow that knowledge upon you quite as quickly as they do now.
The conversations are so much more efficient as, you know, parlance around statistics emerges, which I imagine is pretty similar to poker. That's a really good point that it is just like, look, everybody in soccer, like conceptually understands like good shot versus bad shot. And you can argue about like, is this shot a good shot? Is this shot a bad shot?
Like all of this stuff. But like, you know, when you have expected goals, you have all of it sort of packaged into one thing. And it did, it did give like a really important insight, which is that soccer schemes are won and lost. Players are good and bad.
Teams are good and bad based on creating better and worse shots much, much, much more than finishing the same shots that are worse. What makes you good at scoring goals is getting into good positions more than other people, not being able to keep the ball more accurately from similar positions to somebody else. I am sort of curious, is there something about solvers and the sort of the proliferation of GTO and game theory optimal play that has done something like that in poker that is like twisted the received wisdom like on a pivot point a little bit to sort of like put everybody on the same page with like a basic piece of understanding that maybe wasn't there before? Yeah, let me come back to that one.
I just had a thought when you were talking about kind of zooming into what was important. I was thinking you and I were talking earlier about Nate Silver and he writes previous book focusing on the noise. And I'm thinking about the striker. Can you remind me his name?
The best striker in the world? Let's call it Erling Haarland. There'll be some debate, but let's call it Erling Haarland for now. Excellent.
And you're talking about this ball that comes up waist high on him, and he delivers this, you know, hell of a shot. And I'm thinking, okay, so without fully understanding the signal there is that he made this incredibly athletic play, and it doesn't appear terribly lucky relative, you know, to, and certainly there's probably some rec bias in the first place, knowing he's so fantastic, but this is not a one in a million play for him, like it might've been for someone else. But then how do we judge like his positioning going up to that moment? Maybe someone else positioned slightly better and the ball didn't come up waist high.
It came up, you know, knee high and was an easier blow to deliver. What about who hit that ball in the first place? Is this all noise strikes me probably is pretty noisy without knowing much about it. It seems like that's, that's kind of random variance.
A lot of it, but maybe some of it's not. Maybe, you know, when I think about the jump that an outfielder gets in baseball, you know, we used to judge, you know, who was a good outfielder by who was on SportsCenter diving and catching the ball until we learned some of these guys like Eric Burns, the center fielder for the A's is actually a pretty terrible fielder. He gets horrible jumps and then he's great at diving. I'm a bit 90s New York kid.
Like I grew up watching Derek Jeter dive for me and place like it's just, it is what it is. So on that specific point, there's, there's a couple of ways to think about it. One is that the best, the best forwards in the world do outperform their expected goals by somewhere in the range of 10% to 20%. So there is this idea that yes, you can be better at similar shots.
It's just, it sort of pales in comparison to the other stuff. The other thing to keep in mind is that when it comes, you know, you were saying maybe a better player or, you know, a hypothetical player might be better positioned so that the ball would fall to him more easily. The flip side of that is that like expected goals does capture because Holland gets a shot off from a great position, even very naive expected goals. It doesn't have the height variance captures the idea that he's getting a shot here where other players might have to collect the ball in some way and then figure out a way to lose a defender and get a shot.
So there is like, he is doing something even there that differentiates himself. What he's doing is better captured when you can see in the data that he's making an athletic play to get his foot on a ball that is very elevated as opposed to just sort of, you see eventually in the aggregate, oh look, he gets more shots in and around the box than somebody else does. And, but I do agree that like, there's tons of space there on a shot by shot individual level to look for more information that improves your accuracy and understanding that nets out at the aggregate level, right? Like that's kind of like.
I have this picture in my head of, of three camps where, where you've got, you know, the old school camp where they literally just watch the games and then they come up with their expected goals and whatever other statistics they might, you know, and then you've got the middle camp, which is, you know, a suit soccer fans who know what's going on, who, you know, watch that goal and they say, Oh yeah, that's a hell of an athletic play. And he's one of the few people in the world who's going to do that at the rate that he did. Um, and then you've got the third camp, which are people who barely know the rules of soccer. Like they've been informed just now and they, they don't watch the game at all.
They just, you know, get a statistical, it's just basically written up like a transcript and you watch what the three of them come up with would be really interesting. I, one of the things that I actually took from Nate Silver's recent book is he's talking about the legendary sports better, Billy Walters. And he was, he was saying how she kept all his sources separate. Um, he didn't want them commingling probably for a variety of reasons.
But the reason stated was that he, um, you know, didn't want them influencing each other. And how, you know, naturally agreeable people will be in bias towards kind of seeing, you know, someone else's answers and not wanting to be so far off. But yeah, when I think about something like soccer, it, it strikes me that it would almost certainly be an advantage to know what's going on and to be able to say, Oh yeah, that's a hell of a play. We need to figure out how to quantify that.
But then it could also be noise and it can, it can slow down an optimization of a system of statistics. Yeah. It's challenging. And you know, I, the reality is also that there's not necessarily broad agreement on what you want a statistic like XG to do.
Do you want it to most accurately describe the game that you're watching or do you want it to be optimized for predicting? I mean, this stuff was... I didn't even thought they were in conflict. That's interesting.
Well, okay. So let me give you like a, a simple example. Um, guy goes to line up a shot and as he plants his foot to kick the ball, he slips. And so he goes flailing and the ball goes flying off into the air, right?
Um, XG is going to be blind to that. Um, because the, you know, there's no slips on the grass, you know, figure in there, right? Like, because it's not predictive. Like there's nothing about slipping as you plant that says, Oh, this player slips more than that player when they, when they plant their foot or, or that kind of thing.
But, so like if you're building something that you want to do talent evaluation for a club that you want for gambling purposes, that you want for sort of better, for whatever reason you might want to predict things, um, internal clubs or external clubs. You don't want, you want to be blind to that slip. Yeah. But if I'm giving you a set at the end of the game, that's like, you know, team A had 2.6 XG and team B had 1.4.
Well, if XG was not blind to that slip, it's going to more correctly explain to you what happened in the game than, than otherwise. So, you know, what you want a model to do when you're talking about the sports stuff can be up in the air, which I imagine, like, like at the end of the day, poker is that's one place where poker is very different. Like everything is, is like definitionally zeroed in on results. Yeah.
I think we, especially in online poker where there's less intangibles, It involves having a ball. It is not really the reality. The reality is that basically what you're doing is defending with the ball. It's the equivalent of having a really good running game in the NFL to run down the clock, right?
You can keep the ball in ways that are to protect a lead in ways that are not particularly aggressive, but feel good. And it is absolutely true that doing that from a morale basis, from a coverage basis, the way you're covered by the media, from all of these things, gets better, right? It treats you better. It feels better.
Difficult ways to quantify. That's really tricky. I think that's what eludes the statistics. Right.
But you can overdo it to the point that you are clearly costing yourself. And this, I think, is both sort of what I think about like sort of poker solving and expected goals, is that they provide a real way to mark your sort of like beliefs to market, right? To be like, okay, you know, this player has scored 17 goals, but his XG is 6. You want to tell me he's a great striker who's really great at kicking the ball?
Absolutely. I will hear your argument. But what you have to say to believe that this is a predictive return of goals is that he's the best ever at doing this. Is that true?
Like, is that the argument you want to make to me? And I think that, like, you can, I'm sure, like come up with scenarios in poker in your head where, like, you're jamming 40 big blinds over a raise early in a tournament. And it's the, you know, you can construct for me XYZ reasons that would make it correct in the moment. But now you have something to compare it to, to be like, to evaluate just how extreme that reasoning would have to be to make it correct.
Yeah, and it depends what lens you're wearing. You know, as a poker player, I personally keep a pretty holistic lens where let's say that I'm playing a tournament that's at the bottom of my buy-in point. Oftentimes, I live in South Florida and at the Hard Rock here in Hollywood, it's about 15 minutes from home. And I mean, if I'm playing a tournament and let's say the buy-in was 5K, and I now have 20% of my starting stack left, I'm absolutely factoring in life parts to that equation.
I'm not going to try and optimize for how to play this situation because from an hourly point of view, it's a terrible decision for me. I would say like, hey, I'm willing to, you know, lose a few dollars of expected value to go ahead and free up my night or have a chip stack that's, you know, got more value to it rather than sit here at the table with this, you know, stack that's not worth a ton of money and essentially lose the next five hours to something that I don't regard as terribly enjoyable, playing short stack near the bottom of my buy-in point. And there's something... I imagine at a tournament, say in Jeju, with a much higher buy-in point, you would have a very different perspective on the exact same number of big blinds in front of you.
Absolutely. And especially, there's something to be said about whether or not you have investors because you don't necessarily have that full, you know, it's a little bit selfish for me to be thinking about how I want to spend my night if I'm playing a $100,000 buy-in and $80,000 of it was put up by other people. Then I kind of have revoked that privilege for myself. But, you know, soccer players, athletes, they're human beings.
And hopefully, they're not thinking about what they're going to have for dinner when they're on the pitch in a close game. But only the best can really stay 100% mentally zoned in all the time. And that's where morale does creep in. And there's these holistic decisions where it's like, you know, and this is where rationalization comes up all the time.
Maybe Pep says in a certain spot that, hey, I didn't know that this is maybe 51%. And I'm sure not thinking in this manner, but this is very, very, very slightly the right move. But the shitstorm I'm going to get in the press and all of this effect, and I know how upset, you know, my star forward is going to be if we invoke this strategy. And long term, when I think a little more holistically down the road and how this increases me getting fired by 2%, this is probably not the right move.
You know, whether that is selfish or whether it truly is, you know, a full on outlook. But you can wear different lens. And I think very often one of the one of the benefits of being a critic is you can kind of put on whatever lens and make a great argument that this was a terrible play in a vacuum. But what I think about is in football, you know, maybe it's the fourth quarter and a team is down by 27 with four minutes left to go.
What's their percentage of win? It's got to be under 1%. But, like, if it's fourth and 12 on their own five yard line, should they go for it or not? And that's when coaches kind of tell on themselves, whether they're thinking probabilistically, they're thinking about winning or if they care about not being in the stat book for, you know, biggest loss margin.
Yeah. And then like the flip side, I don't know. I don't know the flip side. It's just another sort of extension of this is like without the stuff to mark yourself to market to, like tether yourself to reality or analysis or however you want to put it.
It's so easy to let run bad affect your decision making. Right. Like, you know, I have friends who work in basketball analytics. And what they will tell you is like, hey, you make a great game plan.
Right. And you're like, we're going to, you know, this is the guy who we want to sag off of and let him shoot threes. We're going to go under the pick and roll against them all game long. And if he wants to shoot threes, you know, that's that's the outcome we want.
And then the first quarter, the guy drills three threes and the coach tears up the plan, tears up the plan. And right or wrong, like you can make arguments either way. But like, are you actually making a good decision at that point? Or are you just responding to the fact that the guy ran hot in the L3?
And, you know, definitely see that all the time in soccer. A lot of it is like, you know, it's really easy to have a hot finishing run for six months. Really easy. You know, great, great soccer forwards take one hundred to one hundred twenty shots a year.
Like, you know, it's the number of shots that a gunner in basketball takes a week. Right. Like it takes a long time to even out. And but like a guy has a great finishing run for six for six months, half a season.
Some team will come in and try to buy him. And a lot of what you're doing with this work is being like, no, no, that's not the one. Like and, you know, making sure that you're making good decisions. And like, I mean, certainly 20 years ago when I was playing, but I got to imagine now, too, that like it's just really hard to withstand the pressures of running good and running bad because you can run really, really good and really, really bad at poker and keep yourself making good decisions.
And I have to imagine that having some of that hard data to fall back on can make life easier in terms of sanity checks. For the online players and for cash game players, the statistics are incredibly important and helpful. For me, for the high stakes live tournament players, it's really the opinions of other high stakes live players because we don't have enough statistics to draw from that they have almost any relevance. If you looked at like who the best statistical high rollers were last year, the correlation is incredibly low, incredibly low.
And that comes down to time horizon. It makes no sense to people on the surface. But what it really is, the way I like to point it out is a year of playing live high roller poker tournaments would probably be like watching two tennis points. It's just like to try and draw who was the better tennis player from two tennis points.
It's not even that fair because I could never beat Novak Djokovic in two tennis points. That would never happen. The idea like if I were to win one, it would simply be because he choked and double faulted. There's no other way for it to happen.
So in that sense, even two tennis points is like overstating how much sample we get. But that doesn't make sense to people because a poker tournament looks like this big journey. And when they've seen it on TV, the announcers have either implicitly or explicitly stated otherwise. So that's what kind of keeps poker shrouded in a cloud of bullshit.
I can tell you, I can tell you personally two points. In 2004, I won the PokerStars Sunday Million the first time I played it. Oh man. If I didn't win it the first time I played it, Lord only knows.
Maybe I just would not have been very interested in poker. I've been playing poker online at that point for a while, but I'm kind of needy with my bankroll. I was not going to plunk down too much dollars for a tournament until I felt right. But then I did and I did.
And then the second one is at the PCA that year in the Bahamas, I came in seventh or eighth. I came in right, right. I busted right before the TV table. And so that was, you know, nice, but it wasn't life changing in any You guys do a bunch of hand breakdowns and stuff like that, which I love, but like, it's fairly high level.
But the interviews, you can get a ton out of even if you're not there for high level poker analysis. You guys have a Discord. You do all the stuff. You're a poker coach.
You play some of the highest stakes tournaments around, and I believe you're working on a book as well. Thanks. I can't endorse that, like, anyone buys GTO Lab products who isn't a serious poker player. It's definitely not for novice at the moment, but I can say, join the Discord.
It's free, and you get to see kind of high-level conversations. There's no necessary engagement, and you can kind of just see how people think about and talk about poker if you're interested in that. Yeah, and then there could be more than just me in our Discord and that Discord. Excellent.
All right, we will be back later this week with more soccer podcasts. I'm actually not sure when all of you will be listening to this, but regardless, we'll be back later this week with more soccer podcasts. Cheers, y'all.