DataTopics Unplugged

#59 Did AI Accurately Predict the Euro 2024 Winners? (Part 2)

DataTopics

Welcome to the cozy corner of the tech world where ones and zeros mingle with casual chit-chat. Datatopics Unplugged is your go-to spot for relaxed discussions around tech, news, data, and society.

Dive into conversations that should flow as smoothly as your morning coffee (but don't), where industry insights meet laid-back banter. Whether you're a data aficionado or just someone curious about the digital age, pull up a chair, relax, and let's get into the heart of data, unplugged style!

In this episode, we dive deep into the fascinating world of AI predictions in sports, with a special focus on the Euro 2024 final between Spain and England. Join us as we explore:

  • AI Predictions Revisited: Reflecting on the previous episode (listen here) about AI predictions and their accuracy, particularly Snowflake's prediction for Euro 2024.
  • Challenges of Predictions: The complexities of predicting outcomes in football due to the group stage setup and other factors.
  • National vs. Club Football: Differences in managing national teams versus club teams and the pressures of player selection.
  • Valuing Players: Methods to measure the value of players, from ELO ratings to valuing actions by estimating probabilities.
  • Psychological Pressure: How high-pressure situations impact player performance, referencing the study "Choke or Shine" with examples like Cristiano Ronaldo's goal and the importance of players who perform under pressure.
  • Technology in Sports: The increasing role of technology in soccer, including goal line tech, offside simulations, and connected ball technology.
  • Subjectivity of Offside Rules: The challenges of interpreting offside rules and the potential benefits and pitfalls of semi-automated offside technology. More info here.
  • Technological Impact on Predictions: The influence of technological advancements on predicting outcomes in sports like NBA and soccer, and the potential future of AI in sports officiating.
Speaker 1:

you have taste in a way that's meaningful to software people hello, I'm bill gates I would.

Speaker 3:

I would recommend, uh, typescript.

Speaker 1:

Yeah, it writes a lot of code for me and usually it's slightly wrong.

Speaker 3:

I'm reminded, incidentally, of rust here, rust this almost makes me happy that I didn't become a supermodel.

Speaker 1:

Cooper and Netties.

Speaker 2:

Well, I'm sorry guys, I don't know what's going on.

Speaker 3:

Thank you for the opportunity to speak to you today about large neural networks. It's really an honor to be here Rust Rust Data Topics.

Speaker 1:

Welcome to the Data Topics.

Speaker 3:

Welcome to the Data Topics Podcast. Hello and welcome to Data Topics Podcast Tim, fan favorite friend of the pod. What's up, tim? How are you?

Speaker 2:

I'm doing good, thank you.

Speaker 3:

Thank you.

Speaker 2:

And I'm joined for a first-time joiner.

Speaker 3:

Yep, yep, sander. How are you doing, sander? I'm good Nervous. No, not at all. Oh, not at all Never is. He's used to the first time. Would you mind, uh, getting giving the people a bit of a background?

Speaker 2:

um, I vaguely remember that I am 0.81 sam stall. That's maybe a bit of a throwback. I don't know if you do that anymore it was hard to keep up.

Speaker 2:

It was hard to keep up okay um, yeah, I to keep up, okay, yeah, I'm Tim. I work in the commercial department of Data Roots, which means that I am today. You're not going to hear the very technical statements from me or the very technically accurate lingo. I'm here as resident NBA expert. I've been informed by Murillo, so I have a big passion for sports analytics, especially in basketball. So, um, with the addition that we're gonna have, I hope that's gonna be a relevant experience I'm sure you will, I'm sure it will.

Speaker 3:

And sunday, yes, for the people that don't know you yet um, is there people that don't know me but the? People like any life updates, you know just love that the people your fans are like manander. Tell me what happened with you. Like what's up.

Speaker 1:

No, no. So I'm Sander. I'm a data engineer as well as the team lead of our platform team, but I'm also a big sports fan. I play football. I watch football like every weekend, so that's why I got involved.

Speaker 3:

Yes, yes, yes, yes. I feel like you guys are saying like, as I snatch you from the hall you gotta sit, you gotta eat your broccoli.

Speaker 2:

You gotta, you know, you gotta talk to me about sport.

Speaker 3:

Maybe a bit too far, but so, yeah, this is the Data Topics Unplugged. The summer edition, let's say so we do have a bit of a different format. Um, the euro just finished, right, the euro 2024, the final big final was yesterday was, uh, spain and england. Maybe I'll just put on the screen here the brackets. Um, and I want to say maybe how much, alex, actually do you remember how much time ago was the other euro predictions was like three weeks ago, yeah, maybe a month or something. So I thought it would be interesting to kind of revisit and see how how these ai predictions do. Also, the plug with nba, because the team's here, so we need to talk about me too. I just came um, but yeah, well, like is what, did they get it right? Did they get wrong? So I was a bit curious in hearing your thoughts. I know sunday you also had some thoughts, and tim as well.

Speaker 2:

Um, the first, congrats to spain, big winners the the true mission of the data topics podcast is holding random medium articles accountable for their predictions that they're making, finding those guys.

Speaker 3:

Oh, you said that, get him, cancel him. Uh, no, that's not the point, but I do think that predicting uh scores in sports I think it's an interesting challenge. I think some sports are more challenging than others. I also think that's. Uh, there's probably some interesting reasons there, and even the more philosophical is like do we want to make it easier to predict which? We'll get to that in a bit so cool. So, uh, maybe to start here, uh, like I mentioned, we did have some predictions already, namely, we had one prediction from uh. I want to say snowflake, using snowflake ml. Let me see if I can find the link. They had predicted, so I'm putting on the screen as well and we'll share the links as well, as usual. They had predicted. I think I want to say Spain. No, not.

Speaker 2:

Spain, France, England.

Speaker 1:

But, Spain.

Speaker 2:

it was either France, portugal or England that was.

Speaker 3:

France, portugal or England.

Speaker 2:

Yes, so yeah, this is it.

Speaker 3:

There we go and hands-on match. I have a Portugalugal won. Friends have come through several times in testing and if I had to pick a top three winning team winners would be england. Friends, maybe before the competition started would you have agreed with this prediction or no? Probably yeah, yeah, you think portugal would.

Speaker 2:

Uh, you were very optimistic about portugal as well I mean, I predicted the final to be england, portugal really oh, wow if you look at, if you look at the pure quality of the players that they have and where they play which teams, then it really makes sense for portugal, portugal.

Speaker 3:

They have a great roster maybe I'm just out of touch.

Speaker 2:

Who's uh the, the portuguese uh stars, I guess but you have bruno fernandez, you have bernardo silva, you have stars, I guess, but you have bruno fernandez, you have bernardo silva, you have leo from is I mean these heroes.

Speaker 3:

We got the record of the oldest outfield player as well as the youngest one. Yeah right, that's crazy. Yeah, so, uh. And the youngest one was the yamal yeah, and his birthday was on saturday really but so he became 17 years old.

Speaker 2:

On saturday, sunday, he became the youngest I promise you, I have a list of, like random statistics right, all right. Tim, take us home he also like as a as a 16 year old. He also made another record. He is the tied uh single tournament biggest assist giver, as in. He gave four assists in the in the entire championship and it's the. It's the highest for spain ever and it's the highest single tournament, tied with others and I don't know the others, but it's crazy oh wow, he's also the youngest, youngest scorer ever, the youngest in everything, basically yeah, 16 the goal against France

Speaker 3:

so if he was born, he's 16 and his birthday he just turned 17. That means he was born in. Actually, I don't need to do the math, don't look it's gonna be painful.

Speaker 2:

Declan Rice. In his in his interview before the game he he said something really funny. He said like oh yeah, when COVID started, this guy was 12.

Speaker 3:

He was born in 2007. So it's crazy man.

Speaker 1:

But did you see like he always got sucked in, like the 80th minute? Did you something like the 80th minute? Did you see why? No, why, like um? In germany it's like, uh, prohibited for like someone who's under the age of 18 to work past the hour of 11 o'clock in the evening, so that meant like yeah, yeah really so he had to be taken off because he had to shower before 11 o'clock really otherwise he was breaking the law it's a lot.

Speaker 3:

It's a lot, it's like it's a lot against child labor in germany, so if the year wasn't in germany and they had a different law in the other country?

Speaker 2:

or if it would have been you know more strict in another country, when, but they would have taken the fine. Probably it was.

Speaker 3:

It was a 30k fine, yeah still yeah, if you're up 3-0, it doesn't matter so let's see popular, popular 2007 movies. Spider-man 3 that's funny, that's the first one. The Bucket List yeah, actually, alex, do you know these films that I'm saying, or no, it's just so. Yeah, okay, the Bee Movie.

Speaker 2:

The Bee Movie. Yeah, okay.

Speaker 3:

The Underdog. Yeah, there's some uh resident evil extinction. All these movies were uh came out. Yeah, mr bean's holidays, you know this one we put this up.

Speaker 2:

We put this up for the next like hour and a half.

Speaker 3:

Yeah, we're just gonna list movies from the rest of the podcast. Actually. No, no, but indeed so. Yeah, super young kid, it's crazy. Like what was I doing at 16? I was like, hmm, maybe. Actually I know an anecdote. I think it was by Leverkusen. They also had a very young player. He was also in high school and I remember they announced on Twitter maybe I can try to find it it was a Champions League match and they were put on Twitter like who's playing and who's not playing, who's out because of injuries, who's out because of yellow cards and all these things. And they're like oh, this guy, they're recovering from injuries. This guy, they received a third yellow. This guy's got a red card. And then this guy, he cannot go, has an important exam with the school the next day.

Speaker 2:

Like really, I think at Mala as well, when he was still in training camp, he was making homework can you imagine he cannot be in the Euro final because the teacher gave him an exam? The next day his grades were not good enough.

Speaker 3:

He was grounded by his dad yeah, it's like you didn't do your chores.

Speaker 1:

Yeah, yeah, florian, florian Wirtz, yeah ah, he's pretty good as well right now yeah, yeah, yeah, so uh, school exam.

Speaker 3:

Let me be sure this is dead he wasn't that good in the championship yeah, but it says school exam made 17 year old Florian Wirtz miss Leverkusen's Europa League clash. Yeah, it's funny, huh, I mean it's funny, but but it's good In Brazil that wouldn't fly In Brazil. You'd be like you're probably not in school anyway, if you're serious about football, you're probably not in school. Okay, so this is one. So actually you put England and Portugal. And actually we can check it out. No, because at Data Roots I think we did a little predictions.

Speaker 3:

Yeah, we did yeah but I haven't even checked.

Speaker 1:

I should, yeah, yeah, we did, yeah, but I haven't even checked. I should have checked before we record. But the thing is like with the, with the data roots one, I think you could like uh, every round of the, it progresses update or you just have to fill it in by then. I see, because the the pro?

Speaker 3:

no, I did, I really had to predict the whole yeah, yeah, yeah, but I think it's like so well, we're talking about the euro, right? So the way the Euro is set up, there are different groups and then the. I think the first two of each group goes through, but then some third places also go through.

Speaker 3:

Something like this no four out of six yeah so I guess it's like, yeah, predicting the actual matches, like who's gonna play against two, and the outcome is super unlikely or even the turds, like the turds, like, it depends on how many goals they scored and stuff so it's impossible to have them.

Speaker 3:

Ah, maybe also talking about the group stage. I think this euro I think was the first time ever that after the second round, uh, all the teams in the group were tied right belgium, romania, slovakia and ukraine yeah, they all had the same. I think was the first time ever. As well, they have four points right because they uh, they know after your third.

Speaker 2:

Even right was after the third. It was an entire group stage in our in our group. Yeah, the belgian group was just. Everybody was tired at the end because I was.

Speaker 3:

I was explaining to my partner, like the the, how the who's gonna classify because she's romanian, and I said she was actually the second, second time, I think, romanian, or maybe first time since 2000 and something that Romania actually goes through and I think maybe the first time that they go to the next phase. So she was super like oh yeah, we're gonna go through, we're not gonna go through, ah, but the points are times, ah, and I was like no, there's a system, you know usually goes this, then it goes cards. How many this?

Speaker 2:

there were two teams that like the deciding factor between them was like a qualifying game they played and there even there was tight and it was something like it was. They had to go to, like criteria number six.

Speaker 1:

But in the end they didn't go to that because one of the teams got a yellow card in the 90th or something yeah it's crazy, it's crazy.

Speaker 3:

Um, but yeah, who did you? Who did you have? Because you mentioned England and Portugal, you had it on the final. Who did you have? Tim, or I don't know if you actually picked anyone, but before the tournament started.

Speaker 2:

I didn't do Prono on the bracket, but my guess was just France.

Speaker 3:

Just France, france Okay.

Speaker 2:

Because just the way that they were able to and they showed it as well during the championship the way that they were able to and they showed it as well during the during the championship, the way that they were able to like ice games they, they do that so good. They weren't like Spain was just better, but just in the mentality that they have in winning games is the way that they become became world champions.

Speaker 3:

Yeah.

Speaker 2:

Still like in terms of talent.

Speaker 3:

Indeed, I actually saw maybe a sure Another time talent. Indeed, I actually saw maybe uh sure another tab. Now, another prediction was from the kill love and uh research center, right, um, yeah, just kind of skipping a bit through. I think we covered this last time as well. They had germany and friends on the final and friends winning. So uh doesn't really disagree with what you were saying, tim. I, before the tournament started, I thought friends. I thought england actually and I thought germany. Those are the three. I don't. I were saying Tim, before the tournament started, I thought France, I thought England actually.

Speaker 2:

And I thought Germany.

Speaker 3:

Those are the three. I don't follow football as closely these days, but Spain, I think, definitely surprised me. I think also, when you look at the roster as well, it's like they have some names that I know, but a lot of them are really young. A lot of them they play really really well together. Yeah, and I think one thing also that I hear from um these national competitions, that it's way harder because the players don't train together the whole time, right. So I think sometimes, like, even if you have big names but they, even if you have like five big names, but in their current teams the people are set up to play for them.

Speaker 2:

Uh, that wouldn't happen as well on the national team I think what you see as well is that, um, especially with france, has become really clear in the tournaments that you in in club football, you have like teams that are just really constructed as in you really like by by buying players here and there and then training some players here and there, you get a team that is usually they're more balanced. And then in these kinds of teams and France, for example, in their midfield they had like three, six, eight players like not a creative kind of midfielder. That was very difficult. Only when Griezmann dropped back. Then you saw that and I think that's in general as well what you see in international football.

Speaker 2:

Because you cannot sort of oh yeah, we want Ronaldo to be a Belgian, no, no, you're stuck with the players that have your nationality and then you get way more unbalanced teams where you have to make a distinction. Kamavinga is a really good player, chouamini is a really good player, kante is a really good player, but they're all sort of the same profile, like these breakers, defenders, um, and then you don't have the creativity and like I think it was at some point in philippe osa said it, one of the commentators he said like the coach is really trying to do the best he can, but he just doesn't have. You don't have the players, you don't have the same balance as you're doing club football.

Speaker 3:

It's not like I think in club football you can kind of say I want my team to work like this and then you can find the right pieces. And I think in national teams is a bit the opposite. Like this is what I have. What can we do?

Speaker 1:

yeah, I also think the thing with national teams is you kind of feel pressure to pick your best players right. Yeah, it's difficult to say like, okay, this guy I saw it with england yesterday like kane, he scored like 40 goals at bayern this year and he's playing like super deep and I'm like play him to his strengths right, but then or don't play him.

Speaker 3:

Yeah.

Speaker 1:

So it's difficult to pick those players which fit your system if that means dropping, maybe, a big name player.

Speaker 2:

Yeah, we saw it a bit Belgium as well. Yeah, like Opelna fit way better with the system at Lukaku you play him, indeed, indeed, way better with the system at Lukaku when you play him indeed, indeed, I think we mentioned he feels sometimes not super objective.

Speaker 3:

We're going to touch a bit more on that as well. Because Lukaku, I think one thing that I I always feel like he makes a presence on World Cups, but during the rest of the year I never hear much about him. I feel like now he's an inter, but I think last World Cup and but during the rest of the year I never hear much about him. You know, I feel like now he's an inter, but I think, like last World Cup, he he's a Roma. He's a Roma, but he was an inter before.

Speaker 2:

He was an inter the year before.

Speaker 3:

Yeah, he left there in really nice terms but I also think there's also like some players like Kieb, I think there's some goalkeepers as well that they go crazy well, and I also think there's a bit.

Speaker 2:

Pickford is a clear example. You don't hear anything from him when he's at Everton, and then he's playing for England but to be honest, even for England I was like man, he's an okay goalkeeper, right he's funny but I think he's like, I think, in 2014, the US goalkeeper.

Speaker 3:

I think Howard's like, I think, in 2014,. The US goalkeeper.

Speaker 2:

I think Howard.

Speaker 3:

Howard yeah.

Speaker 3:

He did crazy, good, crazy. The Mexico goalkeeper as well, ochoa, he was doing super. But, like during the year, we don't see as much and I do think there's a bit of the can. These players, how do they deal with pressure? Yeah Right, and I think sometimes it's a bit hard to say okay, you have this guy that is a star, or even like lukaku, but he has a track record like maybe this other guy's playing well, but maybe for a very short-term competition, maybe he's going to do super great, you know. So it's a bit unpredictable. I think it's very hard to to predict these things.

Speaker 2:

Um, in the in in the nba they call these like floor razors or ceiling razors, like the, the people you have. You have people that can really play exceptionally at a high level, but not always that consistently, and they're like ceiling raisers, like at the top. They can make your top better. They can make the absolute best of your team, make it better. And then you have the, the sort of floor raisers.

Speaker 2:

I think conte is a very clear example of like a floor raiser. He's always going to be super consistent. You know what you got in him and he's just going to make the team the worst your team can be. He's going to make it better, like you're saying. You're not going to be worse than that just because he's there, but he's not going to make you know the best franz better. He's going to be an essential part of it, but he's not going to make it better. That's where mbappe comes in. He's a see and I think you have. You have these kind of players that can be because of their unpredictability. They can be great and they're going to be ceiling raisers for a specific tournament. They're going to make it because everything, like the coincidence and just the great shape and whatever comes, they're going to be the ceiling raisers.

Speaker 3:

And do you, can you quantify? Like, how are? Can you quantify? I mean, you mentioned NBA and we talked a bit before we started recording that NBA in general has way more statistics. Um, is there like a metric or something, then? Be it they use, or is it more like feeling? Or how do they actually say this guy's a ceiling razor or the razor?

Speaker 2:

in the in the nba way more than than in, because, especially now in the NBA, there is not as much position anymore. There's a very large shift towards position of this basketball. So you can, they're trying to have metrics that sort of represent the impact of people and then, and what you can see, like there's metrics that are like there's a metric called LeBron. The metric is called LeBron because it's, you know, it tried to capture what the value was of lebron versus michael jordan, and this metric is used as sort of and what is the? But there's also box plus, minus, whatever, like there's different metrics that try to represent the individual impact the player has, which is very difficult in soccer because you have the positions and different positions already have a different impact. So, um, but there you could.

Speaker 2:

You could just look at the, the, the impact this person has in this, this, in an individual game, and then the, the variability on that, like if, if there's a on a specific game, it's like super high, and but there's also super lows, and then you have players that have a very steady impact because they're all in basketball. Then you have, like, players like drew holiday, who's a very just, like, similar to conte, defensive, disciplined, does all the right things, all the dirty work. They, these kinds of players, super valuable and you can see it in any, any kind of sports. You always need this, but that's the way you could. You could try to measure. I'm not sure if it, if it is done, I don't know to that extent if, because it doesn't hold a lot of value, I think to teams to classify somebody as a ceiling razor or a floor razor, but that is the way that you would measure it in basketball okay, cool, maybe.

Speaker 3:

Um, now there are different ways to measure, like players. Now we're talking about individual players and I want to move on there, but but just before we do, I just wanted to wrap up the discussion on the predictions. We also internally at Data Roots, we're also commenting a bit on this. One thing that someone mentioned I think it was Nick, another data engineer. He was mentioning that the predictions seem very much aligned with how expensive the players were.

Speaker 3:

And then when you look into the actual um predictions right, so the first thing they said is the elo rating. So the elo rating, I think, is like each team in this case has a score and then if you have a high score, you have a expectation that you're going to win, I guess um, and if you don't win, then I guess you lose a lot of points. But if you confirm that expectation with a little bit of points, it's a bit like these point systems. So they actually compute the ELO rating for the recent international match results, which I think makes sense. Then they have individual offensive and offensive ratings. So teams go score and consider in recent games and then, yeah, basically it's not easy, but etc, etc, etc. And not so surprisingly, they include the cumulative market value from transfer market. So I don't know how much this weighs in in the predictions, right, but I mean it would be interesting to see a breakdown, right, how much that contributed to this prediction. But uh, there's.

Speaker 2:

There's such a strange bias there in the cumulative market value. Like a player like modric is a clear issue with that. Like modric is not going to be a valuable player at this point, he's what? 36, 37? His value, his value on something like transfer market is not going to be a lot because a player, a team, is not going to pay 80 million anymore for modric because he's that old. But in terms of impact, if you were to able, if you're able able to measure sort of the individual impact of a player I don't know if there's metrics in that in soccer, um, like the, the forward passes, the assists, the crosses, the accuracy of passing, all these kind of things, sort of combined into a metric, modage would be super high in that.

Speaker 3:

Still, yes, maybe, uh, I also. We also brought it. I didn't put it on the list, but you mentioned like a metric. One thing that I've seen that I thought was pretty interesting is also from the cure live in.

Speaker 3:

Uh, research, which is the vape, is the valuing actions by estimating probabilities, and I think it's somewhat similar to what you were saying. So, like, um, basically, they compute the probability that an action contributes to a goal, right? So I think the example here and I have some video is that there is a people on the baseline and then they pass sideways, basically, actually maybe I don't know if this is the best visualization here um, maybe this is better. So they have, like, let's say, a four-action sequence. There is a pass that is sideways from the defenders, right, and that's arguably a very easy pass.

Speaker 3:

So in statistics sometimes they just say that's a pass completed, but that's a very easy pass. And then there's a through ball from two to three that puts someone in the edge of a corner. So that pass is also just a completed pass, but it creates a lot of probability of scoring a goal. And then actually there's a cross in the middle and then it's a shot. So then, once they know that it was a goal. They kind of back, propagate and attribute the value back, and the idea is that they try to do this with every action in the game. So I think, even in the presentation we saw from Kiel Levin, actually they were showing how, just using the quote-unquote simple statistics, you see a lot of players' value going high. But then once you apply this framework, like Kevin De Bruyne was super valuable because the actions that he does, even if he doesn't have this high percentage of completed passes, the passes that he does complete are super valuable, right. So I thought it was a very interesting and a very data-driven way of assessing these things.

Speaker 2:

Yeah, and actually maybe, while I'm talking about this, so yeah, maybe first question I have is does this, does this have any way to represent off-ball actions?

Speaker 3:

I think so it depends a lot, and I think it's a bit intrinsic to football that the way that they gather data is literally someone watching a game and annotating these things right, uh. But even if you say a pass is a pest where there's a lot of information you need to capture, right, if you manually annotating um, I don't remember if they have uh off ball actions, maybe they do I'm just saying like there's there, were there used to be players.

Speaker 2:

I'm not sure if you you still have the the archetype, but like the alessandro del piero archetype, where you have somebody who's just always the right time and the right spot to and and it's a skill, right yeah to read spaces and but it's very difficult to put into a mathematical model yeah, I agree, I also feel like I feel a bit about that, about ronaldo, to be honest now ronaldo.

Speaker 3:

Yeah, because I feel like a lot of yeah I think, yeah, in the beginning of his career he was way more of a playmaker and whatnot. But I think the the years that he was scoring a lot of goals. A lot of the goals were like tap-ins or headers or something, and a lot of times it's like, yeah, okay, he was just tapping it in, right, but at the same time, to be in that position to do these things like there's, it does take a lot of judgment calls and it's not easy. Yeah, no, uh indeed. So let me see if they have.

Speaker 2:

Uh, I'm not sure, tim maybe I remember it's also difficult to create these models and like it's very arbitrary, like what do you include into a model that sort of represents the impact of the player?

Speaker 3:

yeah, and I also think that every time you make a decision like that, you're also including your bias in a way. Yeah, right, because you're saying that this well, this feature, you don't include. So even if it is valuable, you're not.

Speaker 2:

Uh, most of my thinking on this subject is like shaped by the thinking basketball podcast and and they do a very interesting thing there where they sort of rank players according to their impact metrics and then you look into does this make sense? Like, does this make sense? Does the eye test, if we ask 20 people to rank these players, does it hold true to the impact metrics that we see? And then you often see outliers and you're like, okay, is this player? Does he or she, do they have the narrative against them because of some things that happen off the court, which happens as well. Like, is that's why that? Why people like them less, is um, or is this some? Is there just a part of the game that we overrate? Like somebody who scores a lot, is that a good player or is the person who provides them with the pass that a good player?

Speaker 3:

and yeah, yeah, and I do think in football the build-up for a goal is well, yeah, it's complex, it's really complex, it's really quote-unquote long right, I think in basketball it's more. It's still somewhat complex, but it's still shorter. If you look at baseball, it's even shorter, you know like, and I think that's, and I think that the shorter it is, the more direct or less complex in a way, the analytics. Like if you look at money, ball and all these things, you know it's, you can almost like, you can almost not, you can almost be a scout and not watch any match, you know what? I think that's a bit unthinkable for football Right yeah.

Speaker 3:

You don't like baseball, do you? That's what I was going to say.

Speaker 2:

But yeah, it happens and they try to apply the monoball principles to basketball. I think today you could not apply them to soccer, like not at all. You could try and, you know, adopt some of them and try to be a bit more data-driven, but today you cannot run soccer by no means. And there's a dimensional problem as well. We talked about it during lunch with uh, with yonas, soon as well, shout out yonas, um, but uh, like there's on baseball, you have basically two people that are involved in an action. Like there's a pitcher and there's a one that you know hits it. I think there's the most important people. You got all the outfielders. I don't know what. I'm not a baseball expert, but like in basketball you have 10 players and in soccer you already have 22.

Speaker 3:

Like there's a dimensionality problem, like the amount of coordinates you have to track, and it's true. But then what about for sports like tennis, for example? There's a there's a lot of analytics in tennis there is right, but it's not as much as baseball, I would say um or maybe the reason why maybe I don't hear as much is because tennis is still an individual sport, right? It's not like you're going to recruit someone for I gave a.

Speaker 2:

I gave a presentation on that a couple of years ago on roots conf. There's a. There's a number of factors that play into the analytical mindedness of a sport. You have the amount of people, the size of the court, which sort of, and the how discreet the sport is. Baseball is super discreet sport. It's not continuous, it's discreet because you have like action, stop, action, stop, action, stop, action, stop. Tennis is already more fluent. You have like a number of hits and it's a. You don't know how many hits, like how many times the ball is going to get kicked across the field. Soccer is a super fluid sport like you can.

Speaker 2:

You can have 15 minutes of gameplay without stoppage yeah true, without the ball ever going out, so you cannot divide it into individual parts, which makes it more difficult to model the game. Um, that's why that's why you basically you can see, like the amount of teams in in america, like baseball is almost all the coaches are do have an analytical background and have a heavy scouting team and then, like it drops off, like um basketball and and and ice hockey. They sort of have similar complexities to the game, similar amount of players, similar gameplay, fluidity. Then it drops down to american football already, although with the American football, with the fantasy football, yeah huge increase.

Speaker 3:

And then soccer you barely see any so are you saying that predicting NBA is going to be easier than predicting football?

Speaker 2:

soccer football yeah, just want to make the distinction.

Speaker 3:

Uh yeah, predicting nba is going to be easier because I think so too I know right it's like I didn't find this this morning at all um, predicting to be a champion with machine learning. So this is also medium post right. So I looked I maybe I should. I mean, I wanted to look more into it, but what I have gathered is for the last four years they made some. I have gathered is for the last four years they made some predictions and they actually got the last four years right, but not only that, they were able to predict the eight out of the last ten NBA championships in the last, well, eight out of the last ten years right.

Speaker 3:

So to me this was a, if it is possible, I would imagine, even though so they read, if I correct me, if I'm wrong here, tim. But I think that the way then they works, there's like two conferences right, the east and west, and I think in the beginning everyone plays against everyone or no. Everybody's everyone against everyone, and then the the top of each each or west the conferences. So there's still like the no, actually no.

Speaker 2:

So everyone plays against everyone, and then they have like a knockout stage you play, I think, um, but don't quote me on that one you play four times within conference, uh, against the team you know inside your conference, and two, two times against the team outside your conference, something like that, um, in the regular season. And then you have the playoffs, which is a knockout stage within your conference, which ends up a final, the winner of the you know, the Western Conference champions against the Eastern Conference champions.

Speaker 3:

So there is, there is still like there is a points part, but there is still like a knockout part. Yeah, right, which, as we mentioned before I guess I don't know if you mentioned before I definitely mentioned before the recording that I think that's also something that is super, like the an outlier, someone that is not doing good, like maybe the striker is not feeling well, is sick, or something. It has a huge impact for euros, especially because it's a one match knockout. So in nba there's still a knockout, but the person was still able to predict fairly accurately, right. And one of the reasons why I hypothesize and that's what I wanted to hear from you is that the knockouts in nba's are seven games right. So even if there is an outlier there, it's still diluted a bit across the, the multiple they need to win, for a team needs to win four times at least to go through.

Speaker 2:

Yeah right, yeah, so you have, you know, more statistical relevance.

Speaker 2:

For starters, there's also another trend trend in there that is going to make it very like very much, and I'm wondering how applicable this is to soccer, for example. One of the trends that you see the last couple of years in NBA basketball is the fact that the game speeds up, as in there's more possessions typically per game. They go for shorter possessions, more possessions, and it's a, it's it's statistical reasoning. Um, the better teams want to, you know you want to up the tempo because the more possessions you have it's a lot of large numbers the more possessions they have, the more likely that the best team is going to win. So you try to up the uh, the tempo you you really like as a, as a good team, you really want to up the tempo. And then there's sort of the, the thing where you know good teams are built to have a lot of good possessions and then the less good teams start to look at the good teams, to how they build their teams, and then at some point everybody starts.

Speaker 2:

You can see the pace increasing, and so the offensive rating goes up.

Speaker 3:

When you say pacing, you literally mean the amount of seconds. For example, it would always go down to the amount of seconds it takes for the possession to turn in basketball yeah, they measure it in amount of possessions per game. Oh, wow.

Speaker 2:

And it goes up steadily for a long time.

Speaker 3:

For any team.

Speaker 2:

Average. There might be a team sometimes that slows it down because they specifically have a type of basketball that they want to play because they're not as good, so they want to slow it down.

Speaker 2:

Yeah it happens, it happens, uh, the lakers in 2020. They were a very slow paced team because they had a very good interior defense, so they wanted to sort of constrict it every time to a half court basketball, where this, you know, the, the, the features that they had came out really well. Oh, wow, like, um. But because of this, like, the statistical models that you have will become more relevant because you're sort of already, by increasing the pace, you're averaging out the noise yeah, a bit fair enough so each game becomes statistically more relevant.

Speaker 3:

So as a whole, the predictions should also be more easy to make, because there's more signal compared to the noise as a machinery engineer, maybe also draw a bit the parallel with football, because in football, like you know, I mean it's a bit I don't know how relevant this is today or how accurate this is today, but I know that, like, maybe a few, like four years ago, whatever the Spanish way you know, the tiki taka, high possession and all these things was really the scene as the the top, you know, which is like the opposite of what Tim is saying because you have less turnovers, right, but it's like I think football is different because there's like there's no clock running down right, so as long as you have the ball, that's true that's true, like, yeah, you see what you're saying right, because actually I heard that argument as well, that one of the reasons why possessing the ball is so good is because it's also a way of defending, like if the team doesn't have the other ball.

Speaker 2:

You said the clock runs, right yeah, um, but I the reason why I wondered and I wanted to look this up, so I went into the, into the statistics that you can find on the uefa website and for the current european championship, you can find the amount of possessions and that the teams have, and I think I found it here, um, so there's, for example, spain had 123 attempts on goal, versus England 75 only.

Speaker 2:

But the final no no, just in general in general, the entire tournament but Spain, for example, also had 411 attacks where possessions, whereas England only had 344. So on the same number of games they had like 70 possessions less, which means that on average each game that's 10 possessions less. They played 7 games and I wonder. I don't know if there's any statistical relevance. I hope that there's somebody exploring this. But, like for me as a neutral supporter, I would love it if there's more possessions because, like the fast-paced game where you know they go up and down, up and down, it's, it's, it's amazing.

Speaker 2:

But it should also like if you, if you have a lot of, if you're a good offensive squad, you also want to have a lot of possessions. You want to try a lot of things out.

Speaker 3:

Yeah, I think I also feel like so a bit. I also live in the us, right, and I know that how they feel about soccer, alex, I know how you feel about soccer as well. Um, that, uh, it's a very like the. The scoring moments are not as frequent as like basketball or football and all these things, right, and their eyes. It becomes a bit boring. Yeah, right, um, but then I also think that it becomes way more impactful. Right, like in in basketball, you cannot every time someone scores, they cannot celebrate like this in football, right, it's not like free throw, you know, like, yeah, yeah take off your jersey.

Speaker 2:

Technical foul. Second time you do it, you're out. But uh, and I also.

Speaker 3:

I also think, like, because you mentioned it, you're out, but and I also think like, as you mentioned, yeah, you would like to see more actions, right, like more back and forth, more things like this, and I was also thinking how, yeah, I think also that the hardship to predict football is that these things are very like, it's not like I think, if you were scoring, if the expectation is that you score, are very, um, like, it's not like I think if you're scoring, if the expectation is that you score 100 points per match, maybe there is.

Speaker 3:

Also, it makes it easier to predict who's going to win, because yeah there you know like statistically these things say they, once you get more data, the the likelihood of an outlier, the impact of an outlier, is smaller there's more statistically relevant events in basketball.

Speaker 2:

Yeah, like the things that you're trying to predict, like which are your outcome metrics? In the end? There's like an average game in basketball, I think, the average offensive rating, which is amount of points per 100 possessions. So they try to normalize for the possessions because otherwise you cannot compare between error anymore. In basketball the offensive rating I think it's like 118 last year, which is crazy. Like that means every 100 possessions and there's a there's roughly 100, a bit more than 100 possessions each game.

Speaker 2:

You get like 120 points as a team on average each game that's a lot of things to predict that there's a lot of actions, yeah, plus the fact that you have only five people. So a single action is very is way easier, like there's. It often happens that there's two people doing a pick and roll. A single action is very is way easier, like there's. It often happens that there's two people doing a pick and roll a single type of action the other three are standing still in soccer. It's almost, in think, unthinkable that there's two people doing something and the other nine are just like.

Speaker 3:

Oh, awesome guys, you do you unless you're messy like a few years ago yeah, like messy and Iniesta, but yeah, yeah it's true, but I also think it's like yeah, I think in football it's a bit, maybe in basketball I think it's also more clear who's active in the play, and in football I think it's less clear as well yeah, yeah.

Speaker 1:

I think you can argue that in a football action like everyone's involved actually. I think in a task everyone's involved actually I think it's like he probably has an instruction for every instruction for every player in every situation where he should be very I think it's like it's less of a.

Speaker 3:

I think that the elbow right of influence kind of it's not like an elbow, it's not like it drops, I think it's more like of a slowly fades out right, like someone that is further from the action has. But still, yeah, you know, could right and never know, you know. And I think now we're also talking about on the game level. But I think again, if you take more of the competition level, um, it also becomes it's the same thing. For example, I also think it's probably easier to predict the uh, who's the winner of a points based competition than a knockout competition yeah, very simple thing as well.

Speaker 2:

Like I think and I know, an average basketball season is 82 regular season games, and then you have at least 16 playoff games. So you have, on average, each season 98 games. That's not like in soccer you play less, huh so you already have less statistical relevance.

Speaker 2:

That is true that is true, so that means that a single like, a single sum, like if, if you, we've all seen like random goals happening I think own goal was the top scorer of the european championship um, but like if something like that happens, where somebody kicks in a cross and you hit it with your knee and it ends up in the top bin and you're like, yep, all right, this, this might impact, they have a huge impact on the rest of the season, whereas somebody who accidentally bounces a basketball and it ends up in the yeah, that's true, it doesn't have an impact. Right, that's, that's true, that's in-game, but also because there's 81 other games.

Speaker 3:

Yeah, that's true, that's true. So I guess it's like which statistically makes sense, right, when you have more data, the impact of this, and that it's like the we're doing a long run of explaining the law of large numbers.

Speaker 3:

Yeah, law of large numbers, yeah exactly, exactly, all right, thanks very much. That was the conclusion of this, but I also think there's a psychological aspect, right? So one thing we mentioned as well when we were discussing the predictions from Snowflake, ml and the K-11 research, someone brought up that this reflects on the complexities of the psychological, like the complexity of football, which includes the psychological effect on players. I guess, like you're losing the match it's 80 minutes, what decisions are you making? And in that I think it was the same presentation you were also there, right, sander for the K-11, data Science 11?.

Speaker 3:

They also try to quantify the impact of psychological pressure. So the title here is choke or shine quantify quantifying soccer players abilities to perform under pressure. I think they took the vape still the, the framework before about the decisions and the valid decisions bring. But then from that they try to say okay, when it's high pressure moments or low pressure moments, and they they split it around. So this is a very short post, but they also have the, the banner here, um, but then they kind of say what is high?

Speaker 3:

First, they try to quantify what is high impact and then there's a before match and aftermath and during match.

Speaker 3:

So before match is like if you lost the last three games. Maybe there's a high pressure, right, and there's also the image like you got scored the 80th minute, right, or this is a final. So they also try to quantify these things and then they try to see for each player, what's the impact of that. So, in terms of making the right decisions and in terms of executing those decisions correctly, and, for example, one thing that they see here and the people that are following the live that are just on the audio, this is an image with neymar has some statistics every contribution per per 90 minutes. If it is high pressure, I guess he has higher contribution, I guess uh. But then it also talks about, like, how many shots, how many uh passes, dribbles, and you can see that um, more actions come when there is a well, I think actually this, this graph is a bit confusing but uh, he dribbles more when there's high pressure and these actions have high ratings, I guess it's there that tries to make stuff happen indeed, he tries.

Speaker 3:

So I think it's like there are players that but his contributions drop on average.

Speaker 2:

Like he has a lower average contribution for high pressure situations because the density is more moved to the left.

Speaker 3:

Like the book, I think so I think this graph is a bit confusing. Maybe we can actually open, and I did try to spend. I remember the presentation and I also was trying to, and I remember the conclusion of the presentation is that neymar chokes on the pressure. I'm just gonna ignore that, I'm gonna pretend that I didn't understand that, but, uh, that was something. That was something like that, right, but they do try to. So I think this is a bit.

Speaker 3:

Now I'm opening up for people following the live stream. This is the, the banner. So when you do research, you also have like a banner. You present stuff. So I think this is a bit the same. Uh, and here they show this is really laggy derive insights about players performance under pressure. We aggregate his performance metrics under different pressure levels and then they see what's the actions with high ratings, average ratings, low ratings, and you see amount of passes, amount of shots and all these things. And then they also kind of flip this around into recruitment, for example, player acquisition, right. The use case here is that you have modest, so you're trying to find a replacement for that person, and then try to see which players and how do they perform under pressure, right, so which players actually, uh, perform very well under pressure, the shot execution, for example, under pressure, which ones are average and which ones are like the contributions over 90 minutes.

Speaker 2:

And then you can also see, kind of like I think they have here somewhere- sort of a weird statement to make that you want to have people who perform better under pressure, because there's like a, there's a. There's a reverse statement as well like it's when there's not enough pressure, they don't perform well. Like yeah you could be mario balotelli and be like 90 of the games. You're like man yeah I don't even want to run like you also don't want to have these kind of like.

Speaker 3:

But I think that he probably scores really well on this because high pressure moments, but for example there this guy, for example, they say jiru, jiru, jiru is a clutch goal scorer, so when there's high pressure, apparently he scores a lot of goals, which is good, but at the same time, it's like it's like the same thing you mentioned with the floor razors and the ceiling raisers right, if your team is already good, you also want people that perform well under pressure.

Speaker 3:

Because if you're already Like, I think actually Spain is a good example, at least Spain from some years ago. Right, they had a very possessed ball style, but if they were falling behind, I feel like they had a hard time converting to golf. Yeah, yeah, but I feel like, because they never fell behind, because they always have possession, they were always controlling the game, you know. So I think it's like it's almost like you never get to the high pressure situations because your team what is under pressure performs really well. Yeah, so I'm saying yeah, so I think it's a bit of. I mean, of course, ideally you want someone that performs well under pressure and, uh, without pressure, right, but I think, in reality, if you have some trade-offs, you kind of need a bit of both players you get trade-off.

Speaker 2:

Or you can say like you want, you want people to have um sort of robust curves with regards to pressure yeah, indeed they keep executing the same way. That's what I like I I've been a basketball coach for eight years. That's what I would be looking for is like a player that, when the pressure amounts, they keep executing the way that I expect them to but I do feel there are some players that perform better in the pressure yeah, yeah yeah, but is that because they're not living up to their potential when there's no pressure?

Speaker 3:

I don't know. I don't know, I mean, okay, I don't think I don't think anybody is like like, like for example giroux as well yeah.

Speaker 1:

I find it an interesting example, because he also only got substituted when there was a high pressure. Yeah, Like when he, he rarely started for Chelsea. So when they need a goal in the final 10 minutes, they put him up top and just two balls long balls.

Speaker 3:

Okay, Actually, you're gonna be more impactful right If they tailored the game for you yeah, well, I guess it's like there's always the argument of if they put someone else into the ball for that person? They wouldn't score.

Speaker 2:

Yeah, fair enough, right? Yeah, it's a bit of yeah, but how many? How? How large is this statistical sample size in this?

Speaker 3:

so I don't know I hope it's, because it's kul.

Speaker 2:

It's going to be good probably. I think so too right. A lot of respect for their statistical analysis.

Speaker 3:

And actually I wanted to find someone from the that works on the KUL 11 research for sports to to be with us, but not yet, but one day to get them.

Speaker 2:

I think in these kinds of things it's also like what is interesting is that you can. You can construct your roster to sort of have people that take up this role? Yeah, like you have. We used to have three smertens in belgium who used to, in belgium, always act as a super sub like and it's you come on the court.

Speaker 3:

Super sub is the best, like backhand slap, kind of like. You know, it's like, I feel like it's. I don't know. I don't know if film calls me a super sub.

Speaker 2:

I don't know if I'm gonna be happy or if I'm gonna be like, yeah, you're just calling me, you know I think at some point in in in belgium there was sort of the agreement that driss mertens was one of our best players and yet he was the most valuable for the team coming off the bench yeah, but and that actually I have like I do think there are some players that need either to come out or they need to go win, because sometimes, like in the beginning, the nerves are once the ball is rolling.

Speaker 3:

there's a psychological thing as well.

Speaker 2:

I think like if you know that that's your role and you accept and fully embrace that role, there's a liberating part with mentally where you come on the court and you know they're like, oh, now we expect this from you. And you know like like, oh, now we expect this from you. And you know like, all right, cool, they're looking towards me. I like this, like I've. I've been in basketball. I've been in situations where you come on the court and people sort of look at you like you do this now, and you know like, okay, it's from, it's okay for me to take initiative. Nobody's gonna blame me if I take initiative and it goes wrong, because that's why I came on the court like I'm expected to shoot the ball.

Speaker 3:

But there's also an argument of why. Why do you need to come off the bench? For that?

Speaker 2:

why can't?

Speaker 3:

you just start the match and people just look at you to do the amvc because tactics like uh, how many, how many coaches have started to get like francis?

Speaker 2:

francis played the entire tournament. Like this, like we don't want to have a against goal, you don't want to start by making actions at the midline, because that's not the tactical plan you have in mind. You don't want to lose possession at your own half, because it leads to, but at some point, if you're 2-0 behind, then it's perfectly fine to start making actions at half court.

Speaker 3:

Yeah, no, I agree, I agree, I agree. Maybe, yeah, no, I agree, I agree, I agree, maybe, um, sorry, there we go, um, maybe. One other thing I wanted to mention and I think this is going to be controversial talking to you, tim, because you're very much into it but eventually, like the people that perform well under pressure, I also think of, like the buzzer shots in basketball. I also think of, like michael jordan and all these things.

Speaker 3:

You know, usually there are players who say this is the guy that's going to take the last shot and usually I think, because basketball, I think, like the, the execution time is very short, you know like you need to be the guy that gets the ball and shoots. You know it's not like you get the ball, you dribble and this and this. So I feel like the margin for error, like one small mistake, also costs a lot, and I costs a lot, right, um, and I mean you know more about basketball than me, but I also get the impression that, like michael jordan kind of guys, they were the people that perform really well under pressure, better than without pressure. Or maybe I'll rephrase it a bit the impression I have is that when the games were very high pressure, when the stakes were high, those are the people that really stood out. I don't know if it's because they play better or everyone plays worse, and they just kept the same level.

Speaker 2:

But when you look at, yeah, when the stakes are high, like what's going to happen, they're the ones that there's a lot of research that's done into that and what you see in terms of like, if you look at basketball, you see and you look at the sort of the metrics of people in what they call clutch situations, you see that there are players that perform better but ironically it's not like the like. I think Michaelael jordan is is good, but it's not disproportionately a lot of a lot better than he usually is, better than other players in the court but then I guess he just shows up more because yeah there's an attention part to it.

Speaker 2:

Yeah, there's a, there's a. You remember the situations where he's like game six against the jazz and he makes the game winning shot, and that's what you remember. Those are the moments but I also wonder.

Speaker 3:

I also wonder if there's a like they can be on autopilot and kind of like rest there, you know, like he's like yeah, I'm just going with the flow, I'm just going with it because everything's under control. But oh, things are under control, but I can.

Speaker 2:

I can switch it on yeah, but if you look towards the percentage, the field goal percentage on average, and in basketball they go really far in that they have not just your field goal percentage, but on open, wide open or contested shots, they track everything, everything For every shot, where everybody is. They have so many tracking metrics For example, kobe Bryant, michael Jordan they don't have the best field goal percentage on these kind of shots. For starters, you have a super long tail because there's a lot of players that one time in their career they make this shot and then they are the best in this because they have a hundred percent. Um. But stephen curry, for example, he's the best shooter of all time. I don't even think that's arguable anymore. He also has the best percentage at that point. Um, um.

Speaker 2:

There's another like Joe Johnson, um, he's a way less known player. Yeah, I think he has at this point, the most, the most, the most buzzer beaters, which means that your shot leaves before the buzzer goes and the buzzer goes before it goes through the rim. That's the definition of buzzer. He has the most of them, like I think he has nine of them, which is crazy yeah and that might be an example of a player that shows up when it that.

Speaker 2:

That is, I think, one of the only players. Yeah, there's a number, there's a number of them that are known for that, and it's like joe johnson, rob, or so. There are players that tend to show up yeah but at some point I think it's a self-fulfilling prophecy.

Speaker 3:

Yeah it's true. It's true. I think a lot of it is perception as well. Right so, but that's why I think, looking into the numbers and really trying like I think it's a very good way to just try to remain objective. Yeah, I don't know in the sports that, in an area that is very opinionated yeah, I have a.

Speaker 2:

I have a very like, I'm very opinionated on on these kinds of things. Um, and and I I recently like, I recently gave a presentation on what businesses can learn from sports analytics and one of the one of the easiest lessons you can draw is that the statistics alone are not going to bring you there. Um, we saw that a number of times in the nba, where people sort of just try to look at oh, the best team is doing this, we're going to emulate the statistics, we're going to build our team according to the statistics doesn't work yeah the entire moribund era in houston.

Speaker 2:

Um didn't work, yeah, um, but you always have to have, like, the philosophy and the and the statistics work hand in hand and there's then there's just a huge survivor bias yeah it's always sports super prominently present yes, yes, and maybe uh, now to switch gears a bit.

Speaker 3:

Uh, in basketball, I think there's more tech in like, more in the actual like in the game itself, like in the. Well, I'm saying this because I think now in the euro, one thing that stood out a lot is the amount of technology, yeah, that we're starting to see in the sport right, the var, but also the offside. I think offsides in the zero was crazy. There were some things there's like the guy's foot is five centimeters above and it's like man hell, come on, um, I think it was denmark that scored a goal. That was, uh, against yeah germany.

Speaker 3:

I want to say, um, the handball stuff. So now there are chips inside the ball and there's the goal line technology and there's this and there's that. Um, I had a mini discussion. Well, maybe we can uh to share a bit some things for people that may be following us on video. This is a bit the um spain, england offside right so actually not offside, it was inside right but, uh, they're trying to show a bit the simulations that they did. They drew a line, I think. If you go down, yeah, they have this like 3d rendering thing here and you can see the guy's arm is up in front and his knee is actually in front and that's why he's on side right. And then you see some other examples.

Speaker 3:

Let's see this one uh, this is lukaku, lukaku, right and uh, so this was against romania he's offside right because yeah, yeah, but like he doesn't count, no, the arm doesn't count only the body parts that you can score with to me. It's crazy. It's like come on, man. The guy's like look at this man look at his foot. Look at his feet. If he had cut his nail, he still nails that day, you know his knees will his knees will.

Speaker 2:

Yeah, I said that yesterday about the goal of Oyarzabal accident. Like if you happen to be like one leg extended, one leg behind, you probably will not be offside because the other person is with the knee forward.

Speaker 3:

Like that's the point where you know it's crazy that it just depends on and you cannot time it if you try to kick across because I feel like in american football sometimes, because you like the rules are very clear as well, like for, so for touchdowns or whatever, you need to have both feet on the ground, but then you see like players catching, and you can see they're clearly dragging their feet on the ground so they really like move like they're almost like a dance. You know they do this thing, but I think in football it's so dynamic that it's like you can't it's impossible, it's impossible.

Speaker 2:

The tippy toes in football is one of the most beautiful things.

Speaker 3:

The fact that they drag their toes. But, um, maybe this. Uh, I had a mini discussion with italy. Do you like this technology in football? Maybe we can just do a quick. Do you think this adds?

Speaker 1:

to the sport. This one I think I do because, like for me, offside is black and white okay, so I think everything that helps in making it okay, we're gonna move to some other ones.

Speaker 3:

But uh, and what about you?

Speaker 2:

tim, I think, yeah, I, I do love these technologies, um, but in the case that you adjust the, the rule book alongside with it, as in um, you have technology to very much black and white, decide if something is offside, make it that there's also no room for interpretation. Either you do that or you leave it up to interpretation, but then you leave the technology out of it as well, because if you don't have the combination of the two, like, it leads to frustrating and and one of the examples is the I don't know if you have it the goal of a panda, where he like touches the ball, but like very slightly with the hand and there there was a discussion when, um, when the goal happened.

Speaker 2:

There was a discussion obviously in belgium journalism afterwards where I think some people said, like you have to, it has to be a voluntary movement, which I think obviously is not a voluntary movement. You have to drastically change the trajectory, like you have to impactfully change the trajectory of the ball, which obviously doesn't like. And then it's the question like yeah?

Speaker 1:

I think there was just a wrong call right, but I think I think.

Speaker 3:

But I see what tim's point is like. I guess what he's saying is like either you say a touch is a touch and that's it, and if you happen to touch your hand inside the box it sucks, yeah. But other, if you say like intentional, not intentional, but then you have this technology, that kind of supports saying yeah, it did touch, but now you have to say it's intentional, I think it's a bit weird.

Speaker 2:

The goal of the Netherlands against England, where Xavier Simons was against England, right, xavier Simons kicks it like super hard.

Speaker 1:

It's against France, the one Against France.

Speaker 2:

Yeah, they disallowed it. And like there's a guy standing there and everybody knows that the keeper is not going to get it. Like the ball was, I think, 120 kilometers an hour from inside the box like nobody's gonna like. The keeper didn't even react, yeah. But then there was a guy standing next to him and in some interpretation of the rule book you could say that if he dives like, he jumps against the player, so he's influencing.

Speaker 3:

And then you're like, yeah, okay but that's because of offside or what because of offside yeah, because I guess in offside there is a bit of subjectivity in, uh, if the guy interferes with the play, right. So, for example, if you, I'm clearly passing the ball to you, tim, and you're offside, but then you didn't touch the ball because maybe you couldn't, you weren't fast enough, which, tim, is a bit hard think, but anyway. And then let's imagine the ball goes in and it's a goal, right. Somehow, because the ball was a pass to you, you interfere in that play and therefore that's offside. But it's not like. This is a very hypothetical black and white example, but it doesn't have to be as objective, right. Like maybe I do a through ball and maybe there are two players and one is offside, the other one is not, and maybe the ball goes through.

Speaker 1:

Or maybe, like you know how, do you, but that's also, I think, you. You can't like. I see how you see the rules have to follow if you implement stuff like this, but you can't. I mean you can make a handball black and white, you just can't.

Speaker 2:

Yeah, I yeah. Hockey is like and I think there's way less discussion in hockey it's like it hits your foot. It hits your foot Like you have to. It's your responsibility to make sure it doesn't hit your foot.

Speaker 3:

But I think in football there wasn't a. Isn't there a rule that, like, if your hands is near your body, it's not a handball?

Speaker 2:

But then it's gray.

Speaker 3:

No, like then it's a gray area again. But how it's touching? Is it touching, not touching? Look, it's touching. It's not touching this. Oh, it's even clicked or something. I, I see what you're saying. Yeah, I think it's. Uh, I think I don't know. I feel like the. To me, the, the accuracy of these devices is a bit questionable. You know, when I see the guy with like a little foot and even like in brazil there was a joke that like they zoomed in and you can see the cells and like two cells are above them, it's outside, you know, but to me that's where?

Speaker 2:

yeah, that's that's.

Speaker 3:

That's where I have a bit of a like. Come on, man, like, just let it play you know, the thing is also like.

Speaker 1:

The thing I dislike the most is like far was intended to solve, like clear and obvious errors, maybe if you have to use a microchip inside a ball to check whether someone touched it don't bother right, I mean yeah, that's true or you completely automate it away, but make a choice this is the chip you're saying, right, sander?

Speaker 3:

I think you said that it's from.

Speaker 1:

Adidas there's literally a chip inside that every touch of the ball, like 500 times a second, you get data. So every touch of the ball you record and I think this is also the example right, Because I wanted to show specifically.

Speaker 3:

So basically, Ronaldo said that he scored a goal, but the signals in the ball says that Ronaldo did not touch at all.

Speaker 1:

Yeah, so close not close.

Speaker 2:

So he's a liar in the ball. Says that Rinaldo did not touch at all.

Speaker 3:

Yeah, so close, not close. So so he's a liar. Cristiano Ronaldo, we're calling you out, you're first you're first, well, I'm not your second, right, because we're this is the third year but anyways we got the scoop over Sky Sports.

Speaker 2:

Yeah, but yeah.

Speaker 3:

I do think it's. Yeah, I think those.

Speaker 1:

Yeah, I see what you're saying I think it's interesting and you for stuff like this. It's fun to use and to put a sensor in the ball and stuff, but it shouldn't be used for decision making of the referee like a sensor inside the ball yeah, maybe, I don't know the only thing which is nice about the sensor is you can identify the exact point of context for like offside decisions, because that's always like also did the ball already leave the field or not?

Speaker 3:

that's always a point of discussion. We didn't see that in this. We didn't see that data use like that right, but that's what the semi-automated it also uses the. So tell us about it, tell us more about it. What is the semi-automated thing I'm going to put on the screen here?

Speaker 1:

well, so basically it's, there's like 12-ish cameras across the stadium. They hang from the roof and they track every player with like 29 of these body points at any point in time. So basically, they just track you at any point in time and these body points are across your body and are the ones which are relevant for scoring goals.

Speaker 3:

Okay, so this is a video from FIFA for people following the live stream From FIFA, so it's not like a research group or anything. No, no, it's actually. Yeah, yeah, so that.

Speaker 1:

so this exists, but they're trying right, so now they have the sensor to see where the so yeah, the sensor of the ball then makes it. You can identify the exact point of contact when the ball leaves or the first contact is made okay, and I also wanted like these 3d simulations.

Speaker 3:

I feel like that's what people think programmers do you know, like you have like three screens and like this, and then you have the bug going through the wire, you know, and then it goes to the internet or something um isn't that what you do man. Well, yeah, I do it I keep selling the wrong thing um, it's okay, really cool, and then I guess you have all the cameras with.

Speaker 1:

Oh, wow, like the thing here is like. What I always wonder is like you see these body points right?

Speaker 3:

yeah some people, some players are bigger than other players well, excuse me, more muscular, he has thicker bones no, but do they adjust the model based on the player in question yeah, I think so that's cool um, yeah, I was also thinking about that right, like today we have a referee running around, but I was thinking like, oh, yeah, okay, there's some outside examples what if the actually was this used in the euro or no, this one, yeah, semi-automated, okay. Um, I'm also wondering like we have referees that run around, basically, and they kind of there's actually a lot of training where to position yourselves and whatnot. Do you think in the future we'll have a pure VAR?

Speaker 1:

I don't think so but you could.

Speaker 3:

You could because you have, like, a camera.

Speaker 1:

I mean, you still have someone to blame, but just that they're not on the pitch but then like, what do you do if they're things get heated like two players start pushing each other?

Speaker 3:

let it happen, bro, I mean what's the game?

Speaker 1:

but?

Speaker 3:

what does our first dude they're like they take their ears off and just like come on, I'm from brazil, man. I've seen things like too much like if, if dust comes. How do you say dust comes?

Speaker 2:

the shovel or something, that's it shovel yeah, it is okay, push comes the shovel thank you, I looked at it.

Speaker 3:

I was like Alex, help me, it's okay, when the time comes, they don't do much either.

Speaker 2:

So just, you didn't use it the second time around, did you? No, no.

Speaker 3:

I was too embarrassed. My brain you know, my mouth was ahead of my brain. I was like, yeah, I shouldn't have started with that.

Speaker 1:

But there's times at which I do also think, like what's the guy still on the pitch for? Like, literally, like with the VR, for example. It's like people 50 kilometers away that say go watch this screen, yeah.

Speaker 3:

Or even like the linesman as well. Yeah, they're just going sideways. If it's just a camera, if it's just a camera and there's so much really just controlling with a stick that you can move faster than whatever that guy is doing.

Speaker 2:

You could automate the whole thing.

Speaker 3:

Yeah, right, oh yeah, maybe one last final question. We're talking about like predicting stuff with AI and the impact of technology in sports. I had a discussion with my friends a while ago that we're talking about the point system versus knockouts right, knockout stages, right. So in like in the in the uk or in england you have the premier league and you have the fa cup, which I always thought was funny that everyone says fa cup. But the first time I heard it was like fuck up. I was like anyways, um, and I remember we're talking about like which one do you prefer? That's kind of what the discussion was about.

Speaker 3:

And then in the end we kind of agree that point systems are more fair, the best team usually wins in point systems, but knockouts are more exciting because it's unpredictable. So then I guess it kind of to bring this back to this discussion is even if we could like is it a good thing? Like I think guess it kind of to bring this back to this discussion is even if we could like is it a good thing? Like I think usually we say we're predicting and this and this and it's not fair and this usually seems maybe a bad thing, but isn't that part of what is exciting about sports, the fact that unpredictability of it, like if everything was really like, if all the predictions that we did from either Snowflake or the research center from KU Leuven were correct, would that be good? Is that a good thing or is that a bad thing?

Speaker 2:

So that's. That's where the conversation, like we had we had this conversation I mentioned to Sando before that. So basketball has become rather predictable, as you've shown before. Like eight of the ten last nba championship winners could have been predicted. Like could have been predicted based on data available at that point. Um and the uh. Popularity of basketball has been doing like it's. It's going down. Like there's a less people watching. There's more people.

Speaker 2:

Like more people watch the caitlyn clark ncaa final in women's basketball than they did the final in nba basketball really yeah, wow so, um, it is for, like, a lot of people are criticizing that it is because of the fact that it's been very analytics driven, um, and it's analytics driven in two ways.

Speaker 2:

Like, on the one hand, what the nba tried to do was make it really popular by introducing all these rules where people scored more, which is why, like, at some point you can like there there was at some point they introduced a hand checking rule, like if you touch somebody in front of you with your hand, it's a foul which very much limits you as a defender in what you can do in terms of steering somebody. And then they introduced sort of the they're like James Harden went with with fouls at some point all the way, like he always went in, jumped into the defender who was standing still and was a foul, like you had to jump backwards. Like a lot of rules were introduced where, um, it made it more difficult on the defense and better on the offense.

Speaker 2:

people really went away from it, like they really didn't like it, because everything became a scoring festival, like it wasn't special anymore. Um, and then this year they introduced some rules where defense again got some power back. Um, and and you saw that people that again the, the viewing numbers went up, but still the playoffs were very, were rather predictable. Everybody thought celtics couldn't win. Celtics won, voila. There's more to it, obviously, but and the popularity is still not picked up from. You know, yeah, a couple of years ago. So the unpredictability, the, what is it? The Cinderella, the Cinderella story. Like the, the small team comes up and wins big.

Speaker 2:

Those kind of stories sport needs it like it's obvious like you need to have the Leicester story once in a while the. Union story, although Union is not even close, it's not even.

Speaker 3:

But like, you need these stories because otherwise yeah, teams are, but if you have too many of these stories, but that won't happen right if you have.

Speaker 2:

Yet, there's the economies of scale part. Manchester City proves it year and year again, real Madrid proves it year and year again. Uh, but if you have too many of these stories, yeah, then you also lose out.

Speaker 3:

That's, that's also because I feel like I guess the thing is like if you have too many of these stories, then maybe the rules are not fair.

Speaker 2:

The rules are not like you know, because you still want somehow that the best team will win nba was more popular when, when golden state warriors won like four times in five years, than it is now. Now do we have five times a different champion the last five years in nba. So yeah, maybe there is a like we. We do like and that's what they say in the us as well we do like dynasties yeah we do like real madrid and the legendary galacticos. And true, what?

Speaker 3:

do you think something? What's your thoughts on that?

Speaker 1:

I do agree those man, not the nba part no, but I think like it's nice to have these unpredictable games or champions once in a while, but I do think, yeah, you would also not feel fair if you are the best team and that's not somehow represented in you getting rewarded for it at the end of the road so yeah, cool guys.

Speaker 3:

Thanks a lot, this was fun. Thanks, tim, that wasn't a good clap, that was man, it was just closer to the mic thousand excuses. Thanks y'all.

Speaker 1:

Thanks alex you have taste in a way that's meaningful to software people hello, I'm bill gates I would, I would recommend.

Speaker 3:

Uh, yeah, it writes a lot of code.

Speaker 1:

For me, usually it's slightly wrong. I'm reminded, incidentally, of Rust here Rust.

Speaker 3:

This almost makes me happy that I didn't become a supermodel. You said that, cooper and Ness. Boy.

Speaker 2:

I'm sorry, guys, I don't know what's going on. Thank you for the opportunity to speak to you today about large neural networks. It's really an honor to be here. Rust Data topics. Welcome to the data, welcome to the data topics podcast To you.

People on this episode