Latest / What's Your Problem? / Moneyball, Soccer, and the Gap Between Analytics and the Real World

Transcript

Moneyball, Soccer, and the Gap Between Analytics and the Real World · What's Your Problem?

• 0:08
  Speaker 1: Pushkin. Around twenty ten, Sarah Rudd was rooting for this
• 0:20
  Speaker 1: British soccer team. It was a team called Blackpool. It
• 0:23
  Speaker 1: had been her great grandfather's favorite team. She was really
• 0:26
  Speaker 1: into it, but she started getting annoyed by this one player.
• 0:30
  Speaker 1: His name was Charlie Adam.
• 0:32
  Speaker 2: He just loved to take shots from really long distance
• 0:36
  Speaker 2: and it drove me crazy. Every possession just kind of
• 0:40
  Speaker 2: ended with him taking like these random shots and it's like,
• 0:43
  Speaker 2: what a waste?
• 0:44
  Speaker 3: What are you doing.
• 0:45
  Speaker 1: Sarah was living in Seattle working for Microsoft as a
• 0:48
  Speaker 1: software engineer, but she really wanted to get into sports analytics. Basically,
• 0:53
  Speaker 1: she wanted to do moneyball, but for soccer. And then
• 0:57
  Speaker 1: she discovered that this sports data company called stat DNA
• 1:00
  Speaker 1: was hosting a contest, a contest where people could use
• 1:04
  Speaker 1: the company's soccer tracking data to generate analytical insights. And
• 1:10
  Speaker 1: she thought this could be my chance, This could be
• 1:13
  Speaker 1: my chance to break into the industry, and also at
• 1:16
  Speaker 1: the same time that that guy, Charlie Adam really should
• 1:20
  Speaker 1: not be taking those long shots.
• 1:22
  Speaker 2: I'm a big fan of searching for inspiration elsewhere, so
• 1:27
  Speaker 2: I started just looking around and seeing like, well, what
• 1:30
  Speaker 2: have other people done?
• 1:30
  Speaker 3: In this space, and it turns out for soccer not much.
• 1:35
  Speaker 2: But there was some really interesting papers in other sports,
• 1:38
  Speaker 2: and the one that caught my eye was by a
• 1:40
  Speaker 2: gentleman named Keith Goldner. He had done something using NFL
• 1:45
  Speaker 2: data where he looked at kind of this, if the
• 1:48
  Speaker 2: ball is on the thirty five yard line, it's second
• 1:51
  Speaker 2: and ten with you know, X amount of time on.
• 1:53
  Speaker 3: The clock, like how likely are you to score?
• 1:56
  Speaker 2: And you know, I thought that was really interesting and
• 1:59
  Speaker 2: I thought, well, isn't that kind of similar to Charlie
• 2:01
  Speaker 2: Adam having the ball forty yards out?
• 2:04
  Speaker 1: So Sarah took this methodology it's called a Markov chain
• 2:08
  Speaker 1: just fyi, and she applied it to soccer. She built
• 2:12
  Speaker 1: a model that let you look at a moment when
• 2:14
  Speaker 1: a player had the ball in a given location on
• 2:17
  Speaker 1: the field, and the model could evaluate what the player
• 2:20
  Speaker 1: did at that moment. Specifically, the model let you ask,
• 2:25
  Speaker 1: did the choice the player made passing forward, passing to
• 2:29
  Speaker 1: the side, shooting whatever. Did this choice the player made
• 2:35
  Speaker 1: increase or decrease the probability of the team scoring a
• 2:39
  Speaker 1: goal on that possession?
• 2:41
  Speaker 2: And yeah, it really hadn't been done in soccer before,
• 2:45
  Speaker 2: which was kind of yeah, why it was kind of
• 2:49
  Speaker 2: a revolutionary paper.
• 2:50
  Speaker 3: I hate to use that term, but.
• 2:52
  Speaker 1: Like I'm saying it listed, I'm saying that listed five
• 2:54
  Speaker 1: point thirty eight had like a list of the ten
• 2:56
  Speaker 1: big papers and figuring out this kind of thing, and
• 2:59
  Speaker 1: yours was the first out on the list, so it
• 3:02
  Speaker 1: was chronological to be fair, but still big first is
• 3:04
  Speaker 1: a big deal.
• 3:05
  Speaker 2: Yeah, and you know it was cool because I was
• 3:07
  Speaker 2: kind of like, oh, well, I'm just doing what this
• 3:09
  Speaker 2: guy in the other football is doing.
• 3:12
  Speaker 3: But yeah, it was pretty impactful.
• 3:14
  Speaker 1: And by the way, what did you find out about
• 3:17
  Speaker 1: Charlie Adam?
• 3:18
  Speaker 2: Yeah? Not good to shoot from my hearts out was
• 3:23
  Speaker 2: what I found out, which was really like soul soothing
• 3:27
  Speaker 2: for me, or it was like I was right.
• 3:34
  Speaker 1: I'm Jacob Goldstein and this is what's your problem. Today
• 3:37
  Speaker 1: we have the third and final episode in our series
• 3:40
  Speaker 1: of interviews with people who are working at the frontiers
• 3:43
  Speaker 1: of technology to help a lead athletes perform better. My
• 3:47
  Speaker 1: guest today is Sarah Rudd. She's the co founder and
• 3:50
  Speaker 1: CEO at Source Football, a soccer analytics company that works
• 3:55
  Speaker 1: with professional soccer teams in the US, the UK, and Europe.
• 3:59
  Speaker 1: There are different to frame the core problem that Sarah
• 4:02
  Speaker 1: is trying to solve for today's show, I'm going to
• 4:05
  Speaker 1: go with a problem that generalizes way beyond sports. The
• 4:09
  Speaker 1: problem is this, how do you translate analytical insights into
• 4:13
  Speaker 1: meaningful changes in the real world. In our conversation, Sarah
• 4:18
  Speaker 1: and I talked about analytics and the challenge of getting
• 4:21
  Speaker 1: people to change their behavior, change the way they make
• 4:24
  Speaker 1: decisions based on analytics. But before we got to all that,
• 4:28
  Speaker 1: we talked a little bit more about that first paper
• 4:30
  Speaker 1: that Sarah wrote to prove that Charlie Adams should stop
• 4:34
  Speaker 1: shooting from forty yards out. Sarah told me that paper
• 4:37
  Speaker 1: also led to some less obvious insights.
• 4:41
  Speaker 2: Yeah, I think like one of the first ones was
• 4:43
  Speaker 2: that crossing from kind of like wide situations, which was
• 4:47
  Speaker 2: real really still very popular. So if you think of
• 4:50
  Speaker 2: like a player coming down the wings and then they're
• 4:52
  Speaker 2: going to kick the ball high in the air and
• 4:54
  Speaker 2: then hopefully a teammate is going to head it into
• 4:56
  Speaker 2: the goal.
• 4:57
  Speaker 3: That is also like pretty low value.
• 5:00
  Speaker 2: Oh interesting, yeah, because you have like a really high
• 5:04
  Speaker 2: chance of like turning the ball over from that.
• 5:06
  Speaker 1: And was that finding contrary to the sort of conventional
• 5:10
  Speaker 1: wisdom of the time.
• 5:11
  Speaker 2: Yeah, at that time, it was at that time there
• 5:13
  Speaker 2: was still a lot of teams using that tactic, and
• 5:18
  Speaker 2: I think, you know, going back, like people were starting
• 5:22
  Speaker 2: to just kind of intuitively be like, maybe there's a
• 5:25
  Speaker 2: better way. So it's not like it kind of like
• 5:28
  Speaker 2: broke soccer or football, but it just kind of I
• 5:32
  Speaker 2: think reinforced some of the intuitions that people had where
• 5:35
  Speaker 2: it's like, maybe there's a better way to do this.
• 5:36
  Speaker 1: And just to be clear, in the time since then,
• 5:38
  Speaker 1: in the whatever fifteen years or so since then, have
• 5:42
  Speaker 1: analytics shown clearly that that's a bad call and has
• 5:45
  Speaker 1: that style of play decreased as a result.
• 5:48
  Speaker 2: Yeah, And you know, now it's actually seen as like
• 5:50
  Speaker 2: a sign of like ooh, the offense is struggling, like
• 5:53
  Speaker 2: we're resorting to this really low probability tactic.
• 5:57
  Speaker 1: So okay, so this paper you write, you hired by
• 6:00
  Speaker 1: this analytics company, and then the analytics company gets acquired
• 6:05
  Speaker 1: by Arsenal, the famous London soccer team, and you wind
• 6:08
  Speaker 1: up working at Arsenal for many years. And one of
• 6:13
  Speaker 1: the things I've heard you talk about about that that time,
• 6:17
  Speaker 1: and that clearly is a big important theme that goes
• 6:20
  Speaker 1: beyond soccer is trying to figure out how to get
• 6:23
  Speaker 1: the people who make real world decisions to actually listen
• 6:28
  Speaker 1: to you, to you and the analytics people.
• 6:31
  Speaker 2: Yeah, it's it's really hard because like if you you know,
• 6:34
  Speaker 2: remove sports from it, just getting anybody to change their
• 6:37
  Speaker 2: opinion based on facts or evidence is really really difficult.
• 6:42
  Speaker 2: And now you're talking about people who are in incredibly
• 6:46
  Speaker 2: stressful jobs where they can lose their job really based
• 6:50
  Speaker 2: on like.
• 6:51
  Speaker 3: Some random occurrence that happens over the weekend.
• 6:54
  Speaker 2: Right, Like, right, the result doesn't go other way, you're fired,
• 6:57
  Speaker 2: Good luck to you.
• 6:58
  Speaker 1: Which is not the way analytics works. Right, Like, the
• 7:01
  Speaker 1: way analytics works is you need a large end, You
• 7:03
  Speaker 1: need to do the thing one hundred times, and then
• 7:06
  Speaker 1: sixty times it'll go the way the analytics says it.
• 7:08
  Speaker 1: Forty times it won't. That's just the nature of the
• 7:11
  Speaker 1: probabilistic world we live in.
• 7:13
  Speaker 2: Yeah, and you know, like analytics is also all about
• 7:16
  Speaker 2: separating process from outcomes, and yet these decisions are still
• 7:19
  Speaker 2: being made on outcomes. So I have a lot of
• 7:22
  Speaker 2: empathy for kind of people who are in this situation.
• 7:25
  Speaker 1: And a coach will get fired more likely if they
• 7:27
  Speaker 1: do the thing that the analytics says that's contrary to
• 7:30
  Speaker 1: the sort of conventional wisdom in the sport. Right. It's
• 7:33
  Speaker 1: like why in American football coaches for a long time
• 7:36
  Speaker 1: didn't go for it enough on fourth down, right when
• 7:39
  Speaker 1: the analytic clearly said they should, But everybody would get
• 7:42
  Speaker 1: pissed at them if they went for it and didn't
• 7:43
  Speaker 1: get it. Now that's change right because of analytics. Interestingly, although,
• 7:48
  Speaker 1: was it this last Super Bowl where the coach kept
• 7:49
  Speaker 1: going for it on fourth down and not getting it
• 7:51
  Speaker 1: and people are like, he did it too much? I
• 7:53
  Speaker 1: was like, no, he didn't. Just because it didn't work
• 7:56
  Speaker 1: out doesn't mean it was the wrong decision.
• 7:58
  Speaker 2: Yeah, And so you know, this just kind of goes
• 8:00
  Speaker 2: back to like you need to have kind of top
• 8:02
  Speaker 2: top decision makers saying like, yeah, it's okay if you're
• 8:05
  Speaker 2: going to do something that looks a little bit unconventional.
• 8:08
  Speaker 3: Yeah, we believe in it, we trust it.
• 8:11
  Speaker 1: So okay. So we've talked about why it's hard. Did
• 8:13
  Speaker 1: you figure anything out about how to get people to
• 8:16
  Speaker 1: change their decision?
• 8:16
  Speaker 2: I would say the best thing would be building trust
• 8:19
  Speaker 2: through a common language, And so for us, the common
• 8:22
  Speaker 2: language was video, so we could build a model but
• 8:26
  Speaker 2: like until we could show them the model outputs on
• 8:30
  Speaker 2: video and walk them through it and say, this is
• 8:33
  Speaker 2: what the model sees, this is what the model predicts.
• 8:36
  Speaker 2: What do you think, like, walk me through it. And
• 8:39
  Speaker 2: so then it became kind of like these collaborative, iterative
• 8:42
  Speaker 2: model building processes.
• 8:44
  Speaker 1: It's sort of a version of having an idea and
• 8:46
  Speaker 1: then convincing your boss that that idea is actually their idea,
• 8:50
  Speaker 1: and once they think it's their idea, then they'll do it.
• 8:52
  Speaker 3: It sounds like that, like maybe a little bit less cynical,
• 8:56
  Speaker 3: but yeah.
• 8:56
  Speaker 1: Yeah, yeah. So so eventually you left Arsenal and you
• 9:00
  Speaker 1: started source Football, this company that you're running. Now, what
• 9:04
  Speaker 1: what led you to make that leap?
• 9:05
  Speaker 2: So, you know, part of starting this company was that,
• 9:09
  Speaker 2: you know, I wanted to learn and to experience as
• 9:11
  Speaker 2: much as possible. But the other part is that we
• 9:15
  Speaker 2: felt like there's a huge need for clubs to get
• 9:19
  Speaker 2: help in terms of getting started on this analytics journey
• 9:22
  Speaker 2: where you know, a lot of them just don't even
• 9:25
  Speaker 2: know where to begin.
• 9:26
  Speaker 3: They don't know what's good what's bad.
• 9:28
  Speaker 2: So our kind of main value add is coming in
• 9:32
  Speaker 2: helping them get set up, helping them get started and
• 9:35
  Speaker 2: then once that happens, doing all of the hard work,
• 9:39
  Speaker 2: hard thinking that's impossible to get done within a football club.
• 9:43
  Speaker 2: If anyone has ever been in a training ground, Like
• 9:47
  Speaker 2: I'm sure it's the same across all sports, but they're chaotic,
• 9:51
  Speaker 2: they're noisy, they're loud, Like they're not good places to
• 9:54
  Speaker 2: do deep testosterone. Yeah, a lot of a lot of
• 9:59
  Speaker 2: really loud music, like humping. You know, like if your
• 10:04
  Speaker 2: office is anywhere near the gym, like forget it. Like
• 10:06
  Speaker 2: you got to wait until everybody goes home before you
• 10:08
  Speaker 2: can actually like have a clear thought. So you know,
• 10:11
  Speaker 2: that was that was one of the things that we realized,
• 10:14
  Speaker 2: is like you can't work on solving these really hard
• 10:16
  Speaker 2: problems and like football is really hard to analyze in
• 10:20
  Speaker 2: terms of analytics, and like you just you can't do
• 10:22
  Speaker 2: it within a club.
• 10:24
  Speaker 3: So that's why we want it to be a little
• 10:25
  Speaker 3: bit outside.
• 10:26
  Speaker 2: And then kind of our our long term vision is
• 10:28
  Speaker 2: actually like you know, we're using this consulting business to
• 10:32
  Speaker 2: kind of fund the development of our intelligence platform, and
• 10:36
  Speaker 2: then the ideas eventually to look for like a set
• 10:38
  Speaker 2: of investors or maybe an ownership group and you know,
• 10:42
  Speaker 2: take control of a club and run it in kind
• 10:44
  Speaker 2: of like the modern progressive way that we think they
• 10:47
  Speaker 2: should be run.
• 10:48
  Speaker 1: So what you really want to do is basically buy
• 10:52
  Speaker 1: a soccer team and run it smarter than everybody else.
• 10:54
  Speaker 1: Is that what you're telling me? Yeah, of course, tell
• 10:58
  Speaker 1: me more about the big gream. We'll do the pieces.
• 10:59
  Speaker 1: But I'm curious, Like that's a big, audacious dream and
• 11:02
  Speaker 1: it's fun tell me about it.
• 11:03
  Speaker 2: Yeah, I mean, I think you know, one of the
• 11:05
  Speaker 2: things that probably everybody in analytics has experienced is that
• 11:08
  Speaker 2: unless you're kind of like the key decision maker, it's
• 11:12
  Speaker 2: always going to be kind of difficult to influence decisions
• 11:16
  Speaker 2: because there's always going to be somebody that has, you know,
• 11:19
  Speaker 2: kind of their own perspective and everything. And you know,
• 11:23
  Speaker 2: at Arsenal, we were really lucky because we had like
• 11:25
  Speaker 2: a lot of resources. But what I see at a
• 11:27
  Speaker 2: lot of clubs is that you know, they only hire
• 11:30
  Speaker 2: one or two people. The work that they do is good,
• 11:33
  Speaker 2: but like there's a limit to what two human beings
• 11:36
  Speaker 2: can do, and so like a lot of what they
• 11:39
  Speaker 2: produce doesn't necessarily like match the gut, and so it's
• 11:42
  Speaker 2: hard to get this buy in. And so then they
• 11:46
  Speaker 2: just say like, well, okay, thanks for that. I'm only
• 11:49
  Speaker 2: going to listen to you if I agree with it.
• 11:50
  Speaker 3: If I disagree with it, I'm going to go with
• 11:52
  Speaker 3: my gut, and so it just becomes difficult.
• 11:55
  Speaker 2: But I think also, like there's so many issues within
• 11:58
  Speaker 2: football clubs that go beyond just analytics or you know,
• 12:03
  Speaker 2: making the right decision on players. I mean, there's so
• 12:05
  Speaker 2: much in terms of like building good, good cultures, you know,
• 12:09
  Speaker 2: kind of making sure that everything is kind of set
• 12:12
  Speaker 2: up in like a professional way, because if you think
• 12:15
  Speaker 2: about it, clubs are kind of run by people who've
• 12:17
  Speaker 2: never worked outside of football.
• 12:19
  Speaker 3: But I think there's a lot of.
• 12:20
  Speaker 2: Lessons that you can take from working outside of sports
• 12:24
  Speaker 2: bring them into football, and we've seen it in every
• 12:28
  Speaker 2: other major American sport.
• 12:30
  Speaker 3: Not to be like super ruthless.
• 12:32
  Speaker 2: And say like we got to maximize profit, maximize profit,
• 12:35
  Speaker 2: because I think like they are these weird social institutions
• 12:39
  Speaker 2: that have a lot of meaning to a lot of people,
• 12:42
  Speaker 2: and so you obviously want to respect that. And you
• 12:44
  Speaker 2: also have these amazing platforms to bring positive change into
• 12:48
  Speaker 2: the world, and so you want to take advantage of
• 12:49
  Speaker 2: that as well.
• 12:50
  Speaker 3: But like, there certainly can be a.
• 12:51
  Speaker 2: Lot more kind of professionalism in them than what we
• 12:55
  Speaker 2: experience at a lot of places.
• 12:58
  Speaker 1: So, now you have this company and you want to
• 13:00
  Speaker 1: use it eventually to take over the world. Correct, before
• 13:04
  Speaker 1: you use your company to take over the world, Like,
• 13:06
  Speaker 1: what are the what are the services you're selling to
• 13:10
  Speaker 1: teams to clubs?
• 13:11
  Speaker 2: Yeah, I mean, so clubs are in really different situations.
• 13:16
  Speaker 2: And then also, like you know, it's not like the
• 13:19
  Speaker 2: US where like every single major League baseball team is
• 13:22
  Speaker 2: like loaded with cash. Typically in European leagues, like you're
• 13:27
  Speaker 2: going to have several divisions, so like you're talking about
• 13:30
  Speaker 2: anywhere from like a single a baseball team to like
• 13:34
  Speaker 2: a Major League baseball team, Like there's huge difference in resources.
• 13:38
  Speaker 1: So like orders of magnitude in terms of revenue, how
• 13:42
  Speaker 1: much presumably they want to spend on your services, et cetera.
• 13:46
  Speaker 2: Yeah exactly, And so like the main area where we
• 13:49
  Speaker 2: can help them is really recruitment, so helping them kind
• 13:53
  Speaker 2: of find the best players. So unlike American sports where
• 13:56
  Speaker 2: you kind of like trade players and draft them, everything
• 14:00
  Speaker 2: here is kind of like an open market where you
• 14:02
  Speaker 2: can buy and sell the contracts of these players.
• 14:04
  Speaker 3: And so if you're a small team, and you have
• 14:06
  Speaker 3: a really really.
• 14:07
  Speaker 2: Good player, like you can just sell him keep that cash,
• 14:10
  Speaker 2: or you can invest it back into your club.
• 14:13
  Speaker 1: Huh. This is this is really echoes of Moneyball. I mean,
• 14:16
  Speaker 1: I'm sure I don't know if you're tired of hearing
• 14:18
  Speaker 1: about Moneyball. That book came about twenty years ago, but
• 14:20
  Speaker 1: like that was the basic idea there, Right, it was
• 14:24
  Speaker 1: better scouting essentially, right, Like you had these scouts who
• 14:27
  Speaker 1: were sort of using their guts and these kind of
• 14:29
  Speaker 1: conventional wisdom heuristics, and then you had what I'm sure
• 14:33
  Speaker 1: now seem like primitive analytics coming in and basically doing
• 14:37
  Speaker 1: a better job of predicting player success, right at some level.
• 14:41
  Speaker 3: Yeah, exactly.
• 14:42
  Speaker 2: And so you know, I hate to say that, like
• 14:45
  Speaker 2: soccer is twenty years behind baseball, but like in a lot.
• 14:47
  Speaker 3: Of ways we are. And so like this is still
• 14:50
  Speaker 3: like the main area of.
• 14:52
  Speaker 2: Value add for a lot of clubs because you're competing
• 14:54
  Speaker 2: in this open market with people with varying amounts of knowledge.
• 14:58
  Speaker 2: So sometimes that knowledge is just hey, like we had
• 15:02
  Speaker 2: scouts at that game, We've seen this player once, like
• 15:05
  Speaker 2: this is our opinion of him. And then you have
• 15:08
  Speaker 2: the more sophisticated teams that are like, Okay, well, we
• 15:10
  Speaker 2: have a database of you know, six hundred thousand matches
• 15:14
  Speaker 2: in the world. We know the fifty best players in
• 15:18
  Speaker 2: every league at every position, we know how much they're worth.
• 15:21
  Speaker 2: So we help teams kind of go from like the
• 15:24
  Speaker 2: former to the latter and just be a little bit
• 15:27
  Speaker 2: more sophisticated in terms of the amount of information that
• 15:29
  Speaker 2: they have.
• 15:30
  Speaker 1: I mean, it's sort of crass to put it this way,
• 15:32
  Speaker 1: but it's really like pricing assets right, like the players.
• 15:35
  Speaker 1: And again I realized this is kind of dehumanizing. I'm sorry,
• 15:38
  Speaker 1: but it is analogous to people valuing a stock or
• 15:43
  Speaker 1: valuing anything they might buy. Right, And the better you
• 15:48
  Speaker 1: can model the value of the asset, the better you're
• 15:52
  Speaker 1: going to be at finding mispriced assets. Right. You want
• 15:55
  Speaker 1: to you want to find bargains. You want to get
• 15:59
  Speaker 1: the most you can and uh for your dollar.
• 16:02
  Speaker 2: Yeah, absolutely, And so you know, there's a lot of
• 16:06
  Speaker 2: techniques that we can take from other industries because it's
• 16:09
  Speaker 2: not that different from it. Where it gets hard is
• 16:11
  Speaker 2: that you have to have those assets work well together.
• 16:14
  Speaker 1: They are in fact human beings.
• 16:16
  Speaker 3: Yes, that is you know the.
• 16:18
  Speaker 2: Kind of joke where it's like, well, the problem with
• 16:20
  Speaker 2: football clubs is that it's full of human beings.
• 16:23
  Speaker 1: So let's talk about what you can model, Like, what
• 16:27
  Speaker 1: are you good at modeling in this context, like in
• 16:29
  Speaker 1: helping in scouting, essentially in helping clubs value, you know,
• 16:33
  Speaker 1: players they might acquire.
• 16:35
  Speaker 2: Yeah, so we're pretty good at modeling everything that happens
• 16:38
  Speaker 2: when a player has the ball, and unfortunately in football,
• 16:42
  Speaker 2: that's like a very very small portion of the gaming.
• 16:46
  Speaker 3: Yeah. And so there was this really cool.
• 16:48
  Speaker 2: Development in the last four or five years where a
• 16:51
  Speaker 2: number of companies have come out with a data source
• 16:55
  Speaker 2: that's basically taking the video feeds from TV and turning
• 17:00
  Speaker 2: into tracking data. So basically they're able to track the
• 17:04
  Speaker 2: location of every player that's on screen.
• 17:07
  Speaker 3: So obviously you don't know all the players, and.
• 17:10
  Speaker 2: Then you know, what we've learned is that really what's
• 17:13
  Speaker 2: going on off screen tends to not be as relevant
• 17:17
  Speaker 2: and the location of those players isn't as relevant. And
• 17:21
  Speaker 2: so now all of a sudden, you can start doing
• 17:23
  Speaker 2: modeling on what people are doing off the ball and
• 17:27
  Speaker 2: use it for recruitment. Because prior to this, similar to
• 17:30
  Speaker 2: the NBA, there were cameras in all the stadiums and
• 17:32
  Speaker 2: you would get this full data set, but only for
• 17:37
  Speaker 2: your league, and so you couldn't really use it for
• 17:38
  Speaker 2: a recruitment.
• 17:39
  Speaker 3: And so this has been like a really big change.
• 17:41
  Speaker 2: So you know, we can get much better views into
• 17:45
  Speaker 2: what is a player doing physically, what are they doing
• 17:48
  Speaker 2: defensively in terms of cutting off passing lanes, things like that.
• 17:52
  Speaker 2: There's still a lot we're not good at with that,
• 17:56
  Speaker 2: but what it's also done is allowed us to say, well,
• 17:58
  Speaker 2: if we're going to move a player from this league
• 18:01
  Speaker 2: to that league, how different is it physically? How do
• 18:04
  Speaker 2: we think they're going to adapt? And so this has
• 18:07
  Speaker 2: really opened up a lot of things. So if we
• 18:09
  Speaker 2: want to talk about like risk adjusted pricing of assets,
• 18:13
  Speaker 2: now we're starting to be able to quantify a little
• 18:16
  Speaker 2: bit like what's the risk of bringing in somebody from
• 18:19
  Speaker 2: a really really different environment into this one versus one
• 18:23
  Speaker 2: that's more similar, Or.
• 18:25
  Speaker 1: So you should apply a larger discount when you're bringing
• 18:29
  Speaker 1: a player from a very different league, presumably because you're
• 18:32
  Speaker 1: taking more risk.
• 18:33
  Speaker 2: Yes, yes, and if not, then it's not an undervalued
• 18:36
  Speaker 2: asset and maybe walk away.
• 18:39
  Speaker 1: So tell me more, like, is the output you have
• 18:45
  Speaker 1: just here is what you should pay for each of
• 18:47
  Speaker 1: these players, Like you have whatever ten thousand players or something,
• 18:51
  Speaker 1: and you put a dollar value on each one, or
• 18:53
  Speaker 1: like what's the output?
• 18:55
  Speaker 3: Yeah, I mean I wish it was that simple.
• 18:57
  Speaker 2: We're still like pretty far from putting it all together
• 19:00
  Speaker 2: there and saying like, by this player at this price.
• 19:04
  Speaker 2: You know, a lot of the difficulties is that there's
• 19:07
  Speaker 2: no good data set on pricing information.
• 19:10
  Speaker 1: So then what do you just have a relative like
• 19:12
  Speaker 1: kind of a stack rank.
• 19:14
  Speaker 2: I mean, there are like estimates, so in the in
• 19:17
  Speaker 2: the media, they'll say like, oh, this guy went for
• 19:20
  Speaker 2: ten million, but then a different media source will say
• 19:23
  Speaker 2: this guy went for fifteen million, because like the selling
• 19:27
  Speaker 2: club wants to report a higher price and the buying
• 19:29
  Speaker 2: club wants for lower price.
• 19:31
  Speaker 3: So there's real no truth.
• 19:33
  Speaker 2: And then salaries for players are not public for most leagues.
• 19:39
  Speaker 1: So okay, so that's a very important variable that you
• 19:43
  Speaker 1: don't really have access to. That's a problem. So what
• 19:45
  Speaker 1: is the output of your model? Then?
• 19:47
  Speaker 2: Yeah, I mean, so we we kind of do like
• 19:49
  Speaker 2: a stack ranking and like a you know, projected like
• 19:53
  Speaker 2: what do we think this player would do in this situation?
• 19:56
  Speaker 3: But like the error bars on these things are like
• 19:58
  Speaker 3: are pretty big.
• 19:59
  Speaker 2: So we're still kind of in the like subjective realm
• 20:03
  Speaker 2: of like based on these factors, we think this or
• 20:06
  Speaker 2: we think that, and then you know a lot of
• 20:08
  Speaker 2: the like the markets change every year in football as well.
• 20:12
  Speaker 2: So even if we had like a really precise model
• 20:15
  Speaker 2: that said this guy should be worth five million, well
• 20:18
  Speaker 2: he should he would be worth five million last year.
• 20:20
  Speaker 1: Yeah, I appreciate your candor you know, so the market changes.
• 20:24
  Speaker 1: You don't have pricing data, there's big uncertainty even on
• 20:29
  Speaker 1: the outputs you do have, Like all of this seems
• 20:35
  Speaker 1: so what is the use of what you're doing? How
• 20:37
  Speaker 1: is it valuable to people?
• 20:39
  Speaker 2: That's the point is that we're competing with people who
• 20:43
  Speaker 2: you know, have gone to maybe three or four games
• 20:45
  Speaker 2: watch this guy in you know, unknown conditions. The human
• 20:52
  Speaker 2: brain is like not conditions to be like very objective
• 20:57
  Speaker 2: and so like, you know, there's a host of like
• 21:00
  Speaker 2: very known biases that like people can can fall for.
• 21:04
  Speaker 2: And so really what we're trying to do is just
• 21:07
  Speaker 2: give you like a much more fair and even view
• 21:11
  Speaker 2: of a player, taking into account a lot of you
• 21:15
  Speaker 2: know the factors that these scouts are trying to account
• 21:19
  Speaker 2: for as well, but just doing it more objectively, more rigorously,
• 21:23
  Speaker 2: you know, over a longer period of time.
• 21:25
  Speaker 1: It seems from what you've been saying, like soccer is
• 21:29
  Speaker 1: behind certainly baseball and perhaps other sports in analytics. I
• 21:34
  Speaker 1: mean your first paper, you were following somebody who had
• 21:36
  Speaker 1: written a paper in American football, and you're like, oh,
• 21:38
  Speaker 1: what if we do that for soccer? Is it true
• 21:40
  Speaker 1: that soccer is behind?
• 21:41
  Speaker 3: And if so, why, Yeah, it's true.
• 21:44
  Speaker 2: And I think, you know, the debate used to be like, well,
• 21:47
  Speaker 2: are we making any progress? I think we've made a
• 21:49
  Speaker 2: ton of progress, but we're still really far behind other sports.
• 21:53
  Speaker 2: And so one answer is that we also don't get
• 21:57
  Speaker 2: the funding that other sports get. So so the level
• 22:00
  Speaker 2: of investment just isn't there. And so if we were
• 22:03
  Speaker 2: behind fifteen years ago, we're certainly not keeping pace, you know.
• 22:08
  Speaker 2: And then I think there's other structural issues with it.
• 22:12
  Speaker 2: So I love to use this image that somebody else made,
• 22:16
  Speaker 2: but it shows the relative size of a basketball court
• 22:19
  Speaker 2: to a soccer pitch, And basically a basketball court can
• 22:23
  Speaker 2: fit into like the little tiny penalty area on a
• 22:26
  Speaker 2: soccer pitch, and so you know, the distances and spacing
• 22:30
  Speaker 2: of players is so much more variable in soccer, So
• 22:33
  Speaker 2: you can't say, like, oh, ten meters is a good
• 22:35
  Speaker 2: distance between me and a teammate, because it depends on
• 22:39
  Speaker 2: the situation that's happening where on the pitch, like is
• 22:42
  Speaker 2: it a transitional moment, is it kind of a more
• 22:44
  Speaker 2: controlled moment, And so there's a lot of complexities like that.
• 22:48
  Speaker 3: You know, the game doesn't stop.
• 22:49
  Speaker 2: There aren't a lot of these set pieces where it's
• 22:51
  Speaker 2: like we have a very choreographed idea of what we
• 22:54
  Speaker 2: want to do, and then you know, the sad reality
• 22:58
  Speaker 2: is that most leagues only play thirty eight matches a season,
• 23:03
  Speaker 2: and so you never see a team in the same
• 23:06
  Speaker 2: context twice.
• 23:07
  Speaker 3: You either play them at home or away and then that's.
• 23:10
  Speaker 2: It, or maybe you play them in a cup game,
• 23:12
  Speaker 2: which is like a really different environment, and so all
• 23:15
  Speaker 2: of these things just kind of like add up and
• 23:18
  Speaker 2: it's like, well, we have fewer resources, it's much more complex, Like,
• 23:22
  Speaker 2: of course we're behind on.
• 23:24
  Speaker 1: This still to come on the show, the one big
• 23:30
  Speaker 1: strategic change that Sarah really really wants someone in soccer
• 23:34
  Speaker 1: to try and why nobody has tried it yet. So
• 23:46
  Speaker 1: we talked a lot about essentially scouting, evaluating players as
• 23:50
  Speaker 1: kind of one of the one of the big things
• 23:53
  Speaker 1: you do. What else do you do?
• 23:56
  Speaker 2: Yeah, I mean, so there's a number of different services
• 23:59
  Speaker 2: that we provide, so you know, it could be doing
• 24:02
  Speaker 2: like retrospectives of how did the team play this weekend,
• 24:06
  Speaker 2: or maybe some kind of I guess in US terms
• 24:10
  Speaker 2: they would call it advanced scouting, but delivering kind of
• 24:13
  Speaker 2: like a data profile on their upcoming opponent and then
• 24:16
  Speaker 2: going into like on field stuff, like we're doing a
• 24:19
  Speaker 2: lot of research now around you know, various things in
• 24:22
  Speaker 2: terms of you know, how can we maximize set pieces?
• 24:26
  Speaker 1: A set piece is like a play analogous to sort
• 24:29
  Speaker 1: of a play.
• 24:30
  Speaker 3: Yeah, so, you know, football is really fluid.
• 24:33
  Speaker 2: It never really stops except for these certain moments that
• 24:37
  Speaker 2: are called dead balls and when a dead ball.
• 24:41
  Speaker 3: Happens and kind of like the attacking area when you.
• 24:44
  Speaker 1: Have like a throw in or a penalty kick or something.
• 24:46
  Speaker 2: Yeah, I mean, penalties are pretty pretty straightforward. They're kind
• 24:49
  Speaker 2: of exceptional. It's like we'll just kick it into the goal.
• 24:53
  Speaker 2: But corners corner kicks are kind of like the most
• 24:58
  Speaker 2: common one because they all happen from the same location.
• 25:00
  Speaker 2: They happen fairly frequently. You can analyze them, you can
• 25:03
  Speaker 2: prepare for them. So you have this set piece opportunity,
• 25:07
  Speaker 2: and then it's you know, what's the strategy we're going
• 25:09
  Speaker 2: to use to maximize it.
• 25:12
  Speaker 3: So a lot of research into that.
• 25:14
  Speaker 1: Has analytics, in a general way changed the way people
• 25:17
  Speaker 1: take corners.
• 25:19
  Speaker 2: Yeah, I mean, I think it has influenced how much
• 25:22
  Speaker 2: time they spend thinking about it and preparing for it.
• 25:27
  Speaker 3: So that's another one.
• 25:28
  Speaker 2: Ted Kinnutsen, who's the CEO of Statsbomb, he's been shouting
• 25:33
  Speaker 2: this from up high for the longest, but set pieces
• 25:37
  Speaker 2: for a long time were really valuable, but teams would
• 25:40
  Speaker 2: only train them for like ten minutes a week.
• 25:43
  Speaker 1: It's an interesting level of analytics. That's an analytical claim, right,
• 25:46
  Speaker 1: It's like, your training time is measurable and you should
• 25:50
  Speaker 1: be allocating the right proportion of it to the right things.
• 25:53
  Speaker 1: And he's essentially arguing that you're underfunding in time corners. Yeah.
• 25:59
  Speaker 2: Yeah, the number of goals scored from corners or conceded
• 26:02
  Speaker 2: from corners was not at all proportional to the amount
• 26:06
  Speaker 2: of time spent training them. And so like that's been
• 26:10
  Speaker 2: a big focus on teams lately. And you know, I
• 26:13
  Speaker 2: still don't think they spend enough time training it. But
• 26:16
  Speaker 2: now you'll see, you know, certain clubs even hiring set
• 26:19
  Speaker 2: piece specialists.
• 26:20
  Speaker 3: So they'll have a member of coaching staff whose job.
• 26:23
  Speaker 2: Whose only job is really kind of like thinking about
• 26:26
  Speaker 2: these things and helping prepare the players for it.
• 26:28
  Speaker 1: If that person basically the corner kick coach, I.
• 26:32
  Speaker 2: Mean they you know, there is a guy who's like
• 26:34
  Speaker 2: a throwing coach as well, so you can even have
• 26:37
  Speaker 2: like some specialties to them.
• 26:39
  Speaker 1: But like guy dreams of being the corner kick.
• 26:41
  Speaker 3: Coach, Yeah, gets promotion too.
• 26:44
  Speaker 1: So I mean just thinking more generally, like it seemed
• 26:48
  Speaker 1: well other sports, my US centric sports knowledge, which itself
• 26:53
  Speaker 1: is somewhat limited. Like clearly analytics have changed. You know football,
• 26:59
  Speaker 1: you see people going for it on fourth down. More basketball,
• 27:02
  Speaker 1: obviously the demise of the mid range shot, right, somebody
• 27:06
  Speaker 1: just realized analytically some years ago, like don't shoot from
• 27:09
  Speaker 1: whatever twelve feet out right, shoot from under the basket
• 27:12
  Speaker 1: or take a three. And in baseball there was the shift,
• 27:14
  Speaker 1: like there are these big, big changes in the way
• 27:17
  Speaker 1: the games look because of analytics. Is there anything analogous
• 27:21
  Speaker 1: in soccer?
• 27:22
  Speaker 2: Yeah, probably, Like the earliest one was that the distance
• 27:25
  Speaker 2: from which people are shooting has changed.
• 27:27
  Speaker 1: Huh.
• 27:28
  Speaker 3: So you're not going.
• 27:29
  Speaker 2: To see a Charlie Adam taking a shot from forty
• 27:32
  Speaker 2: yards too often. And lets you know, there's like a oh,
• 27:35
  Speaker 2: the goalkeepers off his line is out of position. I
• 27:38
  Speaker 2: can get it in there, but because of you.
• 27:40
  Speaker 3: You know, I can't.
• 27:41
  Speaker 2: I can't take credit for it. It feels like, you know,
• 27:45
  Speaker 2: the invention of calculus, where there were a lot of
• 27:47
  Speaker 2: people kind of coming to this conclusion.
• 27:49
  Speaker 1: So you're saying, you might not be Newton, but if
• 27:52
  Speaker 1: you're not Newton, your liibnitz.
• 27:53
  Speaker 2: Yes, okay, fair enough, I'll take that. Yeah, that's probably
• 27:59
  Speaker 2: like the earliest one. And then again, like you can't
• 28:02
  Speaker 2: prove causality, but there's been a shift to this tactic
• 28:05
  Speaker 2: of teams putting a lot of pressure up high. So
• 28:10
  Speaker 2: if they have the ball in the attacking area of
• 28:12
  Speaker 2: the pitch and they lose it, they immediately put pressure
• 28:15
  Speaker 2: on the other team to try to win it back
• 28:17
  Speaker 2: as quickly as possible.
• 28:18
  Speaker 1: Uh huh, kind of a full court press.
• 28:20
  Speaker 3: Yeah, exactly.
• 28:21
  Speaker 2: And so I don't know if the origin of that
• 28:24
  Speaker 2: is based in analytics, but like analytics will tell you, yeah,
• 28:27
  Speaker 2: you're more likely to score a goal if you win
• 28:28
  Speaker 2: the ball back up pie.
• 28:31
  Speaker 3: So that's another one.
• 28:32
  Speaker 1: Huh, why do you think people didn't do it before?
• 28:35
  Speaker 1: Is it not intuitively obvious that that's good?
• 28:37
  Speaker 2: Yeah, because if you don't want to be tired, well yeah,
• 28:41
  Speaker 2: so one, it makes you tired and there is like
• 28:44
  Speaker 2: a risk of injury. Trying to kind of like make
• 28:46
  Speaker 2: the guys super fit to do this. But two, kind
• 28:49
  Speaker 2: of going back to like the fourth down analogy, if
• 28:52
  Speaker 2: you don't win the ball back, then you've committed so
• 28:54
  Speaker 2: many players upfield that you're going to concede kind of.
• 28:58
  Speaker 3: Like silly looking up.
• 29:00
  Speaker 1: It's risky.
• 29:01
  Speaker 3: It's risky. Yeah, it's a calculated.
• 29:02
  Speaker 1: Risk, but you expected value is positive, but the variance
• 29:05
  Speaker 1: is high.
• 29:06
  Speaker 3: Exactly what's what's what are.
• 29:08
  Speaker 1: Some interesting frontier problems? I mean they're sort of getting
• 29:11
  Speaker 1: people to listen to you, which we've talked about, but
• 29:14
  Speaker 1: just on the analytics side, Like, what are you trying
• 29:17
  Speaker 1: to figure out on on the analytics side, what's a
• 29:20
  Speaker 1: big problem You're trying to solve.
• 29:23
  Speaker 2: One area that we haven't really exploited yet, but I'm
• 29:26
  Speaker 2: curious about.
• 29:27
  Speaker 3: But we always have this problem that like we.
• 29:29
  Speaker 2: Can't always learn from the data because everybody's doing the
• 29:33
  Speaker 2: same thing.
• 29:34
  Speaker 1: You can't really do an ab test because everybody does a.
• 29:38
  Speaker 3: Yeah exactly, and so it's really frustrating.
• 29:40
  Speaker 2: I mean, like the biggest example would be substitution patterns.
• 29:44
  Speaker 2: Everybody makes roughly the same type of substitution at the
• 29:48
  Speaker 2: same time.
• 29:49
  Speaker 1: And presumably it's not optimal, right, it's just conventional wisdom.
• 29:52
  Speaker 1: You're like, there could be a whole better world that
• 29:54
  Speaker 1: nobody's ever tried.
• 29:56
  Speaker 2: Yeah, exactly, And so I think, you know, with generative AI,
• 29:59
  Speaker 2: like now you're kind of talking about, like what can
• 30:01
  Speaker 2: we do really smart, really realistic simulations on these.
• 30:05
  Speaker 1: Sorts of things synthetic data.
• 30:07
  Speaker 3: Yeah, yeah, that's what we're hoping.
• 30:09
  Speaker 1: Oh so are you trying to figure out a better
• 30:13
  Speaker 1: way to do substitutions? Yes, yeah, you feel like you
• 30:16
  Speaker 1: got it? Do you feel like you have one in
• 30:18
  Speaker 1: your pocket? Or you're not quite ready to say.
• 30:20
  Speaker 2: I mean, we have some theories, we haven't proven them,
• 30:22
  Speaker 2: but like we also need somebody to say, like, yeah,
• 30:24
  Speaker 2: go ahead, like mess with our substitution patterns.
• 30:27
  Speaker 3: We feel comfortable with us.
• 30:28
  Speaker 1: H It is amazing how I mean, like if you
• 30:32
  Speaker 1: think of the shift in baseball or whatever, like a
• 30:35
  Speaker 1: game can just exist for one hundred years and there
• 30:37
  Speaker 1: can be a way better way to do it, and
• 30:40
  Speaker 1: nobody ever does it just because they don't have the
• 30:44
  Speaker 1: imagination or the courage.
• 30:46
  Speaker 3: Yeah, it's it's.
• 30:46
  Speaker 2: Wild and I think, you know, that sort of thing,
• 30:48
  Speaker 2: like I hope in soccer is less common because it's
• 30:51
  Speaker 2: more of like an adversarial game, but it.
• 30:55
  Speaker 1: Maps to the world more generally, right, Like I think
• 30:59
  Speaker 1: a lot of it is just fear, like you won't
• 31:01
  Speaker 1: get in trouble if you do the thing everybody else
• 31:03
  Speaker 1: did and you have a bad outcome, like you see
• 31:05
  Speaker 1: it in you know, in finance certainly, right, Like a
• 31:10
  Speaker 1: financial advisors just exhibit herd behavior because the way the
• 31:14
  Speaker 1: incentives are structured, Like if you do what everybody does
• 31:16
  Speaker 1: and you lose money, you're like, I was just doing
• 31:18
  Speaker 1: what everybody else did.
• 31:20
  Speaker 3: Yeah, yeah, and it's it's really frustrating.
• 31:23
  Speaker 2: And so you know, I think that again goes back
• 31:25
  Speaker 2: to like why we want to get control of a club,
• 31:28
  Speaker 2: because you know, I think there is a lot of
• 31:32
  Speaker 2: ways that we can optimize things, and if we're only
• 31:36
  Speaker 2: reporting to ourselves, then it's like, well, you know, I'm
• 31:40
  Speaker 2: not going to fire myself just because the theory didn't
• 31:43
  Speaker 2: work out.
• 31:44
  Speaker 1: Yes, it's hard to get to a big enough end though, right,
• 31:47
  Speaker 1: it's hard to get to a big enough samplicized there
• 31:49
  Speaker 1: just aren't that many games. So let's talk about the
• 31:53
  Speaker 1: sort of happy outcome for you. Say it's whatever, five
• 31:57
  Speaker 1: years from now and you have found capital, You've found
• 32:01
  Speaker 1: somebody who wants to essentially bet on you, which is,
• 32:03
  Speaker 1: if I understand correctly, what would have to happen, and
• 32:06
  Speaker 1: you are, you know, with your financy financierer partners running
• 32:12
  Speaker 1: a club. What's that look like?
• 32:15
  Speaker 2: Yeah, I mean I think it would look really really different.
• 32:20
  Speaker 2: I think that the typical profile of who would be
• 32:22
  Speaker 2: working in that club would be quite different.
• 32:24
  Speaker 3: So I think the.
• 32:25
  Speaker 2: First thing is, yeah, nerdier for sure, fresh are It's
• 32:31
  Speaker 2: hard because like, you know, I've played a lot of
• 32:33
  Speaker 2: sports growing up, and so it's like, well, you know,
• 32:36
  Speaker 2: I'm not that nerdy. And then it's like, oh yeah, yeah,
• 32:41
  Speaker 2: but yeah, I mean I think it would look nerdier,
• 32:43
  Speaker 2: but I think it would be a lot more like
• 32:45
  Speaker 2: just a totally different mindset. Really a lot more people
• 32:49
  Speaker 2: who are like, yeah, let's do things differently, Let's have
• 32:51
  Speaker 2: the courage to try new things, and then let's have
• 32:56
  Speaker 2: kind of the thinking power behind it to not just
• 32:59
  Speaker 2: like let's try random things, but let's try really well
• 33:02
  Speaker 2: thought out, creative but courageous things.
• 33:06
  Speaker 1: When you talk about, like, you know, being bold and
• 33:09
  Speaker 1: creative and different, is there some particular thing you have
• 33:13
  Speaker 1: in mind? Is there's like one thing where you're like
• 33:15
  Speaker 1: that one thing I wish somebody would just try it.
• 33:19
  Speaker 2: I mean, right now, it's probably the substitution thing, because
• 33:21
  Speaker 2: that's like the lowest hanging fruit, and it like really
• 33:25
  Speaker 2: really drives me nuts.
• 33:27
  Speaker 1: What's your secret theory? What do you think would be better?
• 33:29
  Speaker 3: Oh, I think you should do early.
• 33:31
  Speaker 2: I think you should kind of treat it like line
• 33:33
  Speaker 2: changes in hockey. So I mean, like right now, what
• 33:37
  Speaker 2: they do is they typically try to bring on maybe
• 33:40
  Speaker 2: like a fast player late in the game that you
• 33:42
  Speaker 2: know would theoretically be running against tired legs.
• 33:46
  Speaker 3: But the problem is that like it's.
• 33:47
  Speaker 2: Too late, and so you know, our ideas basically like
• 33:50
  Speaker 2: bring on people who like know that they only have
• 33:53
  Speaker 2: thirty minutes, run as hard as you can, as crazy
• 33:56
  Speaker 2: as you can, get that one goal advantage, your two
• 34:00
  Speaker 2: goal advantage, and then adapts.
• 34:02
  Speaker 1: And nobody has tried that. That doesn't seem crazy, Like
• 34:06
  Speaker 1: there's one hundred soccer teams, all you need is one
• 34:09
  Speaker 1: verson to try it. It's like a free idea.
• 34:11
  Speaker 3: I know, I don't.
• 34:12
  Speaker 2: Well, you know, if we see like a change in
• 34:14
  Speaker 2: substitution patterns after this content, yeah, that's right. Yeah, but
• 34:18
  Speaker 2: you know, like the crazy thing is there used to
• 34:20
  Speaker 2: only be three substitutions in soccer, and so you always
• 34:22
  Speaker 2: wanted to keep at least one in case a player
• 34:25
  Speaker 2: gets injured, so that really leaves you with two. But
• 34:28
  Speaker 2: during COVID, they upped it to five, and so now
• 34:31
  Speaker 2: it's like, well you.
• 34:32
  Speaker 1: Have that has persisted. Yeah, so they've almost doubled the
• 34:37
  Speaker 1: number of substitutions allowed. Has the strategy in using them changed?
• 34:41
  Speaker 2: No, it's like very slightly changed, And a lot of
• 34:46
  Speaker 2: coaches don't even take advantage of all substitutes.
• 34:50
  Speaker 1: Why do you think nobody has, you know, dramatically significantly
• 34:54
  Speaker 1: changed their strategy around substitutions even though the rule has
• 34:57
  Speaker 1: changed so significant.
• 34:59
  Speaker 3: One, I mean, I guess there's like two explanations.
• 35:02
  Speaker 2: The easiest one is like, well, it's always been the
• 35:04
  Speaker 2: way that we've done it, so.
• 35:06
  Speaker 3: You know why change?
• 35:08
  Speaker 2: I think part of what's driving that though, is that
• 35:11
  Speaker 2: the way that rosters are constructed hasn't changed. And so
• 35:16
  Speaker 2: you know, depending on like the finances of a team,
• 35:18
  Speaker 2: you might not have as many like quality players to
• 35:23
  Speaker 2: go that deep onto the bench. But you know, if
• 35:27
  Speaker 2: this is your strategy, then you can change how you recruit,
• 35:29
  Speaker 2: you can change the profile of that squad composition.
• 35:32
  Speaker 1: The second answer is interesting, right, because it requires you
• 35:35
  Speaker 1: to think more systematically. It's like, oh, the rules are different,
• 35:38
  Speaker 1: so therefore we should build a different team, which is
• 35:41
  Speaker 1: kind of next level.
• 35:42
  Speaker 2: Yeah, and yeah, that's that's not typically how teams operate.
• 35:50
  Speaker 1: We'll be back in a minute with the lightning round.
• 36:01
  Speaker 1: We're gonna finish with the lightning round. Okay, soccer or football?
• 36:06
  Speaker 1: I say football, frieser chips, chips, Cookies are biscuits.
• 36:12
  Speaker 3: So my husband is from India and he will kill
• 36:14
  Speaker 3: me if I say cookies.
• 36:15
  Speaker 1: But cookies are there other britishisms. You sign your emails cheers.
• 36:20
  Speaker 3: Sometimes I hate doing it. Mate is another one?
• 36:24
  Speaker 1: Mate?
• 36:25
  Speaker 3: Yeah, yeah, because that's what.
• 36:27
  Speaker 1: You call everything brilliant. I love talking to British people
• 36:29
  Speaker 1: because they say everything I say is brilliant, but they
• 36:31
  Speaker 1: don't mean it. The way I think.
• 36:32
  Speaker 3: Yeah, it's brilliant, bloody. Yeah, just little things.
• 36:36
  Speaker 1: By British guy once said blind I loved that.
• 36:40
  Speaker 2: Yeah, I've not not done.
• 36:43
  Speaker 3: That, but yeah.
• 36:43
  Speaker 2: I think when I was little, my brother and I
• 36:45
  Speaker 2: used to antagonize my dad until he would like yell
• 36:48
  Speaker 2: at us in British slang, and that was like success
• 36:52
  Speaker 2: for us.
• 36:53
  Speaker 1: Give me your impression of your dad yelling at your
• 36:55
  Speaker 1: British slang.
• 36:56
  Speaker 3: It would just say, like, oh, bugger off, the.
• 36:58
  Speaker 1: Bugger off is good. What's one thing you learned working
• 37:03
  Speaker 1: at Microsoft?
• 37:06
  Speaker 3: It's hard to get things done in large companies.
• 37:10
  Speaker 1: Do you have any tips for getting things done in
• 37:12
  Speaker 1: large companies? Or is it just leave?
• 37:15
  Speaker 2: I mean my answer was leave, But yeah, I mean
• 37:19
  Speaker 2: I think there's there's certain behaviors that maybe I don't
• 37:22
  Speaker 2: necessarily possess or like love, but like aggressive, loud people
• 37:26
  Speaker 2: tend to get things done.
• 37:28
  Speaker 1: There were there in the Balmer area era. That's very
• 37:33
  Speaker 1: Steve Balmer vibes. Who's the most underrated player you've ever seen?
• 37:41
  Speaker 2: Ooh, it might be this guy named Manu Tregueros. He
• 37:46
  Speaker 2: plays for a club in Spain called Viriao, and I
• 37:51
  Speaker 2: think he's brilliant he was kind of born at a
• 37:54
  Speaker 2: time when Spain had like loads of really really talented
• 37:57
  Speaker 2: players in his position, so he never got called up
• 38:00
  Speaker 2: the national team or anything like that.
• 38:02
  Speaker 3: But his running style is like a little bit nerdy,
• 38:04
  Speaker 3: Like he runs very.
• 38:05
  Speaker 2: Much on his toes, and you know, he doesn't look
• 38:08
  Speaker 2: like an athlete. He actually is like a like while
• 38:13
  Speaker 2: he was playing professionally, he was doing his master's in education,
• 38:16
  Speaker 2: so he was like a student teacher.
• 38:18
  Speaker 1: Like, it's just it's just that you love him. It's
• 38:20
  Speaker 1: not that he's an amazing player. Is that I love him?
• 38:22
  Speaker 2: I love him, but like he was also an amazing player,
• 38:26
  Speaker 2: but I think like having this air of like nerdiness
• 38:29
  Speaker 2: around him kind of kept him a little bit under
• 38:32
  Speaker 2: the radar.
• 38:33
  Speaker 1: Who's the most overrated player you've ever seen?
• 38:36
  Speaker 2: Ooh, that's a yeah, that's a really hard one.
• 38:45
  Speaker 1: Are you afraid of getting in trouble? Is there a
• 38:47
  Speaker 1: name in your mind and you just don't want to
• 38:48
  Speaker 1: say it because you don't want to antagonize anybody.
• 38:51
  Speaker 3: Yeah, definitely, there's some of that.
• 38:53
  Speaker 2: And then there's like a little bit of like hindsight bias,
• 38:55
  Speaker 2: where it's like there were guys who were overrated at.
• 38:58
  Speaker 3: The moment and then like kind of crumbled and failed.
• 39:02
  Speaker 2: So it's like, well, if I say like Marlon Shamack,
• 39:05
  Speaker 2: who was a kind of like a notoriously disastrous signing
• 39:09
  Speaker 2: for Arsenal, everyone would say like.
• 39:11
  Speaker 1: Oh, yeah, too easy. Yeah, what's one piece of advice
• 39:16
  Speaker 1: you have for women working in male dominated fields?
• 39:21
  Speaker 3: Oh, that's a really good one.
• 39:26
  Speaker 2: It would probably be that just you belong and know
• 39:29
  Speaker 2: that you belong, and don't let people try to make
• 39:32
  Speaker 2: you feel that you don't. And one of the ways
• 39:35
  Speaker 2: that you can do that is talk to other women.
• 39:39
  Speaker 2: When I got started fifteen years ago, like there weren't
• 39:42
  Speaker 2: any other women. Now there's loads of women working in
• 39:46
  Speaker 2: all different sports and all different sorts of roles. And
• 39:51
  Speaker 2: a couple of years ago there's conference started called Women
• 39:53
  Speaker 2: in Sports Data. This year it's could be held in
• 39:56
  Speaker 2: Philadelphia September seventh, I think. But it's like a great
• 40:00
  Speaker 2: place to connect with people because to me, I find
• 40:03
  Speaker 2: it really powerful to be in like a gymnasium with
• 40:07
  Speaker 2: like a room literally full of women who are interested
• 40:10
  Speaker 2: in sports and data and technology. And yeah, just I
• 40:15
  Speaker 2: don't know, it's good for your mental health, it's good
• 40:18
  Speaker 2: for you know, your self esteem and everything. But yeah,
• 40:22
  Speaker 2: just know that that you belong, that you can do it,
• 40:25
  Speaker 2: and that yeah, nobody should tell you otherwise.
• 40:29
  Speaker 1: So Brad Pitt played Billy Bean and Moneyball. Who's going
• 40:33
  Speaker 1: to play you in Moneyball too? Don't call it soccer?
• 40:37
  Speaker 2: Oh yeah, I have I have no idea. I don't
• 40:41
  Speaker 2: watch a lot of movies, So, like, I couldn't even
• 40:44
  Speaker 2: name an actress right now. Maybe maybe Kiera Knightley because
• 40:47
  Speaker 2: she was in Bendettlake Beckham, So I'll go with Kiera Knightley.
• 40:54
  Speaker 1: Sarah Rudd is the co founder and CEO of Source Football.
• 40:59
  Speaker 1: Today's show was produced by Gabriel Hunter Cheng. It was
• 41:03
  Speaker 1: edited by Lyddy jeene Kott and engineered by Sarah Brugeer.
• 41:07
  Speaker 1: You can email us at Problem at Pushkin dot Fm.
• 41:10
  Speaker 1: I'm Jacob Oldstein and we'll be back next week with
• 41:13
  Speaker 1: another episode of What's Your Problem