Counterpoints /

The Sports Analytics Podcast from MIT SMR

Second-Grade Soccer Superstars

play Listen
« Prev Full Series Next »

Even as soccer has become a multibillion-dollar business, with superstar transfer fees exceeding nine-digit figures in euros, the best clubs can still gain a competitive advantage through their youth academies. Ajax, FC Barcelona, and Manchester United are just a few examples of teams that have consistently developed diamonds in the rough into championship contributors. But to develop those players, first you have to find them — and find them before anyone else does. At Chelsea Football Club, home to an academy that has recently produced burgeoning talents like Callum Hudson-Odoi and Ruben Loftus-Cheek, some of that responsibility falls on the head of research and innovation, Ben Smith.

Transcript

Paul Michelman: OK, first some business: This is not a pledge drive. It’s a ratings and review drive. If you enjoy our show, please take two minutes to rate us and, even better, post a review either on Apple Podcasts or wherever you get Counterpoints.

All right, onto the show. When you think of prodigies in American team sports, a few names spring to mind. LeBron James and Bryce Harper graced the covers of Sports Illustrated at just 16. Meanwhile, college stars like Trevor Lawrence and Zion Williamson could be found on recruiting boards for years before they played their way into the national consciousness. But while high school is about the earliest you’ll see the athlete hype machine get revved up in the U.S., it’s a little bit different across the pond. Take, for instance, this quote from a youth coach at AS Bondy, a small soccer club in the Paris suburbs: “You could tell he was different. In Paris, there are many talents, but I’d never seen a talent like him.” That was about a young forward named Kylian Mbappé. And when I say young, I mean he was all of 6 years old.

Ben Shields: Sure, he might have a few precocious dribble tricks up his sleeve, but how do you really know what path a talented 6-year-old will follow? When I was 6, I was the best shooter on my friend’s 8-foot basketball hoop, but I’m not currently raining threes for the Miami Heat. And yet the coaches at AS Bondy were confident what they were seeing was no fluke. Just over a decade later, Mbappé broke AS Monaco records as their youngest professional ever. And today, the kid turned bigger kid is lighting up the world for his club — PSG — and the French national team. As every club on Earth searches for their own Mbappé, the World Cup-winning superstar serves as a prime example of the payoff for pursuing a prodigy’s potential. I’m Ben Shields.

Paul Michelman: I’m Paul Michelman and this is Counterpoints, the sports analytics podcast from MIT Sloan Management Review. In this episode, it’s all about the kids. We dive into the world of elite soccer academies and discover at just how young an age the top teams can identify professional-level talent.

Even as soccer has become a multibillion-dollar business with superstar transfer fees exceeding nine figures, the best clubs can still gain a competitive advantage through their youth academies.

Ben Shields: While the academy system is gaining traction here in the U.S. In Europe, it served as the backbone of successful clubs for decades. Ajax, FC Barcelona, and Manchester United are just a few examples of teams that have consistently developed diamonds in the rough into championship contributors.

Paul Michelman: But to develop those players, first you have to find them — and find them before anyone else does. At Chelsea FC, home to an academy that has recently produced burgeoning talents like Callum Hudson-Odoi and Ruben Loftus-Cheek, some of that responsibility falls on the head of research and innovation, Ben Smith. Here our Ben speaks to Chelsea’s Ben about how football academies can develop talented 6-year-olds into professional soccer players and what kind of role data and analytics play in creating the next generation of stars.

Ben Shields: You know, I want to start with a little bit of context for our listeners. Can you kind of give us a sense of what the academy does at Chelsea Football Club?

Ben Smith: So our job is to create players for our first team. That’s about understanding what talented footballers look like. So we’re going out, and we’re able to recruit players of exceptional potential as well as exceptional ability and then putting in a really strong development process to maximize their talent, so that they’re ultimately able to cope with the demands of our first team and possibly contribute towards the needs of our squad.

Ben Shields: I want to talk a bit about the talent identification piece because this is really interesting to me. At what age are you able to identify high-potential talent?

Ben Smith: We have development centers — we have 12 development centers — that are sort of within (I think it’s) an hour-and-a-half of the training base. And in each of those development centers there’s approximately 20 boys. And so those boys are typically, of course, under 7 and under 8, and they have been recruited in our local area and seen as boys that are high-potential players for their age. We train and help develop them. And those boys — there are then 20 selected to join our under-9s as registered Chelsea footballers.

Ben Shields: When you identify these players, what is your success rate in converting them into professional footballers?

Ben Smith: So when we look at the data, we kind of make a break point of approximately 12 years ago, which is... some of it is around quality of data, but most of it is really around when the academy took a strategic direction, and head of youth development Neil Bath came in and started shaping and influencing recruitment and every aspect of the academy. So prior to that stage, it’s kind of a different era. Most of our data is looking at the era that we’re currently in. And of that time frame, we’ve got 36% of boys that have ever registered for us as an under-9 have gone on to be professionals.

Ben Shields: So let’s unpack that a little bit. What is the mix between scouting the right player and then the development process? Can you talk us through kind of the ingredients here to the success rate that you’ve been able to achieve there?

Ben Smith: Sure, I’ll happily go through all of that. But it is important to say from an evidence point of view, we can’t differentiate between the quality of our recruitment and the quality of our development processes — because, you know, we’re confident we bring in high-potential individuals, and we’re confident we’ve got an exceptional development program, but you can never tell which has had the influence. We don’t know if we’re bringing in average players and we’re training them to be incredible (although we don’t feel that’s the case), or if we’re bringing in unbelievable players that would make it — irrelevant of our development process. So we can look at the quality of both of those processes. We just can’t measure the difference between those, unfortunately.

Ben Shields: Do you think you’ll be able to get to that point?

Ben Smith: No, I don’t think so, because you can’t quantify the potential of a player at 9. You can look to see: Do they have key traits which we know are vital? And, you know, what should they look like in an 8-year-old? They have ingredients that will give them the potential to be really, really good. However, then you’ve got 12 years of development in which anything can happen. Some of it is under our control; a lot of it is out of our control. That’s the randomness of the development pathway.

Ben Shields: So that’s interesting. When you talk about the identification, for instance, of an 8-year-old, how much of that process is about the experience that your talent scouts have versus perhaps the data that you have on that boy in order to understand whether or not it’s a high-potential talent?

Ben Smith: I think it’s fair to say that it mostly comes under the qualitative assessment, but it’s a structured qualitative assessment. You know, we’re not just kind of going: Tell us if he’s good or not. We’re kind of giving very defined areas of how we want our coaches to use their expert assessments and report back to us. There’s a huge amount of education that goes on with our scouts and recruitment staff so that there is very clear alignment between how we view a player: What do they need to do? What should it look like at that age? So that if different people see a player, there should be a high level of clarity and resonance between how they’re reporting back on that player. We’re not just kind of randomly going: He’s good here or he’s good there. So that quality assurance across our scouting processes is something that the recruitment department have done fantastic work on and really helps our process.

Ben Shields: So even with a qualitative assessment process, you have in place criteria (for lack of a better term) to help systematize that process across the entire organization. So that’s really helpful. OK, so the kid gets in the door. You identify the talent. Now, let’s talk about the development process. So once that 8-year-old is now part of the academy, what are some of the data-based approaches that you take to develop that player’s football skills as a kid?

Ben Smith: I would say the industry is definitely developing in this area, but I wouldn’t say it’s hugely sophisticated yet. The ability to create a data-evidence-led approach that still works operationally on the ground for coaches and for staff — I haven’t seen that in a large number of environments on a development level. I have seen it, obviously, in a huge amount of first team environments. But it’s hard to create a really sound narrative that works for 10, 11, 12 years about how a player progresses from a data point of view because of all the different things that happen. And so I think that where we’re at now is trying to create a model, again, that frames a lot of subjective expert assessment, but then you’re able to track a player over four, five, six years in how they’ve developed in these core competencies that we consider necessary in football and then evaluate how effective we’ve been in developing those areas.

Ben Shields: So can you share an example, perhaps, of a player who maybe is on the current roster and what that not only talent identification story was with that player but also how you’ve developed him to be a professional contributor on the team currently?

Ben Smith: So there’s a few boys on the first team squad at the moment. Ruben Loftus-Cheek is one of our academy boys that we’re really proud of. He’s — let’s see now — Ruben’s, I think, 20 or 21, and he joined the academy as an 8-year-old. And so he’s someone that would’ve gone through a really high-quality coaching process. He had fantastic talent. However, helping him maximize that talent and learn some of the holistic skills that go alongside technical expertise so that he can go on to be successful on our first team environment is kind of a key process that we do. Because our first team environment is — it’s not a normal environment. And if you just take a high-talent individual, but with a regular psychology, it’s not necessarily going to translate into success on our first team. You’ve got to have a very particular sort of set performance skills in terms of your mental approach and how you manage yourself to be able to even get through the door, never mind be successful and thrive and take the opportunity.

Ben Shields: That’s an interesting point about the mental side of a first team player. One of our first episodes of this show was about measuring basketball IQ. Are there data-based approaches that you take within the academy to measure and perhaps improve the IQ of a footballer?

Ben Smith: I don’t think we would look to improve the IQ of a footballer. What we work really hard on is: There’s a core set of psychological skills, the PGCs — psychological skills for developing excellence — that make up a skill set about on-pitch performance. But then we also want to develop our players holistically, to be successful off the pitch, which obviously then has an impact on the on-pitch abilities. And so there’s a football education program that starts under 9 and goes all the way through — about upskilling our players in areas that would probably be considered normal life for most kids. But we are taking these highly talented boys at an age, and to a lesser degree, to start, we’re bringing them out of normal life. And we’re putting them in a very unusual environment, it’s an exceptional environment, but it’s very unusual. And some of the things that just naturally develop as a regular kid growing up — you no longer develop in our environment because it’s so focused. So we have to make sure that our formal education program gives our boys not just the skills to thrive in our first team, but the things that you just assume everyone develops, but inadvertently they’ve missed out on by being in such a niche specialist environment.

Ben Shields: So how many current players on the Chelsea first team are products of the academy?

Ben Smith: Without meaning to be elusive, it’s a difficult question to define, because you have your 1 to 11. How many people start consistently in our 1 to 11? It depends on who the manager is. Last year, Andreas Christensen was starting consistently. This year, we’ve had a change in formation and so he’s now more often on the bench. Ruben Loftus-Cheek was out on loan; he’s now back and playing. Callum Hudson-Odoi is in certain games and doing really well but isn’t a consistent starter in the Premier League games. And then we have boys who are more like squad players, like Ethan Ampadu, who are doing tremendously well. But it’s not a black-and-white question, because the context changes so much.

Ben Shields: Right. I think that’s fair. I’d like to ask you a question about specialization. Here in the U.S., there are often narratives of players that actually play a bunch of sports and then only start to specialize maybe in high school, and then the highest talented ones make it to the pros. And you’re telling a story here, a successful one, but of the benefits of specialization. Is it possible for a professional footballer to not specialize and focus only on soccer for his entire life and still make the pros? Or is there something about soccer where you really have to start young and focus on it 100% in order to be successful?

Ben Smith: Specializations are really interesting concepts, certainly around the U.K. model. I think the answer to that question differs on where you’re growing up, because perhaps you can still be successful in that model in North America. But in the U.K. and to my knowledge Europe, that’s harder, because the children who have specialized earlier have got such an advantage. And sometimes, technically, you cannot develop that level of expertise unless you’ve started early. However, we’re not unaware of the challenges of early specialization in terms of physical competencies across loads of foundational skills — movement skills — that you don’t develop unless you’re exposed to lots of different types of movements that early specialization would typically limit. The psychological flexibility and adaptability to lots of different approaches is harder with early specialization. So our approach to specialization is not to say we shouldn’t specialize — early that is — because if you don’t specialize early, our competitors will just sign up the talent. If we say, all right, we won’t recruit until we’re 15 or 16, there’s no one left to recruit. And whoever we do recruit, there’s almost no chance that there’ll be able to progress at the rate that we need. So the approach is to accept that to be competitive, we need to recruit at these early ages and to understand the challenges of early specialization and then to proactively address those challenges with our development pathway. So we would have very specific multisport activities that all our boys are exposed to. And they would do yoga, gymnastics, badminton. We understand there are some core sports and skills that expose you to physical movements that are not the same as football but complement the football demands really well in developing really good robustness and core strength and footwork — movement patterns that really help you translate into a talent environment but also build up wider competencies of those core skills.

Ben Shields: That’s a fascinating approach actually, Ben, that even though the boys are within the academy and focused of course on professional football, that you are also exposing them to other sports as a way to help round out their development. All right — couple more questions for you. The first is regarding the data that you would like to have. And that’s what’s so fascinating about the sports industry. We’re getting new sources of data and information about athletes in the future. So if you think about the talent identification process or the development process, are there types of data that you wish you had that you don’t have today that can help the academy do their jobs better and help benefit these boys?

Ben Smith: When I think of data that would be really useful, I’ve got to think: OK, what are our core challenges? You know, I’m not someone that wants to be data indulgent for the sake of it. I want to use data as a tool to address our key challenges and problems and solve some things and make everyone’s life easier. So if I think, OK, well how can you use data to find talent more effectively? And then I think I would probably come down to: How do we create a new level (a new first level) of recruitment that covers so many more players. We’re limited at the moment by the physical number of scouts and how many matches they can watch. And you know, it’s a human limitation. If we could access data that covered all players at grassroots, we could do some really cool stuff that certainly wouldn’t replace scouts (I can’t imagine a situation where computers do any of that). However, we could start to much more sensibly, strategically, align where those scouts go to because we consider them high-talent areas or high-talent potential areas, because we’ve already got this low-level information that can help identify so our scouts can then be going to games of higher quality or high-potential quality to do their job that they do really, really well, but they’re just going to less games that are: Ah, there wasn’t anyone there. Does that make sense?

Ben Shields: It makes perfect sense. And it was actually something I was thinking about because you’ve got to scout not just your local area. I mean, you’re scouting regions all around the world. And it is seemingly a very difficult task to not miss a player. Right? I mean, you want to be able to be everywhere, but you can’t, based on some of the resources that you have. So that makes perfect sense to me.

Ben Smith: Yeah, there’s always a situation every time that a player progresses in a different club that he’s technically on our patch. You know, because it’s a highly competitive environment here where we crossover with lots and lots of clubs. And so, there’ll always be a thorough examination of: How did that one get away? Why did that person not come up on our scouting process? Often we’re able to go: Well no, they did — we’ve seen them here, here, and here. But because of, I suppose, the nature of performance and not being consistent, those times that we saw them were just unlucky or they came into trial, but they didn’t do very well in our environment. And, you know, they have developed really well in another environment. But where we go, we didn’t ever hear about them. Either those guys are really good at keeping those situations away from normal conversations — I doubt that, I just don’t think there are lots of players that we miss because we have a really, really strong scanning network who work really, really hard to make sure players don’t get missed.

Ben Shields: All right, Ben, my last question is coming back to where we started. You shared some pretty compelling data — about 36% of boys under 9 within the academy making it to first team. What other success metrics are you able to share about the work that the academy does in terms of not only identifying talent but also developing them fully?

Ben Smith: Well, I think when we look at what we’re trying to do, we’re trying to develop boys who are successful on and off the pitch. And so we have metrics about boys that go on to become professional footballers. Within that we can differentiate between our first team, Premier League, football league, or just kind of professionals in the broader game. And that’s absolutely a core aspect of what we do. And from a success point of view, over the last 12 years, it’s on a really positive trend. As we’ve developed our professional processes and improved our quality, we’ve been able to be increasingly more successful. The conversion rates that I think we have, it’s where, we say, there’s 36% have become professionals; it’s 21% are playing at a football league level or above; 11% play at a Premier League club. But those numbers are very different to the numbers advertised in mainstream media. Part of that is the fact we have an exceptionally talented group of players, but we also feel like we do a really good job of helping development. The bit that is harder to evidence is when we talk about the off-pitch success. We’re helping boys become more than just professional footballers. We haven’t managed to quantify that value yet, but we’d be really interested in that so that we can measure the success of boys outside of their professional contribution in football.

Paul Michelman: This has been Counterpoints, the sports analytics podcast from MIT Sloan Management Review.

Ben Shields: You can find us on Apple Podcasts, Google Play, Stitcher, Spotify, and wherever fine podcasts are streamed. And if you have an idea for a topic we should cover or a guest we should invite, please drop us a line at counterpoints@mit.edu.

Paul Michelman: Counterpoints is produced by Mary Dooe. Our theme music was composed by Matt Reed. Our coordinating producer is Mackenzie Wise. Our crack researcher is Jake Manashi, and our maven of marketing is Desiree Barry.