September 25, 2022

It was 2015 and Antonio Gagliardi was doing what he was paid to do: thinking about football. Something about the way we talk about football didn’t quite sit well with the Italian Football Federation analyst, and then a light bulb went on: “Roles shouldn’t be defined by position, but rather by function.”

– O’Hanlon: Could Haaland and Nunez signings lead to revival of centre-forward role?
– Stream ESPN FC Daily on ESPN+ (US only)
– You don’t have ESPN? Access immediately

In other words, it makes little sense to talk about a right-back when different players in that position are called upon to do different things. What Benjamin Pavard is doing for Bayern is very different from what Trent Alexander-Arnold is doing for Liverpool. Or take Chelsea’s Jorginho and Real Madrid’s Casemiro, both deep central midfielders, right? Yet while their heatmaps may be similar, what they actually do on the ground is entirely different.

Gagliardi has teamed up with analytics firm Soccerment to come up with an entirely different way of categorizing players, based on function. Using Opta data from the past five seasons and some proprietary algorithms, they looked at Europe’s five Big Five leagues and came up with 13 ‘clusters’ to define outfield players by their contributions or, better yet, their attempted contributions. . For example, a loose player who repeatedly attempts to go one-on-one and beat an opponent rather than, say, crossing or breaking inside to build play is defined in a certain way , whether he succeeds or not.

You might be wondering if there isn’t a bit of a “chicken and egg” dilemma here. In other words, have they defined a type of player in terms of, for example, X many cross attempts, Y many dribbles and Z many shots? Or did they just let the algorithm do its thing and sort players by statistical category, grouping natural clusters?

They did the latter, to help remove any bias they might have had about what a function should look like. And they ended up with 13 pretty distinct groups based on a player’s tendencies. All that remained was to name the clusters (and here the naming may seem a bit hokey, but bear with them, this is their first attempt).

There are ball stoppers (Tottenham’s Cristian Romero), construction starters (Liverpool’s Virgil Van Dijk) and frontline breakers (Napoli’s Kalidou Koulibaly). You also have wide controllers (Bayern’s Pavard), wide makers (Achraf Hakimi of Paris St Germain) and chance makers (Kevin De Bruyne of Manchester City).

There are ball stealers (Eduardo Camavinga of Real Madrid), construction managers (Thiago Alcantara of Liverpool), box-to-box raiders (Nicolo Barella of Inter) and individual explorers (Vinicius Junior of Real Madrid). Up front you’ll find mobile finishers (Mohamed Salah of Liverpool), versatile finishers (Robert Lewandowski of Bayern) and target men (Sebastian Haller of Borussia Dortmund).

And, of course, there are hybrids. Exiting some players places them in multiple groups. PSG’s Kylian Mbappe ranks as both mobile finishers and one-on-one explorers. Joao Cancelo ranks high in both the wide maker and chance maker categories, which makes sense to anyone who has watched City play and the Portuguese full-back (which he might have been ranked there). has a generation) entering and serving assists. Real Madrid’s David Alaba is perhaps the most hybrid of them all: he appears in five different groups.

A side effect of the consolidation project reveals what many already suspected: certain functions are much more prevalent in successful teams. The best teams tend to have more builders (both in the lucky and loose categories), more players who help build, and fewer ball blockers and ball thieves. Part of that is down to design. They have more balls and they need more players who can do things with them. However, some of this is likely a reflection of the resource imbalance in the game, with the top clubs attracting (some might say hoarding) the most talent.

The creators of the clustering project admit that there is huge room for improvement. For starters, there is more advanced and detailed data, both in terms of event data, pressing data and tracking. And, obviously, what a player does on the pitch is influenced by other factors, who his teammates are, how his team plays, what instructions a manager gives his team. They see their work as a starting point on which to build.

But the concept behind it is obvious. And it shows how the game is evolving and, perhaps, has moved beyond traditional nomenclature. Just like, in many ways, basketball. We used to have a point guard who dribbled and passed, a goalkeeper who shot, a small forward who drove to the basket, a more powerful forward who trailed close to the paint, while the center was the biggest in the game. team and worried about rebounding, blocking shots and posting. Much of that is now gone, and the focus is on individual skills rather than defined positions.

Basketball is obviously a more fluid game, but football has its own fluidity. Is it logical to say that Liverpool play in 4-3-3 when, in possession of the ball, Alexander-Arnold and Andy Robertson are often more advanced than the midfielders? Or think of Sergio Busquets at Barcelona. It’s also a nominal 4-3-3, but when they have the ball the full-backs push up, the centre-backs split and Busquets slips between them, turning him into a de facto back three.

We already recognize, often unwittingly, different skill sets. Raheem Sterling won’t play on the left wing like Sadio Mane does, let alone Cristiano Ronaldo, when they stick him there. Most understand this. Consider this a possible next evolutionary step in our understanding of the game.