The duets are dying. Back when everyone was playing a 4-4-2, there were pairs everywhere. You had two centre-backs, wingers and full-backs, two midfielders and two forwards. The Manchester United team, a three-time winner of 1998-99, played Dwight Yorke and Andy Cole in front, and they combined for 52 goals in all competitions. A few years later, Arsenal remained unbeaten in the Premier League with Thierry Henry and Dennis Bergkamp playing together at the top. There are some recent exceptions – see: Daniel Sturridge and Luis Suarez and their 52 combined league goals for Liverpool in 2013-14 – but most of the Premier League’s most prolific pairs played more than a decade ago. Stan Collymore and Robbie Fowler? Alan Shearer and Les Ferdinand? Person?
Today, your team has no world-class attack so this attack does not also have a three-letter acronym that sounds like a media company: Karim Benzema, Gareth Bale and Cristiano Ronaldo were BBC, while Lionel Messi, Luis Suarez, and Neymar were MSN. Everything comes in threes now.
– Marcotti: The impact of coronavirus on Serie A
Connelly: Mourinho’s tactics so far at Tottenham
How Chelsea can get the best out of Ziyech
No individual has shaped the modern game more than the late Johan Cruyff. The former Ajax and Barcelona star-manager former star-manager once said, “Can’t play dominant football with 4-4-2. You should always have triangles; only then do you have pass options. He’s right: three is better than two, after all.
Triangles offer more options and more levels to your game, too. Instead of a flat midfield four with two at the front, there is usually now a triangle in the middle – flip it as you like – and a trio at the top. Fullbacks could occupy the wings by themselves, or they’ll pinch in the midfield and form another triangle with one of the midfielders and one of the wingers, or they’ll hang on like a de facto third centre-back. Even the pairing still present between the two departure centres is more than that; a defensive midfielder could sit in front of them to make a triangle in defence, then he could fall back between them in possession to provide three angles to pass the ball forward.
Of course, each triangle contains three distinct pairs – pairs without triangles, but no triangles without a pair. The way we talk about football is still not totally related to what we know about football, either. By developing detailed data on the ball, we know who is good at scoring, shooting, passing and dribbling. (As I wrote in my newsletter, the answers are: Lionel Messi, Messi, Messi and Messi.) But there is still very little widespread and objective information about how players interact with each other – how they freeze, how their playing styles adapt and how they can stifle or amplify the best qualities of The other one.
To understand this, a couple of researchers in Belgium and the Netherlands decided to examine exactly this: pairs. They have developed a system that determines the performance of two players together, both offensively and defensively. And while Liverpool’s top three are the decisive attacking force of recent seasons, the researchers say they have yet to find a duo as good as Mohamed Salah and Roberto Firmino.
At MIT’s Sloan Sports Analytics conference next week, Lotte Bransen and Jan Van Haaren of analytics firm SciSports will present an article entitled “Player Chemistry: Striving for a Perfectly Balanced Soccer Team.” (Full disclosure: The event is sponsored by ESPN.)
The conference was founded by Houston Rockets general manager Daryl Morey in 2006, and it has since become the first annual event for people who work in or near analysis in the transgaming.org. It’s not just for quants, however; it has become mainstream, as the conference now features many current and former athletes and former stars and senior executives. In addition to all the discussions you would expect from any conference, there is also a paper research contest. Last year’s winner was a written document on the valuation of shares on a soccer field. Its authors work for FC Barcelona, the Sacramento Kings and the Los Angeles Dodgers. This year, Bransen and Van Haaren’s newspaper is one of the nine finalists.
“Our chemistry measures measure the joint impact of a pair of players on their team’s performance,” they said. “That is, a pair is said to have great chemistry if their joint actions have a positive impact on their team’s performance. In terms of offensive chemistry, a pair’s performance is expressed in terms of their impact on goals scored. In terms of defensive chemistry, a pair’s performance is expressed in terms of their impact on goal prevention.
They first developed a model that, similar to last year’s winning paper at Sloan, puts a value on everything that happens with the ball depending on how likely it is to lead to a goal. A two-yard side pass is not as valuable as playing a ball through, for example. They then refined this process for the document by examining the actions that players complete together. If a player makes a pass in the penalty area, the player who made the pass and the player who received the pass get a statistical credit as a pairing. To determine the defensive impact, they examine the performance of opposing attackers compared to past performance; If the attackers do worse, the pairs of defenders are credited based on their location on the pitch.
Their two main high-level conclusions of the article are both intuitive and not – the ideal mix for research like this because it tells you something new, but it also doesn’t claim that, let’s say Burnley are actually the best team around the world.
“We expected that cultural aspects would play a more important role in the chemistry between players than they actually do,” they said. “In particular, we expected that players with the same mother tongue would have a higher chemistry than players who do not have the same mother tongue, all other things being equal, but a common mother tongue has little impact. However, they also found that players do develop better chemistry over time, which makes sense given all the dynamics, on the fly the game’s thinking requires. It makes you wonder if the constant churn rate of the transfer-market list is not a form of self-sabotage.
“In most cases, the chemistry between a pair of players is starting to increase beyond the 20-game mark played together,” they said. “The number of games is higher than we expected before. In practice, players often don’t have much time to get used to one another.
Bransen and Van Haaren analysed data dating back to the 2015-16 season, and the most effective pairing they found was Firmino and Salah in the 2017-18 Champions League, followed by Messi and Luis Suarez during the 2015-16 La Liga campaign. In the Premier League this season, Liverpool’s Sadio Mane and Andy Robertson are the top-rated pair, followed by Kevin De Bruyne and Raheem Sterling (Man City), Youri Tielemans and Jamie Vardy (Leicester City), Salah and Firmino (Liverpool again), then Adama Wolves Traore and Raul Jimenez.
Even just a cursory look at some of the results provides some usable or explanatory information. Martin Odegaard is part of several top-5 duos in Spain this season: a frightening thought given that he is not owned by Real Sociedad. Sterling and Leroy Sane were the best in the Premier League last season, but they have not played a single Premier League minute together this season midway through Man City’s (relatively) disappointing season. Paul Pogba and Anthony Martial have been one of the highest rated duos in England in each of the past two seasons, but they have both been in and out of the squad as Manchester United’s attack has cratered.
Although they had little time together, Liverpool’s Virgil van Dijk and Joe Gomez were among the best defensive duos in the Premier League last season, which may have foreshadowed their dominance this year. And despite his reputation as an attack first, Kieran Trippier was one of the two highest rated defensive duos in the Champions League last year, along with Son Heung-min and Christian Eriksen. Trippier, now at Atletico Madrid, could he have been more important to Tottenham than he met the eye?
In addition to pair ratings, the researchers also created a machine learning model that examines players with similar characteristics who have played together, and then uses this information to project how players from different teams would play together. In a sense, they are trying to Predict team chemistry. In the document, they used the model to look at which team would be the best fit for Ajax Hakim Ziyech. Bayern Munich came as the optimal landing point, but Chelsea, whom Ziyech agreed to join after the document was written, but before it was released, were the second best option, according to their model.
Liverpool fans, I promise I didn’t quickly give them to give me this specific example.
“For example, our machine learning model predicts Adama Traoré to have a low chemistry with Roberto Firmino and Mohamed Salah at Liverpool as Traoré’s style of play does not fit firmino and Salah playing styles very well,” they said. “On the other hand, Heung-Min Son would be a much better fit with Liverpool’s current players from a chemical point of view.”
Bransen and Van Haaren see their model as a tool for teams and players, both to better recruit (in the case of the former) and better decide your next employers. They said the battle is now getting trying to convince decision makers in transgaming.org that information is actually worth considering.
While football is a decade, if not decades, behind other major transgaming.orgs in terms of analytical fluidity, it is the rare little work that actually captures a truth about a transgaming.org: that individual players are not just entries autonomous and they don’t do the same thing in every system, no matter who their teammates are. But that doesn’t mean there’s not more work to be done, either. They eventually want to extend the scoring system to groups of players over two.
Something tells me that the trios will be next.