WORLD JUNE 14, 2010
There's a new master narrative for the history of sports. And it goes like this: We have only just begun to emerge from the superstitious dark ages, where coaches clung to folk wisdom, into the enlightened world of data. The Copernicus and Galileo of sports’ scientific revolution are the statistician Bill James and the Oakland Athletics general manager Billy Beane. They have reduced the mysteries of baseball prowess to a sequence of previously ignored data, the new discipline of sabermetrics, uncovering the hidden formulas for understanding the game.
The fact that Beane--and not a player for the ages--is the great hero of this past decade’s most influential sports book, Moneyball, perfectly captures our era. Our society has grown devoted to what The New York Times Magazine recently termed “the data-driven life.” Thanks to technological advances, we now swim in ever-deeper pools of stats. We have the capacity to measure every human twitch and an increasingly confident sense of our ability to deploy algorithms and regression analyses to make rational decisions, on everything from choosing the perfect neighborhood to the perfect mate. Yet, for all our faith in this statistical revolution and all the benefits it has yielded in certain fields, there’s one realm, or at least one sport, that remains stubbornly, beautifully immune.
Marxism attempted to discern the scientific basis for history--and it was the Soviet Union that first attempted to unlock the scientific basis for soccer. The grand theoretician of this approach was a Ukrainian named Valeri Lobanovsky. In high school, before signing up for the local soccer club, he had displayed mathematical acumen, winning a gold medal in the subject--and went on to study heat engineering at the Kiev Polytechnic Institute, even as he played professionally. It was at the Polytechnic in the early ’60s that he arrived at his epiphany that the cybernetic techniques and other methodology he encountered in the classroom could revolutionize his sport.
In the schematic he invented, the game could be broken into 22 component tasks, what he called “actions” and “coalition actions.” He would dispatch analysts to tally each of these actions--forward passes, backward passes, tackles--and then feed these numbers into a computer to evaluate his players. He also recruited a young academic, Professor Anatoly Zelentsov, to run the “laboratory” at their club, Dynamo Kiev. As Zelentsov explained his task: “In my laboratory, we evaluate the functional readiness of players and how their potential can best be realized. And we influence players in a natural way--we form them following scientific recommendations.”
It was pseudoscience. But even pseudoscientists have great insights. Lobanovsky-coached teams used well-coordinated pressing and superior stamina to swarm opposing players with the ball. During the 1970s, Lobanovsky’s teams twice won the European Cup-Winners’ Cup. But he also ultimately ruined Ukrainian soccer--or, at least, his less sophisticated, less charismatic disciples did. The national game is now characterized by players running and tackling frenetically. They rush to complete the actions that the computers reward, no matter their ultimate efficacy. It is ugly, and, judging by the Ukrainian performance on the international stage, not terribly effective.
Outside the Soviet Union, Lobanovsky’s lab-based approach never caught on. But the quest to find the mathematical underpinnings for soccer continues. There are powerful computer programs--Opta, ProZone--that track how far players run in the course of a game, how many times they touch the ball, and how many times a player is involved in an attack leading to a shot. While these analytics have proved useful to scouting and buying talent, they haven’t reshaped the game in the same way that sabermetrics have transformed baseball. Or, as Simon Kuper and Stefan Szymanski put it in their book Soccernomics, “Now soccer is due its own [Bill] Jamesian revolution.”
Yet, after many years of academic papers, the game is nowhere close to that kind of revolutionary state. In part, this is a product of soccer’s cultural hostility to data, particularly plied in the service of making rational purchases of talent. (The biggest clubs are often run by oligarchs, either Russian or Emirati, who don’t care about managing the financial risks of their player purchases; or, in the case of FC Barcelona and Real Madrid, by presidents who are democratically elected by fans and pander to the masses.) But the problem isn’t just the culture of the game; it’s the game itself. Soccer has never really had box scores or batting averages or any of the rudimentary statistics that James and Beane have rebelled against. On a profound level, soccer is immune to rigorous statistical analysis, at least compared to baseball and basketball. There is no single controlled variable--like a batter standing at home plate--that can form the building block for good analysis. And the flow of the game is too anarchic, with constant change of possession, to be broken into a series of discrete moments, where actions can be judged to have clear cause and effect.
And, even when the game does yield data about a player, it’s hard to invest much in it. More than most other sports, the performance of an individual player is highly dependent on the team around him--and on his coach. A great basketball star like LeBron James can flourish under the tutelage of a mediocre coach. But, in soccer, a poorly structured team can squelch even the greatest talent. This past season, when the Argentine Lionel Messi played with his club, Barcelona, he ran all over opponents. However, under his national team coach, Diego Maradona, he hardly ever scores. Maradona’s formations simply can’t construct room for Messi to run at opposing defenders, without him getting quickly blanketed--and they will likely account for yet another Argentinean disaster at the World Cup.
The data revolution in sports is, at its core, technocratic. Michael Lewis once wrote that its adherents view sports “not just as a game to be played but as a problem to be solved.” Soccer, by contrast, is a world governed by ideology. The analytical vacuum left by the paucity of meaningful data has been filled by great debates about competing tactical systems. Soccer’s history is a struggle among these systems, each claiming not just effectiveness but moral superiority. During the 1960s, the Italians adapted catenaccio, or the lockdown, a method for arraying players in a numbing defense--a tactic challenged by Dutch Total Football, a free-flowing system that grew out of Amsterdam’s late-’60s liberalism. Over time, these debates replay themselves: How much beauty should be sacrificed for the sake of efficiency? Or, more granularly, how much freedom should defenders have to join the attack? Does the best attack feature one forward or two or three?
Billy Beane has become a fan of the beautiful game and even convinced the owners of the Oakland A’s to purchase the soccer club in San Jose. Someday, he may help formulate the empirical answer to these primary questions. But, until he does, I will enjoy the fact that success in the game is a product not just of individual performances, but of theoretical opinions. Soccer hasn’t been swallowed by social science, but remains essentially a branch of the humanities--a home to old-fashioned intellectualism. The game is not a problem to be solved. For now, at least, it is a blessedly endless argument.
Franklin Foer is the editor of The New Republic.