More than a decade after the "Moneyball" revolution in baseball, a new data age is dawning in the soccer universe -- and some of the early insights are already challenging century-old beliefs. Corner kicks don’t have any bearing on a game’s outcome, the numbers show. Superstars are less important than the worst players on the field. More ball possession doesn’t translate into more goals scored.
Soccer -- or football, as most of the world calls the sport -- is cautiously embracing data analytics. The trend is beginning to influence the way professional players are evaluated and choices coaches make when planning on-field tactics. Top clubs in Europe and the U.S. are hiring computer scientists and mathematicians to try to gain an edge.
“We haven’t had the a-ha ‘Moneyball’ moment yet in football,” says David Sally, co-author of "The Numbers Game: Why Everything You Know About Soccer is Wrong," which will be published this summer. But, he says, soccer epiphanies are just around the corner.
The first big step has been simply to gather new forms of soccer data. Baseball teams have gathered detailed statistics going back to the 1870s. But baseball is a game of many isolated, measurable events -- an at-bat, a fielding chance -- while soccer is more free-flowing and amorphous. Little data was ever recorded or analyzed.
But a company called Opta, founded in 1996, began collecting new kinds of information from English Premier League games and now does the same for Major League Soccer in the U.S. For the first time, a club can know how far each player runs during a game, how many tackles each defender makes, and even which two players pass most often to each other. Little by little, club management has been getting interested.
Other companies have jumped in. Prozone tracks player movements and collects and analyzes on-field data. The Castrol Performance Index uses an algorithm to create player ratings. Northwestern University’s Luis Amaral, who studies complex systems, founded a soccer data analytics firm called Chimu Solutions. It analyzed statistics and, for instance, threw into doubt the notion that the team that dominates ball possession usually wins. Chimu showed that winning teams typically hold the ball less than half the time.
Meanwhile, managers and executives of the world’s top soccer clubs have been flocking to the annual MIT Sloan Sports Analytics Conference in Boston to learn more.
One of the first applications of analytics has been in player evaluation. By crunching data, clubs are finding they can virtually scout more players around the world to find talent that might otherwise have been overlooked. Data is also changing ideas of how to build a winning team. Sally and his co-author, Chris Anderson -- both Ivy League professors -- run a consulting firm that crunches data for pro soccer teams. As Sally explains, data is showing that the relative strength of the weakest players on a team has more to do with winning than the relative strength of star players. The conclusion: spend resources on making sure every player on the field is strong, instead of splurging on a couple of superstars and filling out the roster with cheap so-so players.
The emerging frontier involves using data to plan and adjust on-field tactics. The New York Red Bulls of Major League Soccer aggressively adopted data. David Lee is the team’s “performance analyst.” He sorts data about opponents and discovers tendencies.
“It can open the coaches’ eyes to look for things the data is showing,” he says. Some of the insights described in Sally and Anderson’s in "The Numbers Game" might yet deeply effect soccer coaching: Now that data shows that corner kicks don’t alter the outcome of the game, teams might try something different on the field when they earn a corner kick.
In coming years, Sally says, soccer data analysts are expected to apply network theory to the burgeoning reams of data about the passes players make and who they connect with. That should lead to fascinating discoveries about passing networks -- which groups of players have the most effective connections and why. It could change the way managers think about how to assemble a winning team.
There’s a reason, though, that soccer hasn’t yet seen the kind of earthshaking data breakthrough that changed baseballs Oakland A’s, as dramatized in the “Moneyball” movie. Data is still new to soccer. A lot of traditionalists don’t like it. Teams that have embraced data, like the Red Bulls or Manchester City in the United Kingdom, are still trying to figure out how to best use it.
“Football went from no data to huge databases,” Sally says. Soccer clubs haven’t had the long history of statistics and data that were inherent to baseball. “There’s still a disconnect between the technology and the ability inside clubs to handle it,” he says.
And yet, as always happens, the technology and the insights from the data will improve and grow more valuable. Younger coaches, reared in the Internet age, will understand what the data can do. Sooner or later, like Billy Beane in "Moneyball," one of those coaches will use data to change the sport forever.
The contents or opinions in this feature are independent and do not necessarily represent the views of Cisco. They are offered in an effort to encourage continuing conversations on a broad range of innovative technology subjects. We welcome your comments and engagement.
We welcome the re-use, republication, and distribution of "The Network" content. Please credit us with the following information: Used with the permission of http://thenetwork.cisco.com/.