Has Orlando really gone crazy over soccer?
Sure, its government leaders have been cheerleading the sport a lot lately as they try to attract a major-league team to the city. But it’s hard to ignore that Orlando City Soccer recently saw its biggest regular-season home crowd as the Lions beat the Seattle Sounders Reserves earlier this week.
Now the sport’s fan base, particularly here in Orlando, might be surprised to learn that a new scientific understanding of the game’s dynamics is being advanced by none other than Disney-related researchers. And what the Disney Research scientists have uncovered just might change the way road games are coached.
- 2015 events at Walt Disney World
- Pictures: The rides of Walt Disney World
- More Disney coverage
- Pictures: Events at Walt Disney World Resort
- Pictures: Disney Hollywood Studios sorcerer's hat through the years
- Pictures: Walt Disney World Marathon 2015
See more photos »
- Disney video
- David Beckham
In a pithily titled piece of light reading called “Assessing Team Strategy Using Spatiotemporal Data,” researchers at Disney Research, Pittsburgh, have found that the time-honored adage about wins being easier at home than on the road just might have more to do with the coaching decisions made by away teams rather than the perceived bias of the home-team referees. Thus, as soccer teams hit the road and play less aggressively, their defensive-oriented strategy reduces the likelihood of winning.
How did the Disney Research scientists get to this conjecture?
Well, it wasn’t a review of Pele game tape or an analysis of David Beckham affair rumors published in British tabloids.
Instead, researchers Patrick Lucey, Joe Roth, Peter Carr and Iain Matthews – plus Dean Oliver, ESPN director of production analytics – applied artificial-intelligence observations of data from all 380 games of the 2010-2011 season of a 20-team professional soccer league season. Anyone familiar with “Moneyball,” the film and book about how analytical, evidence-based observations of game and player data aided a Major League Baseball team’s competitive efforts, will recognize the Disney researcher’s underlying efforts.
And, according to a press release about the research, what’s more unique is that the soccer data wasn’t just a listing of player positions during the games. Instead, the researchers used so-called “ball action data” – time-coded info about the ball in play. For the season the researchers studied, that amounted to 760,000 ball-action notations.
Then the researchers used software to infer the position and possession of the ball, the release stated, and finally the research team divided the field to count how many times the ball was in a specific area and for how long.
This info, in turn, gave the researchers a model for a team’s behavior in certain parts of the field. Then comes the moneyball shot of insight from their research. As stated in the release from Disney Research: “At home, the team had the ball in its opponents' defensive third more often – and thus had more shots on goal – than when on the road and played a more defensive, counterattacking style. It was a pattern the researchers discovered held true for nearly every team.”
This came, the researchers said, despite the fact that shooting and passing percentages were similar for both home and visiting teams.
The team reported its findings at this month’s Conference on Knowledge Discovery and Data Mining in Chicago.
You can read the full press release about the Disney Research efforts here.