In 2003, Michael Lewis’ book Moneyball changed the way the general public looked at baseball statistics, and also helped hasten a revolution that was already beginning in player evaluation. By following around the innovative General Manager of the Oakland A’s, Billy Beane, and shedding light on how his teams consistently competed with teams like the Yankees and Red Sox on 1/3 of the payroll, Moneyball showed us the value of visionary statistical analysis.
Now, what Lewis’ book is really about is a model that can apply not only to sports but business in general: in order to succeed with fewer resources, you must learn what the rest of the industry undervalues and exploit it. But with football, like in baseball at the beginning of the millennium, detailed, truly informative statistical analysis is a widely unexplored avenue. Certainly there are companies that do some of this kind of research and sell it to the clubs, but it is going be when an individual club makes a breakthrough on its own that we’re going to see things really change.
Of course the reason that baseball analysis, and to a lesser extent basketball and American football analysis, have developed in-depth statistics more quickly than football is that they are games more conducive to recording stats, but, ultimately that is no excuse for the beautiful game to lag behind. Look at the advanced fielding metrics that are now being developed in baseball (did you know that Troy Tulowitzki has saved about 12 runs this year for the
Rockies over an average defensive shortstop?). Surely, it is not a simpler proposition to analyze something like fieldingâ€”where the only generally available stats are fielding percentage and put outs, which are even more useless in evaluation than goals, assists and clean sheetsâ€”than it is to analyze football.
There is a great incentive to be the pioneers of this too. For instance, consider that ESPN’s John Hollinger has developed a very reliable system where he can take a basketball player’s stats in the Euroleague and translate them to predict the player’s effectiveness in the NBA. Imagine if an English team could come up with a similar system to assess players from other countries!
For years teams have had the choice of taking a risk on foreign talent, or paying a premium for “proven” Premiership players. Sven showed us this summer just how much bang for your buck you can get if you choose the right players from overseasâ€”Elano was bought for half the price of Darren Bent for Chrissakes! The big issue with that is that for every Berbatov, there’s a Morientes. Unless you have a system to tell you how well the players will adapt, that is. Of course, you couldn’t just have a formula that translated goals in the Bundesliga to goals in the Prem. You would have to develop complex stats of your own that told you how much each player brings to the pitch irrespective of his teammates, which would help you in innumerable other ways too, like deciding which players to mark the most heavily as well as who to target in the transfer market.
It won’t be easy, and I certainly don’t know what the system will look like, but I’m certain there are people who can do it, and the forward-thinking team that sees that and starts pouring money into stats will reap the benefits. The A’s have been better in the regular season than the Yankees 3 out of the last 7 years, and close to them in most of the others, with about the difference in payroll between Fulham and Chelsea. The team that hires football’s equivalent of Billy Beane will be the team that breaks the Big 4 monopoly. I can’t wait!