Moneyball in football?

Part of the story of Moneyball is the clash between old-school, instinctive, subjective analysts and their culture and the statistical, more-objective analysts. It's interesting to see the same thing is happening in football. SFGate has a story about the 49ers version of Paul DePodesta and his clashes with old-school football folks.  I think it's clear that moneyfootball isn't "working" yet in 49er land, but it's unclear whether any system could have saved the team and its salary cap woes and generally bad personnel decisions.

One wonders whether baseball is inherently different from the other major American team sports to make it susceptible to statistical analysis.  Here are some thoughts about why baseball is different.

  • Individual contributions are generally much more clear and discrete. Note that the places where it isn't (e.g. defense) are the hardest to analyze.
  • The game is mostly time-discrete. You can get a lot of information by treating baseball as a series of discrete events ignoring their length and sequence. In particular, you can usually ignore fatigue, and ball and field changes. Again, the places where you can't (e.g. pitcher fatigue) are the hardest to analyze.
  • Emotions play less of a role. Baseball is a finesse game. Being stronger, willingness to take more pain, being faster, getting angry and emotional do not immediately translate into better performance. I would argue that in football, this helps the performance of everyone but the quarterback and kicker. In baseball, many analysts like to ignore the role of emotion (e.g. the 'no-clutch-ability-exists analysis').

  • As a result, I would argue that moneyfootball is going to be much harder to sort out than baseball. A system is easier to understand when the parts are more independent, and thus can be examined in isolation. In a game like football, the parts are surprisingly correlated. If the QB gets sacked, is it because he was slow, or a lineman missed a block, or the running back didn't pick up the blitz, or the defensive end just made a spectacular move, or the defense felt comfortable blitzing because Randy Moss didn't feel like playing that down?

Will any other sport be analyzed in the same sabermetric detail with success?

  • Football. Seems hard. Lots of coordination and emotion involved.
  • Basketball. There is more isolation possible than football, and there are fewer people. But the best teams do have a lot of subtle coordination, and emotion does seem to be a factor.
  • Hockey. A sport I once loved, which has ceased to exist at the top level. Again, lots of coordination and emotion involved, but fewer players than football. Statistically, they do try to recognize total contribution through +/- and giving an Assist to the pass that makes the pass that makes the goal.
  • Soccer. I don't know. It seems like there is coordination, but a lot of individual matchups as well, especially since the field is so darned big.
  • Lacrosse. Hey, I just saw a sign that there is a professional league! Lacrosse is probably as analyzable as hockey.
  • Cricket. I know it's not American, but it's the natural choice after baseball, though time is more of a factor, e.g. batter fatigue is more of an issue when you are batting for six hours, and the ball and pitch changes.
  • Halo. Hey, when our kids are adults, I bet professional computer gaming will be as popular as soccer is now.