Thursday, September 06, 2007

Team salaries and performance.

Check this out. Via Kottke.

1 comment:

B said...

That interesting, although I found myself wishing that they'd set up their methodology a little differently. In effect, because the Sox have the second highest payroll in the league, they were going to be found wanting (red line) unless they had the best record in the majors. The chart is glitsy, but the underlying observation isn't all that illuminating.

It would have been more interesting if (for example) they calculated the average payroll cost per win for the entire majors (sum of all team payrolls divided by sum of all team wins (presumably 82*30=2460)). Then calculate, for each team, how much they spent per win at each point in the season and show: were they getting their monies worth? The Sox, even with the second-highest payroll, might be above-average in wins-per-buck or they might not.

My hypothesis would be not; if only because I think that as teams get better, the incremental cost of each additional win goes up. I.e. it's cheaper (on average) to get from 50 to 60 wins then to get from 90 to 100. The former often requires something like replacing a couple below-average players with above-average players, who are fairly common and not that much money. The later often requires dropping big bucks on someone like Dice-K or Schilling. This is just a hypothesis, and there have certainly been exceptions (the 'moneyball' Oakland A's come to mind). It would be interesting to see what the data said.