Friday, September 11, 2009

Sabermetrics and Tennis

This might be a dumb question, but has anyone ever done the math on whether or not a tennis player would be better off hitting nothing but first serves, rather than hitting a lesser second serve?

Just as an example, let's take a player who hits 68% of his first serves in and wins 85% of those points. (My guess is that this is somewhere near the average.) Let's also (charitably) suppose that he hits 100% of his second serves in, winning 50% of those points. Over the course of 100 points, that player will win 73.8 of the 100 points on his serve.

If he hits nothing but first serves, sure, he'll have a lot of double faults (10.24 of them), but he'll wind up winning 76.3 points, giving him a 2.5 point advantage.

Has some Bill James of tennis already gamed this all out?

1 comment:

codeandculture said...

there is tennis strategy work in game theory going back to the 70s. in the 90s behavioral economists (thaler?) found it very amusing that tennis players generally don't adhere to the strategies laid out in game theory. unfortunately i can't remember any of the cites, but the literature is out there.