# Four Up, Four Down

On October 3, at the start of the baseball playoffs, Bruce Bukiet, an associate professor of Mathematical Sciences at the New Jersey Institute of Technology, ran his mathematical model for baseball games and issued his predictions of the winners: the Yankees, Twins, Dodgers and Padres.

By October 9, all four teams had lost.

What are the odds?

(According to strict probabilities, about 15 to 1.)

Bukiet offered an explanation: “You could say the probabilities weren’t wrong — the teams just didn’t perform as they were supposed to.”

Hmm… that is one way to look at it.

Bukiet has been conducting probability models of baseball games since 1987, when he read an article on mathematical modeling for tennis and wondered if it might work for baseball. He published an article on the subject in 1997, in the math journal Operations Research, and has delivered several talks on the subject. For the past six years, he has posted a Web site that makes predictions for how one should bet on baseball games — not that he encourages gambling, mind you. He insists that his gambling page is only for academic interest — betting on baseball, he notes, is illegal — and besides, if he did gamble, “My wife… well, I’d be in the doghouse.”

Aside from baseball, the professor has studied detonation theory at Los Alamos, modeled the mechanics of diseased hearts and helped the Orthodox Union develop an inspection procedure to certify vegetables as bug-free. So how did the professor get it so wrong?

One reason — which Bukiet readily acknowledges — is that the probability theories omit the human factor. That Yankee captain and clutch hitter Derek Jeter would have a .500 batting average for the playoff series, while tormented Yankee superstar Alex Rodriguez hit a miniscule .071 wasn’t mathematically predictable — but then again, it wasn’t exactly a surprise to baseball fans, either.

The other reason is that a playoff series is, in a sense, a terrible way to measure which is the best team. With such a short series, the results depend on lucky bounces, nervous miscues, blown calls and journeyman ballplayers who make unexpected clutch plays. These tricks of fate and human passion play hell with probability theories. But it makes them heaven to watch.

And at least when it came to the Dodgers, the professor was happy to be wrong — he’s a Mets fan.