Wednesday, March 2, 2011

Tennis? Anyone? Anyone? Beuller?

I have elsewhere noted that my alma mater, in addition to being the primary bastion of defense against the advance of Illini-led communism, was also a egg-headed liberal arts school that taught us about old poetry and literature and stuff.

But not all of us were tweedy types who spent hours trying to figure out how to create the perfect collegiate Shakespearean sonnet (hint: Fourteen lines, a-b-a-b, c-d-c-d, e-f-e-f, g-g rhyme structure and includes references to pizza, beer and "Louie, Louie"). Some of us were the classic kind of nerds that thought of just about everything in terms of scientific experiments and mathematical formulae. And at least one of us was the kind of language snob who insists on using proper Latinate plural forms of certain words.

Anyway, one of those classic nerds (we actually called them "weens," which is another story for another time) turned his attention to the sport of tennis. Filippo Radicchi, a researcher at Northwestern and tennis fan, decided to apply something called network analysis to try to rank male tennis players. According to the story, he used an algorithm, or special kind of equation, like the one Google uses to rank web pages. The algorithm increased a player's ranking, for example, if he had more wins over other higher-ranking players. Thus, a single victory over say, Björn Borg, would be of more value than a bunch of wins over Dominik Hrbaty. Or me.

And so, according to the algorithm, the greatest player of "all time" is Jimmy Connors. "All time" is in quotes because Radicchi's study only included players with one Association of Tennis Professionals (ATP) match played between 1968 and 2010. Radicchi points out that Connors rates so high because his long career (20-plus years of playing tournament-level tennis) gave him the chance to earn victories over several top opponents, and that current players who might very well have bested Connors in his prime are ranked lower because they're still playing and accumulating value in the algorithm.

On the one hand, serious nerd-ism, to be sure. On the other hand, Radicchi is my favorite kind of specialist -- someone who takes his or her specific field of study and applies it to an area he or she enjoys in the real world, both just for fun and to try to learn something.

No comments: