Thu Sep 23 02:39pm EDT
What's the most important statistic in tennis? The answer may surprise you. (Or it may not. It's not like it's "second-serve points hit into net" or anything.)
But before we get to that, a thought about tennis and stats. Over the past few months, I've been hearing and reading some laments of tennis fans and writers that the statistical revolution which has hit other sports has yet to make its way to our game.
Baseball has long been considered the sport best suited to statistical analysis. The head-to-head matchup between pitcher and batter is one of a number of factors that helps provide a bevy of data which can be analyzed to produce mathematically complicted stats that help define a player's true worth. For decades, baseball fans had to rely on largely arbitrary numbers like RBIs, runs, batting average and wins. Now WAR and OPS+ give better representations of individual talent.
Even football has intelligent stat analysis, despite the fact that the team nature of the game should make it impervious to such things.
Through all this, tennis has been left out. We're still looking at the same numbers we did 30 years ago. First-serve percentage and break points and serve speed. It's 2010 but analysis of tennis is stuck in the pre-computer days.
The sport should be perfect for smart, reasoned mathematical research. Its head-to-head format leaves even less room for variables than in baseball, where the speed of a center fielder or range of a shortstop puts some question marks on data. Tennis is one player against another. The ball is either in or its out. In the net or not. Backhands, forehands, overheads, volleys, break points, game points, match points, holding serve, aces, length of rallies, first set vs. third set -- I could go on and on. There's a goldmine of data waiting to be tapped and decoded.
This week, the Harvard Sports Analysis Collective, a student organization dedicated to the analysis of sports strategy, started to do just that. The group posted findings about which tennis stats best predict ATP rankings. They looked at five variables -- aces per game, service-hold percentage, return-points-won percentage, break-points-saved percentage and break-points-won percentage -- and compared them to rankings points to look for the best correlation. Their findings were marginally surprising.
According to the HSAC, service holds are the best predictor of who will top the rankings. Most of us could have predicted that; win your serve and it's difficult to lose the match. What was a bit surprising is that the percentage of break points won isn't the best predictor of rankings success.
It is certainly surprising that player performances in break point situations are insignificant determiners of world ranking. If you were to look at who currently leads the ATP tour early on this season in break point stats, names such as Rafael Nadal, Roger Federer, and Andy Roddick would frequent the list. Break points are considered the "clutch" moments in tennis, and the players that win them often attain insurmountable advantages in matches. One would assume that winning or fighting off break points (representative of break points won % and break points saved %) would result in match wins which would increase ranking points.
If you want to read the mathematical jargon stating why, let the HSAC explain it you.
The study isn't perfect, as its authors would be the first to admit. There's no causation, most of the stats are dependent on the quality of opponent, and percentages don't always tell the whole story.
For instance, let's say Player A and Player B both break their opponents' serves. Player A did so after getting out to a 40-0 lead, but then having to hold on to win in a second deuce. Player B got out to the same 40-0 lead and won on his first break point. Player A's break-point percentage is 20 percent, Player B's is 100 percent. The result was the same.
On a similar note, in the U.S. Open final, Novak Djokovic was 3-4 (75 percent) on break points. Rafael Nadal was 6-23 (23 percent). Which would you rather have?
But these are quibbles that can be ironed out down the road by people much smarter than me. I'm just glad there's a growing dialogue about this. It would benefit fans (who's clutch, who's not, does clutch even exist?) and players (which player can't hit a backhand on break point, is coming to net more effective on a second serve on clay courts?) alike.
The revolution is coming to tennis. Get out your calculators.