QR Code
September 22 • 03:50
flag image

What Stats Suggest about Quarterfinals

September 06, 2011
Before you bet too many beers on predictions, here are a few stats for the contenders remaining. Of course these stats - like all stats - signify little in the abstract, and less if anything at all in the reality of competition (since Serena is pre-ordained as the champion again, this is even more academic on the women's side). None-the-less, some of it reveals a little insight and humor about about how the game is and isn't being played differently by winners and losers, and by men versus women.


To start with a stat which - to my surprise - is remarkably similar for the men and the women, an average of 25-30% of points were decided on net approaches, with about 2/3 of those points being won by the approacher. This seems to lead to a promising inquiry: did those who came in more often and won their approaches more often end up more often as winners?

NO. The winners didn't necessarily approach more or win more at net than their opponents. With the exception of Fish and Tsonga, few of the players probably had much intention of getting to the net as a way of winning. For others, net approaches may have resulted from the fact that powerful, well-placed groundstrokes sometimes yield short return that drew them into the net. The best woman volleyer (Serena) didn't even bother to come to the net (she won 1 point in 4 approaches). Why should she? She's also the best server and the best from the baseline. Ana hit more winners than Serena, and might have hit even more and made fewer errors if she'd had a target to pass at the net. Though statistical anomalies don't refute the value in statistical norms when there is enough good date to analyze (there isn't here, or anywhere else), this example helps illustrate the limits of the insight to be gained from analyzing such stats: you win in tennis by playing your opponent well enough to win, not by playing the best tennis.


1st serve percentage:

This should be a key stat related to holding serve, but in this sample of only 12 matches there were 4 exceptions to what you'd expect to see (low percents leading to more breaks). Three of these exceptions were in the 8 women's matches, notably:

- Stosur served at 56% and suffered 3 breaks, while Kirilenko served at 75% and was broken 7

- Kerber served at 61% (good) and lost serve twice; Nicolescu served at 91% but lost 5

(of course, her average 1st serve speed was an appalling 67 mph, the lowest I ever recall)

There's at least as much humor in this as insight.

Break points faced:

In all but 1 men's match and 1 women's match, the winner earned more break point chances. This is a no-brainer, of course, and neither of the exceptions were dramatic.

As we'd expect on average, women faced more break points and were broken more often than men. Perhaps the actual percentages of break points saved and converted won't be startling either:

- Break points saved by winners: Men (25%); Women (47%)

- Break points converted by winners: Men (50%); Women (56%)

So, can reviewing data averages like this provide any useful insight about the play of the contenders?

Little, if any. The data of previous matches may be worth looking at before individual match-ups in the same event or in previous head-to-heads. But how Roger played against Monaco will have little to offer about how he will play against Tsonga, and since Fish plays more a net-rushing style than Roger, what worked for Tsonga against Fish may be different from what he'll need to do against Roger. In their last two H2H meetings, the only statistic that could remotely relate to the outcomes is the percentage of 2nd serve points Tsonga won in Canada (64% vs. 45% for Roger - high for Jo, low for Rog). That might suggest Jo took more chances than usual on Rog's 2nd serves, playing aggressively, or was able to do so on hard courts. Will Jo be able to do this again in NY? The courts are slow, so maybe.

More valuable is what I covered in my earlier post today about how many times each of the top 4 men had been broken and faced break points. This would tend to suggest that Roger or Joker should be favored in a final with Nadal or Murray this year, unless the data evens out between them in the next two rounds.

Reader Feedback Submission
Use this form to submit Reader Feedback. Your submission will be reviewed by our staff before appearing on the Web site.
* required value
Your Name*

Email (not shown on website)*