Men, By the Numbers
by Pete Bodo
Have you ever wondered what, if any, correlation exists between producing impressive statistics (first-serve percentage, break points saved, etc.) and the rankings? Is it possible to be outstanding in a number of statistical departments yet not win enough matches to earn what might seem an appropriate ranking? Over the coming days, I'm going to be looking at some of the year-end statistics, partly to see what I can glean from them in relation to the rankings. And we'll start today with the year-end ATP Top 10, and how they stack up in the statistics department.
I wondered, going in, if the rankings would remain identical if you somehow assigned values to statistical excellence in the categories tracked by the Ricoh ATP Match Facts program. So I did the most basic and obvious thing: I assigned points for appearing in the top 10 in each statistical category, for each of the top 10 players. The number of points each man earned is the reverse of where he was ranked—e.g. if Novak Djokovic finished No. 1 in 2nd Serve Return Points Won, (which he did) he received 10 points, while the No. 10 man in that category (if he's in the ATP Top 10) received a single point. This way, the guy who accumulates the most points also ends up being the top-ranked "stats man." It just seems clearer that way.
Also, please note that I did not include the Aces category in these calculations, as I felt the other nine categories best represent a players' all-around game. In addition, all results were compiled from matches on all surfaces, though you can limit the results to a specific surface, if you want to have some fun.
So here's the ATP Top 10 organized not by ranking but by how well they performed in the statistical categories. Each man's name is followed immediately by his "real" ATP ranking, and then the number of points he accumulated by my measure. As always, feel free to double-check my math as I'm only human, at least in theory.
1. Novak Djokovic (ATP No. 1), 54 points: As I compiled the figures, I thought I had hit upon a gold mine of counter-intuitive data. With just four categories to go, Djokovic had a grand total of just 16 points. I was prepared to be declared an investigative genius for demonstrating that, going just by statistics, Tommy Robredo might be a better player than Novak! Ha. Djokovic earned a pair of 10s and a pair of 9s in the final four categories (respectively, Break Points Converted, 1st Serve Return Points Won, 2nd Serve Return Points Won, and Return Games Won). Djokovic failed to earn points in just one category: 1st Serve Points Won. That's a pretty good measure of the degree to which he dominated.
2. Rafael Nadal (ATP No. 2), 46 points: Nadal earned points in six of the nine categories. His best two results were his second-place finishes (nine points) in 2nd Serve Points Won and 2nd Serve Return Points Won. Call him the master of the second serve at both ends of the court. The biggest surprise is that while Nadal ranks sixth in 1st Serve (Percentage), he's not even in the top 25 in 1st Serve Points Won.
3. Roger Federer (ATP No. 3), 35 points: Roger scored almost perfectly in 1st Serve Points Won and 2nd Serve Points Won (9 and 10 points respectively), but he really tailed off in Break Points Converted, 2nd Serve Points Won, and Return Games Won—he didn't finish in the top 10 in any of those categories and thus earned no points. This suggests that Federer's hold game was lethal, but his return game was relatively lackluster.
4: Andy Murray (ATP No. 4), 32 points: Like Djokovic, Murray's numbers looked pretty grim until the final four categories, which really are about the return game. That's pretty interesting to me, and has to be good news for him. Murray was tops in 1st Serve Return Points Won, and he was outstanding returning second serves as well as breaking serve. He was as good in the return-game department as Federer was in the service-game areas; look at the mere three-point difference in their final tallies. Among the Top 4, Murray is the only one who finished in the Top 10 in just four categories.
5. David Ferrer, (ATP No. 5), 24 points: The return game was Ferrer's great strength in 2011. He finished fourth in Return Games Won as well as 2nd Serve Return Points Won. He also did well in 2nd Serve Points won, finishing in the top 10, but he wasn't effective enough on his own first-serve points to do the same.
6, Jo-Wilfried Tsonga (ATP No. 6), 19 points: The big Frenchman was weak in the return game, but he's right up near the top in 1st Serve Points Won, as well as Service Games Won and Break Points Saved—all of them heavily serve-biased categories.
7: Tomas Berdych (ATP No. 7), 10 points: Just as there's a significant drop-off in points after Murray, there's one after Tsonga as well. Berdych and Tsonga's stats are similar, both did best in 1st Serve Points Won, but Berdych made the top 10 in at least one return category, 2nd Serve Return Points Won (he finished ninth.)
8: Janko Tipsarevic (ATP No. 9), 7 points: This is our first—and only—departure from the ATP rankings. Tipsarevic gets bumped up a notch ahead of Mardy Fish mainly because he was an outstanding fifth (good for six of his seven points) in 2nd Serve Return Points Won. His other point came as the No. 10 man in 2nd Serve Points Won. Keep working on winning those first serve—and return—points, Janko.
9: Mardy Fish (ATP No. 8), 2 points: Fish finished ninth in 2nd Serve Points Won, a testament to his serve. The mystery is why that serve didn't get him more traction in the various other serve-biased categories.
10: Nicolas Almagro (ATP No. 10), 1 point: You'd probably never guess where he got that lone point, either. It was in 1st Serve Points Won.
To tell you the truth, I was a little bummed when I realized that the stats-based rankings would be nearly identical to the regular rankings. But the synchronicity is really impressive, and it supports the idea that the statistics are actually an excellent barometer and predictor of excellence. So much for the romantic, perhaps over-complicated notion that it really isn't all about quantifiable performance, that how many break points you create means nothing compared to how many you convert. The message here is that if you're going to create a large number of break points, you'll convert enough to earn a high place in the rankings.
We sometimes want to downplay statistics, but while I can't exactly say they don't lie, chances are that the lies they tell are white ones. Over time, as this little exercise suggests, it all kind of evens out, and the players who do some or most things better than their peers and rivals will enjoy a commensurately greater degree of success.