Skip to main content

ATP World Tour Analysis: Rethinking Serve Stats

By Zach Bigalke

I decided to wander over to the ATP World Tour website for no other reason than to see if the leaders in the various statistical categories were the leaders in the rankings. By and large what I saw was what I expected to see. There were a few oddities there at the top of some of the categories, for sure, but for the most part the top players in the game were the statistical leaders as well.

Of course, the statistics there on the home screen are really just ten simple categories that hardly tell the full story. Aces are a simple story — but even then, they list the top 100 players not by aces per game but by pure aces. They have first-serve percentage, and first-serve win percentage, but those too only tell part of the story. These sorts of things simplify and yet obfuscate the whole story.

So let’s do some manipulation this week and come up with a few new ways of looking at success within the statistics that tennis does offer up. Here are three new ways that I’ve determined to look at the statistics more clearly. Some are elementary, others are more complex, but they all should prove of value in one way or another…


It all stemmed from one question that was on my mind: Why is Feliciano Lopez leading in aces?

It turns out that he does hit a lot of aces… but by the new metric he ranks sixth, not first. Rather it is a guy I suspected of being the best ace man on tour, Croatia’s Ivo Karlovic, who topped the charts. Long considered one of the most feared servers on the ATP tour, his spot as the most prolific man per match — with almost seventeen aces every time he competes — is hardly a shock.

Do you know what is a shock? None of the men in the top ten in ace ratio are in the top ten in the world rankings. Further, three of the ten — Karlovic (126), Dustin Brown (123) and Frank Dancevic (179) — are not even in the top 100.

The highest ranked among the top ten of ace ratio? Andy Roddick, ranked twelfth in the world currently, the hard-hitting American who serves up 13.4 aces a match on average. After Jo-Wilfried Tsonga (9th in ace ratio, 16th in the rankings, 11.24 aces/match) no other players are even in the top 25.

So aces are hardly a worthy measure of a man’s worth on the tennis court… they are often a sign, it appears, of a player who is relying too much on his serve to bail out other deficiencies in his game.


The ATP has two statistics on its website about first serves — first-serve percentage and first-serve points won. Both are percentage calculations, the first about how many first serves a player makes legally and the second concerned with how many of those successful attempts are ultimately won by the server.

Both alone tell a story that makes no real sense. Sure, Potito Starace is stellar at getting his first serves into play. But why is he not blazing a trail through the tennis world? Why has the 30-year-old Italian veteran never ranked higher than 27th in the world, or advanced into the second week of play at a Grand Slam tournament?

It’s quite simple… you need to look at both numbers to see what percentage of total serves are ultimately winners coming on a first serve. To do so is a matter of taking the two percentages and multiplying them together. Doing so gives you the percentage of all serves that equal success on the first attempt.

What we see here is more accurate. Again we see John Isner highly rated here, his serve proving an effective weapon. Karlovic, the one man with more aces per match than him, doesn’t even rate among the 57 qualifying players.

So this statistic obviously illustrates a greater potential for success. Roger Federer and Novak Djokovic are ranked third and fourth in success rate, winning just under half their service points on first serves; four of the top ten players in the world are on this list. So we’re on the right track…


And this is where it all comes together to evaluate what percentage of points in his serve go toward a winning effort in a given service game. By combining the success rate for both first and second serves, and taking those combined success rates and multiplying them by the percentage of service games won, you get serve efficiency.

This is perhaps the truest stat I’ve seen of a statistical measurement of success. Everyone high on the list makes sense, really, and we’re seeing how various components of serve statistics can demonstrate efficiency.

Take Isner’s breakdown once more, since he seems to have been a recurring player in this week’s column. Over 63 percent of Isner’s service points go toward holding serve. Likewise many of the top names can be found here.

The beauty of this number is that it not only shows how well each player does on a particular serve, but shows how infrequently he can waste his serve…

So there you have it… as we’ve learned once more, statistics can show only part of a story unless they are combined to make further sense of things. Unless you manipulate the figures to sensibility, it is all just a bunch of numbers that really make little sense toward determining both the worth of a player and the worth of the statistic being compiled…

Get more great sports analysis over at the


Popular Video