[We’re up to 158 matches now]. Since 2007 some of us have been counting stats in old and new matches. We now have stats for  men’s and women’s matches going back to Wimbledon in 1959. Below I’ve listed all the data, according to the highest rate of winners hit by an individual player in an individual match. For example, the highest rate we’ve found so far belongs to Stefan Edberg, who hit an average of 2.045 winners per game in his semifinal victory over Mats Wilander at the 1990 Australian Open. Just below Edberg is a rate of winners by Steffi Graf in the 1988 Wimbledon final, then a rate by Roger Federer, etc. There are two lists, organized according to: 1) rate of winners per game, not including aces or service winners 2) rate of winners per game, including aces These lists can be slightly misleading because they present averages per game, and certain matches have unusually long or short games. So I’ve included a column where I provide, if I know it, all the points played in the match. That way you can see how many winners a certain player hit, out of the total points played. I’ve included that information as a percentage. For example, Edberg finished 35% of all the points played in his match with his own winners. In some cases we don’t know what the player did on a few points missing on our copies. In those cases you’ll see the percentages in italics and a smaller font. When 4 points or more are missing from a match, I haven’t bothered with providing these percentages. In those cases you’ll see at least one game listed in the column for “Games Not Watched”. Whenever any games are listed as unwatched, the winners “per game” means only the games we actually watched, not the total number of games played. Points of clarification: 1) All the winners and aces presented here are clean winners. We have tried not to include any balls that the opponent touched with his or her racquet before the second bounce, no matter how slightly. This is an objective way to count, avoiding judgment calls. 2) As such, these lists don’t include service winners. We did count service winners for some matches. But the difference between an ordinary return error and a service winner is a judgment call, and we were not able to find an agreed-upon method of counting. Service winners have often been included in the total winners reported for a given match; however, they have not always been counted the same way. In some matches, the players are officially credited with very few service winners; in others they have more; in still others, all the service return errors are counted as service winners. We’ve even found one instance where the term “service winners” included the aces, so even terminology is not fixed. It’s an issue we’re still looking at and we’re happy to get any help on it – especially since the official counts of service winners are rarely reported. There is no quick way to get a lot of data on service winners and analyze it. To know how many service winners are in the published count for a given match, you either count the clean winners yourself and subtract from the official total of all the winners, or contact the statistician. 3) There are very few women’s matches in the lists, because when I first started doing these stats I wanted to stick to the men, just to keep things as simple as possible. When the Graf-Navratilova final at Wimbledon 1988 was uploaded to YouTube, I casually counted the winners, just to see what would happen. To my surprise, Graf had a higher rate of winners than any men’s performance that I knew about at the time. So I proofed and included the data, and did another Graf match for comparison. I did a few others in order to check published numbers that looked incorrect. (One of the most interesting things about this project has been the number of mistakes we’ve found in official counts). Since then we’ve done more, but the women’s matches in these lists are few and far between, and not chosen with any consistent rule in mind. We did a lot of men’s matches though the only “rule” or goal we had in mind was to do a lot of famous matches in our personal collections. 4) We counted all of these winners ourselves. The only place we use stats from other sources is when we don’t have the total number of points in the match, though often we got that number ourselves. The following is a chronological list of all the matches. I did the stats myself, unless otherwise noted.