Upsetopsy: Garcia-Lopez over Nadal

It was September of 2010, and Rafael Nadal was on top of the tennis world. Earlier in the month, he had won the US Open, thereby completing the career Grand Slam, extending his streak of consecutive Slams won to three, and further securing the top spot in the world rankings. The next event Nadal played was a small tournament in Bangkok; he had a first-round bye and then won his next two matches with ease (dropping 5 total service games in each one).

His semifinal opponent was fellow Spaniard Guillermo Garcia-Lopez. He entered the tournament with an ATP ranking of 53 and a respectable 21-19 record on the year, having lost in the first or second round of all four Slams and with a solitary final in Eastbourne to his name. The three wins that put him in the semis were all hard-fought, with scores of 6-4, 7-6; 7-5, 6-7, 7-5; and 7-6, 4-6, 6-3. And while all three came against top-75 opposition, none of them approached the level of the world’s best player, one who had bested Garcia-Lopez with ease in their two previous encounters.

The final statistics of the match look mostly like one would expect: Nadal won more points than his countryman, 112-107, and did so despite serving far less. Rafa won a healthy 71% of points on his serve (60 of 84), while Garcia-Lopez took a mere 61% (83 of 135) when he was serving. (The ATP Tour average in 2010 was 63.4% of points won on serve.) This discrepancy was in turn reflected in the number of break points faced by each player – Nadal had an unbelievable 26 chances to take Garcia-Lopez’s serve, while the lower-ranked Spaniard earned only one solitary opportunity in the match.

All of this suggests that Nadal was the superior player – and, of course, we know that he was superior overall, and remains so. But as anyone who has followed sports for any length of time at all knows very well, the better player (or team) does not always win.

It certainly appeared that he would in the early going, as Nadal secured the opening set, 6-2, thanks to a pair of breaks. But his success on break points dried up in the second set, and even though he held his own serve with ease, Garcia-Lopez managed to stay on serve, and took the resulting tiebreak 7-3 to level the match.

And in the third, leading 2-1, the #53 player in the world finally earned a break point, and promptly converted. Nadal had chances to break back throughout the remainder of the set, but failed to capitalize, and went down in stunning defeat by a final score of 2-6, 7-6, 6-3.

Garcia-Lopez would go on to victory in the final against Jarkko Nieminen, securing the second (and, to date, last) title of his career. Nadal, meanwhile, recovered promptly enough to grab a larger title in Tokyo the next week, his seventh of the year. As a further postscript, the two have faced each other twice since this match, and Nadal has prevailed both times in routine straight sets.

So, outside of the already-known-but-worth-repeating fact that the better player does not always win, is there anything we can learn from this match?

The main point that jumps out to me is that this establishes the practical limit on how much a player can be out-performed over the course of a match and still come out with a victory. In 2010, there were 2,573 ATP World Tour matches that were played to completion. Of those matches, only 184 were won by a player who won a lower percentage of his own service points than his opponent, just over 7% of the total. (Incidentally, five matches were exact ties by percentage of service points won.) Only 24 (less than 1%) featured a victor who trailed his foe by 5% or more, and only four (including this one) had a difference that rounded to 10% or higher.

Let's examine the 24 matches that were won by a player who was exceeded by 5% or more in service point percentage. Only one of the matches came in a Grand Slam (Florian Mayer’s second-round upset of Mardy Fish at Wimbledon), but ten occurred in Masters events. Only three players were on the winning end of one such match and the losing end of another – Robin Soderling, Richard Gasquet, and Marco Chiudinelli. Three players won more than one match of this type – Olivier Rochus won two, as did Ivan Ljubicic, while John Isner won an unbelievable four matches in which he was badly out-performed in service point percentage, including two of the four that round to 10% or higher. (Isner's performance may be a trend worth examining in other seasons.) Ljubicic’s pair of victories are particularly notable, since they were the semifinal and final of his surprise victory at Indian Wells.

And on the other end of things, there were two players who lost more than one match in which they easily outplayed their opponents in service point percentage. Thiemo de Bakker lost three of them, which is rather remarkable. The other player who was on the wrong end more than once? Nadal himself, who was also Ljubicic’s victim in the aforementioned Indian Wells semifinal. Which means that, great a year as Rafa had in 2010, it could have been even better if he could have caught a break in either of those two contests