Blogs

Post a Comment

How unlikely was Wawrinka's Australian Open victory?

January 30, 2014

In this article, I will examine the prior probabilities that Stanislas Wawrinka, or a player with a similar seeding in the tournament, would go on to win the 2014 Australian Open. I will also review any prior matches similar to the final between Stanislas Wawrinka and Rafael Nadal to determine his chances of winning such a final based on previous results in the Grand Slam tournaments since 2005.

Stanislas Wawrinka was seeded eighth at the 2014 Australian Open, which matched his ATP Tour ranking going in to the tournament. Winning the tournament gave him a significant boost in the rankings and he is now ranked number three on the list that was released on Monday, January 27.

To put Wawrinka’s achievement into perspective, let us examine how unlikely his victory was, not just winning the final against Nadal, but to actually go all the way from being seeded eighth to clinching his first Grand Slam title.

A lower seeded player winning a Grand Slam title

Only once since 2005 has a player who was seeded lower than fourth won a Gentlemen’s Singles Grand Slam title, and that was when the sixth seeded Juan Martin Del Potro defeated Roger Federer in five sets at the 2009 US Open. The Gentlemen’s Singles Grand Slam titles have been dominated by the “Big Three,” and more recently by the “Big Four” with Roger Federer and Rafael Nadal each having taken 13 of the titles since 2005.

table_1.jpg

Marat Safin also holds the 2000 US Open title, which he won as the tournament’s sixth seed. Roger Federer holds a total of 17 titles, with his first coming at the 2003 Wimbledon (fourth seed). He took another three titles in 2004 (as the second, first and first seed, respectively) that are not included in the table above. The only Grand Slam title that Roger Federer did not win in 2003 was Roland Garros, which was won by the unseeded Argentinean Gaston Gaudio, proving that while it is improbable for an unseeded player to win a title, it is not impossible.

table_2_0.jpg

The table above shows the likelihood of a given result (shown across the top) based on the players seed in the tournament (shown down the side). The table should be read across, with the most likely outcome for a player at a given seed being highlighted with a green background: the darker the color, the more likely the outcome. The table shows that there is a fairly good correlation between the seed and the expected result for the player, but there are some discrepancies; for example, the second seeded player is more likely to win the tournament than to be the runner‑up.

Wawrinka was seeded eighth .The most probable result for a player with a seed between five and eight would be to reach either the Round of 16 (with a probability of 24.6 percent) or the Quarterfinals (with a probability of 23.9 percent). His chances of winning the title, however, were only 0.7 percent based on the history of the Gentlemen’s Singles since 2005, and even his chances of reaching the final were just 5.0 percent—a 0.7 percent chance of taking the title and a 4.3 percent chance of being the runner-up.

Wawrinka’s chances in the final against Rafael Nadal

Since 2005, there had been 40 matches between a first or second seeded player and a player seeded between five and eight. Six of those matches (15.0 percent) were won by the lower ranked player, so, based on that alone, Wawrinka would have about a one-in-six chance of winning the final. At the Australian Open, however, there had been 14 such matches, and only one of these was won by the lower ranked player, reducing his chances to 7.1 percent.

Then take into account the record of previous matches between Nadal and Wawrinka; having met twice before at the third round of the 2007 US Open and at the Quarterfinals of the 2013 Roland‑Garros, Wawrinka was still to win a single set against Nadal at a Grand Slam tournament, as he lost both of those matches in straight sets.

table_3_0.jpg

Wawrinka’s true chances of winning going into the final were probably somewhere in between these different estimates and perhaps even slightly better. He had already proven himself twice in 2013 against Novak Djokovic: first in the quarterfinals of the Australian Open and then again in the US Open semifinals. Although he lost both of these matches in five sets, he had defeated the fifth seeded Tomas Berdych and the third seeded Andy Murray on his way to the US Open semifinals; Wawrinka showed that he was not to be taken lightly by anyone and that he was on an upward trajectory.

A match summary by the numbers

The final itself was dominated by Nadal’s back injury, which affected his play throughout the match and in the second set in particular.

The IBM “Keys to the Match” identifies key performance indicators: what each player needs to succeed in a match. During the match, each player’s performance is measured against their keys and updated in real time in the IBM SlamTracker application on the tournament website—see this previous blog post about the “Keys to the Match,” and watch this videochat where I discuss it.

The “Keys to the Match” for Nadal and Wawrinka from the Australian Open final are shown in Figure 1 below, which also shows the match summary of how the two players performed against each of the Keys.

figure_1.jpg

These Keys tell the story of the match from a standpoint of the most relevant statistics for the match. The statistics are the same for the two players, but with different ranking of importance; the Keys are listed from top to bottom in descending order of importance for each player, and with different target values.

You can see that over the course of the match, Nadal did not reach the target value for any of his three Keys, while Wawrinka exceeded the target values for each of his three Keys.

However, you can also see that Nadal’s injury changed the way both players approached the match, and this is reflected in their statistics. Most notably, it appears that Wawrinka took the pressure off his first serve in the third set after having won the first two sets and after seeing that Nadal was injured and was having trouble moving. His first serve percentage in the first set was just 38 percent (which might have been a little too aggressive) and it increased to 77 percent in the third set, which was the only set he lost. The increased first serve percentage came at the cost of service speed, which went from 193 km/h (120 mph) in the first set to 188 km/h (117 mph) in the third set. In itself this is not a major change, but it could be seen as a symptom of a different approach or strategy by Wawrinka in the third set compared to the previous sets—a strategy that was unusual for the Swiss and did not work out for him.

While this change did not directly impact the “Keys to the Match,” it appears to have shifted the focus from the statistics that would have been important under normal circumstances to other statistics in the third set—a development that was impossible to foresee prior to the match. This shift in focus, along with Nadal’s prior status as the clear favourite to win the match, might help explain why Nadal was able to win the third set despite only reaching one of his Keys, while Wawrinka reached the target value for all three of his Keys in that set.

Unlikely, but true

You can look at any number of different statistics related to Stanislas Wawrinka going into this tournament, and you will find his victory to fall on the side of “unlikely,” but he persevered, he beat the odds, and he took his first Grand Slam title.

Read more on my background and interests, and follow me on Twitter: