Serena Williams dominates my most recent WTA hard court rankings, so it’s no surprise that she’s favored to win the U.S. Open. As was the case before Wimbledon, it’s remarkable to see how chaotic the women’s field is. While Novak Djokovic has a 28% chance of winning the men’s event, Serena is the only woman in double digits, at 14.2%.
Because of Serena’s low seeding at #28, a decisive match may take place in the first week. Assuming some easy wins for both Williams and Victoria Azarenka, the two ladies will face off in the third round. My algorithm gives the American a 59% chance of winning that match, meaning it could be the toughest test she faces in the entire tournament.
Behind Serena, Carolina Wozniacki has a 9.8% chance of winning the U.S. Open, followed by Maria Sharapova at 9.2%. Next up are Petra Kvitova at 8.0% and Vera Zvonareva at 7.9%. An amazing 21 women (compared to 13 men) have at least a 1% chance of going home a champion. These include the unseeded Venus Williams (1.8%) and the 32nd seeded Maria Jose Martinez Sanchez (1.0%, or 0.25% per name).
The conspiracy-minded among you might note that top seeds Wozniacki and Zvonareva have the most favorable first-round odds, despite my system ranking them only 3rd and 7th on hard courts. Their opening opponents, Nuria Llagostera Vives and Stephanie Foretz Gacon, are the 16th and 23rd weakest players in the draw, according to the current WTA rankings. (My system isn’t as reliable that far down the WTA list.) Caro and Vera are the only two players with a better-than-90% chance of winning their openers, though both Sharapova, Marion Bartoli, and Andrea Petkovic are at an even 90%.
Here are a few interesting first-rounders. In each of these, my system gives neither player a better than 55% chance of advancing, with the favorite in bold:
- Greta Arn vs Vania King
- Eleni Daniilidou vs Michaella Krajicek
- Anne Keothavong vs Chanelle Scheepers
- Francesca Schiavone vs Galina Voskoboeva (the 7th seed doesn’t have good results on hard courts)
- Alla Kudryavtseva vs Anastasia Rodionova
- Misaki Doi vs Laura Pous-Tio
- Melania Oudin vs Romani Oprandi
- Anastasija Sevastova vs Vera Dushevina (nearly 50/50)
- Maria Kirilenko vs Ekaterina Makarova (Kirilenko is another vulnerable seed)
- Coco Vandeweghe vs Alberta Brianti
Hi, I found it interesting to see the difference in predictions for men’s and women’s draws, with Djokovic dominating men’s and no dominance to the same degree of the women’s draw.
I was wondering if there would be any way to quantify the ‘open-ness’ of a draw.
I considered the standard deviation of the win percentages for each player.
For womens I got about 2.1, and men’s about 3.2. Clearly the higher number means a greater dominance
I wonder if this would be a useful measurement across different years. I would like to investigate the ‘open-ness’, or ‘one-sidedeness’ of grandslams past, and see if anything interesting happens.
For example, whether or not women’s tennis was always this open, maybe that has something to do with the best of 3 rather than best of 5 sets condition?
I don’t have the data, but I was wondering if you had any thoughts about this.
Yes, I’ve been curious about that as well. I don’t yet have draws in my database going back that far, but I suspect that both men’s and women’s draws vary over time. Recently, we’ve seen a wave of dominant players in the ATP and a particularly even group of women, but there have certainly been eras in the not-too-distant past where the opposite was true.
I don’t think the 3 vs 5 set thing is an issue — I suspect you’d get similar results if you looked at the spread of ranking points in the top 100, and ATP results are based heavily on the best-of-3 matches in most tourneys.
By Win percentages I mean probabilities