Playing the Australian Open at age 39, Serena Williams is trying again to lock down one of the last puzzle pieces of her magnificent career.
Widely regarded as one of the greatest tennis players ever, Williams’s only remaining snag is the Grand Slam record of a troublesome Margaret Court, who played in an era with weaker competition. Idling at 23 Grand Slam singles titles, Williams needs one more to tie and two to overtake Court in the overall tally. She’s come vexingly close, winning plenty of Grand Slam matches, but has suffered a string of late-round trip-ups since her comeback from having a child in 2017.
Who’s the most chaotic fictional football coach? | FiveThirtyEight
So what is it that’s gone wrong in the eight Grand Slam losses since her comeback? Observers have pinned the blame on everything from age to nerves to poor conditioning. But an analysis of the stat sheets from those eight Grand Slam losses reveals a fairly obvious umbrella of differences between what happened in defeat versus how Williams normally plays.
It’s not her serve speed. In 2015, when Williams dominated the sport by capturing three of the four Slams, her average first serve speed was 109 mph — identical to what it was in her last Grand Slam loss at the hands of Victoria Azarenka. Williams’s percentage of first serves landed has been mixed in these matches. She’s been at or close to the ideal of around 65 percent of first serves landed in half of those eight Grand Slam losses. Her percentages in the remaining half have been either too high or too low, but the spottiness of performance in this area isn’t enough to point to a trend. In this data set of Grand Slam losses, her averages mostly hovered around the mean of 65 percent of first serves in, which makes it unlikely to be a chronic problem at the moment. In addition, her first-serve win percentage is steady: In her three most recent Grand Slam losses, Williams’s win percentage on her first serve points was 74 percent, 70 percent and 72 percent — all in line with the Top 10 players on the WTA Tour in 2020. There has to be something else going on.
Imagine James Carville sitting across from a group of tennis pundits grasping for why Williams hasn’t broken through yet. To those pundits, he might say, “It’s the return, stupid.”
Outside of metrics relating to serve, a mountain of meaning is built into the statistical category “return points won” because points played on return make up roughly half of all points played. Like many stats in tennis, it’s most helpful to view this category as a percentage rather than a raw number of points, since tennis matches can vary wildly in the overall number of points played.
Williams seems to sense that her problem lies in her return. After losing to journeywoman Wang Qiang in the third round of last year’s Australian Open, she said, “I didn’t return like Serena. Honestly, if we were just honest with ourselves, it’s all on my shoulders. I lost that match. … I literally can’t do that again. That’s unprofessional. It’s not cool.”
In that three-set match, Williams committed 30 errors on her return of serve. She still nearly pulled the match out, losing 7-5 in the third set. Imagine if she had found a way to put just half of those 30 returns in play. And no, blasting a bunch of return winners to offset all the errors did not even things out — she had only seven return winners to those 30 errors.
Looking at Serena’s percentage of return points won in her eight Grand Slam losses since her comeback, the trend is off her overall season-encompassing benchmarks.
|2019||French Open||Kenin||3rd round||28/78||35.9|
|2020||Australian Open||Wang||3rd round||39/114||34.2|
These losses show a trend of return point win percentages mostly in the 30s, a departure from her normally solid return game. By comparison, overall in 2020, she won a markedly better 44 percent of her return points, and in Williams’s banner year of 2015, her share of return points won was 48 percent. Examining her return-point history over the past decade, it’s possible that her success rate in this category needs to be in the 40s in order for her to win, while percentages in the 30s leave her vulnerable.
In 2019, the last full year of professional tennis before the pandemic, the top 60 players on the WTA Tour all had seasonlong return point winning percentages in the 40s. Simply put, it’s the range where the most successful women in tennis reside.
Consider, too, that there’s more to winning a return point than the return stroke itself.
Historians agree that Williams is one of the best returners of all time. She’s a woman capable of returning a male player’s 138 mph serve with a blast for a winner. Her return speeds sometimes far exceed the speed of the serve into which she tore.
Is Serena not putting enough returns in play — either by getting aced, flubbing the stroke or going for too much?
The answer is: sometimes.
According to Tennis Abstract, most of the top women in the world put between 75 and 83 percent of their returns in play. This means they are hard to ace, but also, they make very few errors on their return strokes.
In the 2018 loss to Angelique Kerber, Williams’s share of returns in play was an acceptable 71 percent. However, in the U.S. Open final loss to Bianca Andreescu, Williams’s returns in play slipped to just 63 percent. That was just a two-set match in which Serena committed 19 return errors — or nearly two return errors per receiving game.
The return isn’t the only time a player can lose on a return point, of course. There is at least some indication that Williams might also be struggling with her very next stroke following a return hit in play. Where rally length data is available for these Grand Slam losses, in at least four of them, Williams lost the battle of short points, which tennis stat keepers classify as from 0 to 4 shots. In the tournaments that ended with her losses to Wang, Naomi Osaka and Karolína Plíšková, Williams had previously won all the short-point battles leading into her defeats. It’s an indication of a quick flameout in points — at least in the Grand Slam matches she’s losing.
What is potentially most egregious about Williams’s diminished performance in the category of return points won is the caliber of servers she’s letting slide. In these eight Grand Slam losses, she faced two notably elite servers in Osaka and Plíšková. It’s almost pardonable to not return well against those two. But the rest? You’d think Williams would feast on their serves. Andreescu, Azarenka, Kerber, Wang, Simona Halep and Sofia Kenin are nowhere near the 2020 WTA Top 10 in aces or percentage of first serve points won.
However, that group of players contains supreme defenders and fast runners. The only way to find out why Williams is slipping in her return game against these players might be to get inside her head, but it’s possible that the specter of getting into long, extended running points against them is weighing on her, consciously or not.
Although Williams doesn’t need the Grand Slam record to cement her greatness, it would be satisfying for her and her fans if she completed the task. A straightforward shoring up of her return game — taking care to put returns in play and seeking to extend receiving point rallies — could help her clear the last obstacle.