Aryna Sabalenka Under Pressure

Also today: January 26, 1924

Aryna Sabalenka at Wimbledon in 2023. Credit: Adrian Scottow

It felt like a pivotal moment. Aryna Sabalenka had taken a 5-2 first-set lead in yesterday’s Australian Open semi-final against Coco Gauff. Gauff kept the set going with a strong service game for 5-3. Sabalenka lost the first point on her serve, but bounced back with a plus-one backhand winner.

At 30-15, the American struck again. She took advantage of a Sabalenka second serve to drag the Belarusian into a backhand rally, ultimately drawing an unforced error on the ninth shot and putting the game back in play.

Then, still just two points from the set, Sabalenka double-faulted.

The narrative practically writes itself. Aryna hits hard, aims for the lines, and keeps points short. Let her do that, and she will destroy you. Her first five opponents in Melbourne managed a grand total of 16 games against her. On the other hand, if you keep the ball in play, she’ll start pressing, trying too hard to dictate with her serve, going for too much when a smackable groundstroke presents itself.

Gauff, by this reading, is Sabalenka’s nightmare opponent. She won the US Open final by denying the Belarusian one would-be winner after another. Not only can she take Sabalenka’s game away from her, but Coco–at least on a good day–won’t give it back on her own serve. When she lets loose, Gauff wields just as much power as her more tactically aggressive opponent.

As it turned out, Sabalenka did lose that service game. Several twists and turns later, Gauff led the set, 6-5. Only then did Aryna regroup, winning four straight points from 30-love to force a tiebreak, then dropping just two more points to clinch the set. Gauff kept the second set close, but Sabalenka never allowed her to reach break point. The contest closed with a narrative-busting move: Facing match point, Gauff pulled out a 12-stroke rally, the kind of point that has been known to steer her opponent off course. But instead of compounding the damage, Sabalenka came back with two unreturned serves. Game over.

What to believe, then? Was the apparent first-set turning point a reflection of the true Sabalenka? Or is this the new Aryna, who slams the door when challengers sniff opportunity? Or is it something else, the all-too-common story in which someone looks like a clutch hero or a constant choker, only for us to discover, after crunching all the numbers, that she’s impervious to momentum and plays pretty much the same all the time?

Recovering at a disadvantage

Sabalenka’s serve games do follow a pattern after she loses a longish rally. But the results are not entirely straightforward.

On the next point (assuming the lost rally didn’t end the service game), Aryna is more likely to miss her first serve:

Year   1stIn%  post-rallyL-1stIn%  Change  
2019    61.2%               55.9%   -8.6%  
2020    61.5%               57.0%   -7.3%  
2021    58.6%               52.6%  -10.3%  
2022    60.0%               59.9%    0.0%  
2023    61.1%               61.3%    0.4%  
2024    63.3%               62.5%   -1.2%
----  
TOTAL   60.5%               57.6%   -4.8%

Most of the effect is concentrated in the earlier years of her career on tour. Yesterday, the trend ran in the opposite direction: She made nearly 76% of her first serves overall, but after Gauff won a rally, she landed 88% of them.

The trend is clearer–and persisting to the present–when we look at double faults after losing a rally:

Year     DF%  post-rallyL-DF%  Change  
2019    8.6%            10.4%   20.8%  
2020    6.2%             8.4%   36.9%  
2021    7.9%            11.8%   50.3%  
2022   10.7%            10.1%   -5.5%  
2023    6.2%             7.2%   16.5%  
2024    3.4%             8.3%  144.7%  
----
TOTAL   7.9%             9.6%   22.5%

2022 was Aryna’s year of the yips; she was more likely to bunch double faults together than hit them in particularly nervy spots. (Put another way: Every spot was a nervy one.) The 2024 number will surely come back to earth, but it is still revealing: Sabalenka has made so much progress in this aspect of her game, but her second-serve struggles continue when she faces the threat of getting dragged into another rally.

Some of these effects persist even longer. From those service games that last long enough, here are Sabalenka’s first-in and double-fault percentages two points after losing a long rally:

Year   1stIn%  +2 1stIn%  Change    DF%  +2 DF%  Change  
2019    61.2%      55.8%   -8.8%   8.6%    8.7%    1.2%  
2020    61.5%      50.5%  -17.9%   6.2%    7.2%   17.1%  
2021    58.6%      56.0%   -4.5%   7.9%    8.7%   10.5%  
2022    60.0%      63.1%    5.3%  10.7%    7.8%  -27.1%  
2023    61.1%      59.2%   -3.2%   6.2%    8.4%   35.6%  
2024    63.3%      57.1%   -9.7%   3.4%    2.4%  -30.1%  
----
TOTAL   60.5%      57.1%   -5.6%   7.9%    8.0%    2.0%

She continues to miss more first serves even two points after the rally setback. To some degree, the memory should have dissipated–after all, something else happened on the intervening point. On the other hand, she’s back in the same court. If a reliable serve didn’t work in the deuce court at 30-love, there’s reason to doubt it at 30-all.

The double fault trends are less clear, in part because our sample size is shrinking and double faults are blessedly rare. If nothing else, it’s safe to conclude that the explosion of double faults on the point after the lost rally doesn’t continue to nearly the same degree.

Tallying the cost

Now, this all seems bad. Sabalenka possesses one of the best first serves in the game; her whole attack is built around it. Her emergence as a superstar came after she got control of the service yips and cut her double faults down to manageable levels. After losing a long rally, she needs her serve more than ever, and–at least by comparison with other situations–it isn’t there for her.

Except… it doesn’t matter! At least not on the first point. Here is the bottom-line figure of service points won:

Year    SPW%  post-rallyL-SPW%  Change  
2019   59.6%             63.8%    7.2%  
2020   60.3%             56.6%   -6.0%  
2021   61.5%             61.3%   -0.3%  
2022   57.2%             59.9%    4.7%  
2023   63.7%             63.9%    0.4%  
2024   66.7%             70.8%    6.3%  
----
TOTAL  60.7%             61.7%    1.6%

Fewer first serves, but more serve points won. It isn’t supposed to work like that, but Sabalenka bounces back strong from lost rallies. A shift of +1.6% in her favor is solid enough, and it’s even better if you look solely at the last three years.

Part of the explanation is that she tightens up the rest of her game–exactly the opposite of what my off-the-cuff narrative suggests. Under pressure, I hypothesized, she would try too hard to end points. Instead, after losing a long rally, she’s more willing than usual to play another one: She commits 14% fewer plus-one errors than her usual rate, implying a lower rate of aggression when she has an early chance to put the point away.

On the second point after losing a long rally, the bottom-line outcomes are more mixed:

Year    SPW%  +2 SPW%  Change  
2019   59.6%    53.9%   -9.5%  
2020   60.3%    55.3%   -8.3%  
2021   61.5%    58.5%   -4.9%  
2022   57.2%    61.5%    7.4%  
2023   63.7%    60.7%   -4.7%  
2024   66.7%    71.4%    7.1% 
---- 
TOTAL  60.7%    58.2%   -4.0%

While these aren’t as rosy as the next-point results, focus on the last few years. Since the beginning of 2022, Aryna has won more service points than usual when she returns to the serving direction where she recently lost a long rally–despite landing fewer first serves. She is even stingier with plus-one errors on these points, coughing up 29% fewer than usual.

These trends did not hold in yesterday’s semi-final. While Sabalenka made more first serves on the two points after Gauff outlasted her in a rally, fewer of them ended in her favor: 4% less on the first point, 12% less on the second. We can’t read too much into single-match totals with stats like these: 4% is a difference of one point. And Gauff is a far superior returner and baseline player than the typical WTAer, one who is unlikely to lose focus after going toe to toe with Sabalenka for a point or two. The average player pushes Aryna to a seventh shot barely one-tenth of the time; Gauff did so on one of every six points yesterday.

All of this leads us to an unexpected conclusion: Does Aryna Sabalenka have nerves of steel? First serves and double faults are just components in a larger picture; when we measure her results by points won, Sabalenka serves more successfully right after an opponent makes her uncomfortable. The yips are gone, and the on-court histrionics are a diversion that deceived us all. Aryna under pressure may be even more fearsome than her typical, terrifying self.

* * *

January 26, 1924: Suzanne’s longest day

Suzanne Lenglen wasn’t accustomed to spending much time on court. In eight tournaments since the 1923 Championships at Wimbledon, she had lost just ten games. Her doubles matches, especially with net maven Elizabeth Ryan at her side, were often just as lopsided. She never missed, she could put the ball anywhere on the court, and most opponents were lucky just to win a single point.

Lenglen and Ryan in 1925 at Wimbledon. Colorization credit: Women’s Tennis Colorizations

In January 1924, Lenglen eased her way back onto the circuit. Battling some combination of illness, anxiety, and hypochondria, she didn’t return to singles action until February. (She’d win her first three matches before dropping a game.) But she was a celebrity on the French Riviera, and she was prevailed upon to compete in doubles. She won the mixed at the Hotel Beau-Site tournament in Cannes to ring in the new year, and she entered both the women’s doubles–with Ryan–and the mixed at the Hotel Gallia tournament a few weeks later.

On the 26th, Lenglen and Ryan completed their waltz through the draw, defeating a British pair, Phyllis Covell and Dorothy Shepherd-Barron, 6-3, 6-4. Suzanne’s most aggravating foe was another Brit, a line judge with the temerity to call a foot-fault on the five-time Wimbledon champion. She tried to get the man removed and ultimately had to settle for his “voluntary” departure. “It is unfair,” she said. “The English are pigs.”

Her nerves would be tested even more severely in the mixed doubles final. Lenglen partnered Charles Aeschlimann of Switzerland, while Ryan teamed with the 43-year-old Canadian Henry Mayes. Both men were better known on the Riviera than in the tennis world at large, more clubbable than talented. Lenglen and Ryan–herself one of the top few woman players in the world–would be the stars of the show.

Lenglen and Aeschlimann took the first set, 6-4; Ryan and Mayes came back with a 6-1 frame of their own. The underdogs–that is, the team without Suzanne–built up an early lead in the third, thanks to Aeschlimann’s inconsistency and Ryan’s glittering play. Mayes served for a 4-2 advantage, but a lucky netcord halted their momentum, and the deciding set settled into a rhythm it wouldn’t break for 20 more games.

Only at 13-14 did Ryan finally give in. She gifted a double fault to her opponents, and Mayes’s fatigue–he had played a four-set men’s doubles final beforehand–began to tell. Lenglen and Aeschlimann broke serve, securing the 6-4, 1-6, 15-13 victory. It would stand as the longest set of Suzanne’s unparalleled career.

* * *

Subscribe to the blog to receive each new post by email:

 

How To Play One-Set Shootouts Like Daniil Medvedev

Daniil Medvedev in 2023, practicing… something. Credit: Hameltion

In yesterday’s Australian Open quarter-final match against Hubert Hurkacz, Daniil Medvedev came through with his second five-set win of the tournament. In the decider, Hurkacz’s level dropped, Medvedev kept his ground game tight, and the Russian converted the one break point on offer. Four hours of tennis, compressed into a few crucial moments, and Medvedev has a place in the semi-finals.

Not long ago, Medvedev gained a reputation as a disappointment in deciding sets. He lost 11 of 15 three- and five-setters in 2022, and yesterday’s match was the first time in nine tries–going back to Melbourne two years ago–that he had beaten a top-ten player in a climactic set.

But such trends are easy to exaggerate. For one, three of those eight consecutive losses were clustered at the 2022 Tour Finals, where the Russian managed, remarkably, to drop third-set tiebreaks in all of his round-robin matches. Not the best way to ensure a restful offseason, but hardly an indictment of his ability to hang around late into matches with the best players in the game.

Further, except for the 2022 season, Medvedev has developed a knack for cleaning up close matches with everybody else:

Year   Decider W-L  Decider W%  
2024           2-0      100.0%  
2023          14-6       70.0%  
2022          4-11       26.7%  
2021          14-5       73.7%  
2020           9-4       69.2%  
2019         10-11       47.6%  
2018          16-9       64.0%  
2017          13-6       68.4%  
2016          23-9       71.9% 
---- 
Total       105-61       63.3%

2016 shouldn’t really count, since it’s a mix of ITFs, Challengers, and early forays onto the main tour, but given the results, I figured it was worth including. Wherever you draw the line, it’s hardly the case that Medvedev struggles in such matches. Recently, I looked into what a player’s third-set record “should” be, given their skill level, and a mark above 60% is better than expected for nearly anyone.

You might argue that the Russian shouldn’t have racked up so many deciders. He was expected to finish off Emil Ruusuvuori much more quickly than he did in the second round in Australia, and even on clay, he should never have gotten dragged to a fifth set at Roland Garros by Thiago Seyboth Wild, much less lost it. But everyone takes the scenic route sometimes. 14 of Medvedev’s deciding sets last year came against the top 50, 10 of them against the top 20.

The final set shift

When a match is reduced to a one-set shootout, it becomes a bit less serve-centric. This is a persistent finding in all high-pressure situations, from tiebreaks to break points to fifth sets. Servers get a bit more cautious, returners heighten their focus, and quick points are harder to come by.

The effects are small but real. In the 1,200-plus men’s deciding sets since 2017 logged by the Match Charting Project, servers win 1.1% fewer points in the final set that they did in the first two or four. They land fractionally more of their first serves, but only by increasing their margins: The percent of unreturned serves falls by more than 5%. The average rally increases from 4.1 strokes to 4.3.

There are two fundamental ways to benefit from those changes. First, you can buck the trend, continuing to serve big while your opponent succumbs to the natural tendency toward caution. That’s part of the reason that John Isner and Roger Federer were two of the very few players to win more tiebreaks than expected over long periods of time. It’s not easy, especially if fatigue is setting in. But if you can keep serving the way you did for two or four sets, you have a minor edge in the decider.

Second, you can be the type of player who excels in deciding-set-style tennis. If you had to pick between Medvedev and Hurkacz in a contest where more serves would come back and points would last longer, the choice is simple, right? It’s no guarantee, to be sure: The shift is a minor one, and it may not show up in any given match. Yesterday, more points were decided in four shots or less in the fifth set than in the first four. But on average, the trend moves in the other direction, right into the Russian’s wheelhouse.

Evidence shows that Medvedev follows these prescriptions, maintaining his attack on serve while taking advantage of more cautious opponents. Other top players, to varying degrees, do the same.

Let’s start with the basics. For each stat, I calculated every player’s performance in deciding sets, and in all previous sets. The numbers I’m about to show you are the ratio between those numbers, a measure of how much their tactics change when the final set begins. Positive numbers mean they do more of it in the decider, negative means they do less. We’ll look at the four Australian Open semi-finalists, plus Carlos Alcaraz (because of course) and Hurkacz (because of his deciding-set notoriety). Keep in mind that Novak Djokovic’s figures are limited to matches since 2017.

Here are the rate of serve points won, and the rate of first serves in:

Player             SPW%  1stIn%  
Carlos Alcaraz     3.9%    4.4%  
Jannik Sinner      2.6%   -1.2%  
Novak Djokovic     1.5%   -1.1%  
Hubert Hurkacz     0.8%   -1.9%  
-- Average --     -1.1%    0.7%  
Daniil Medvedev   -1.2%   -1.7%  
Alexander Zverev  -4.5%    3.2%

Medvedev is in line with tour average when it comes to winning service points: He doesn’t hold on to as many in deciding sets. Average isn’t bad in this case, though it looks mediocre in this company. A more encouraging sign, at least in terms of the tactical approach, is the change in first serves in. The Russian, in line with Djokovic, Hurkacz, and Jannik Sinner, seems to take a few more chances in the shootout. Alcaraz defies gravity, serving more conservatively yet winning more points, and Zverev looks out of place, a caricature of prudence.

Now let’s look at the percentage of serves that don’t come back (Unret%), as well as the percent of service points won in three shots or less (SPW% <=3):

Player            Unret%  SPW% <=3  
Novak Djokovic     10.9%      5.4%  
Carlos Alcaraz      0.2%      1.0%  
Daniil Medvedev    -0.6%     -2.0%  
Hubert Hurkacz     -1.1%      0.2%  
-- Average --      -5.7%     -3.6%  
Jannik Sinner      -7.4%      0.3%  
Alexander Zverev  -13.4%    -11.2%

The first rule of writing about men's tennis: Whatever the topic, you'll eventually end up showering praise on Djokovic. In recent years, he has learned how to get more out of his serve, and he turns that knob even further in deciding sets. Most players struggle to simply stay above water in the final set; Djokovic starts serving bigger.

Medvedev's rate of unreturned serves is the sort of positive sign it takes a connoisseur to appreciate: "-0.6%" doesn't turn up on many Hall of Fame plaques. But when the typical player serves so much more carefully, the Russian's consistency works to his advantage. His three-shots-or-less win rate does not stand out as much, but it is still less of a step backward than the typical tour player takes.

Once again, deciding-set Alexander Zverev is an unusual beast.

Opportunistic returning

If the challenge on serve is to keep attacking in the final set, the task on return is to take advantage of an opponent who probably isn't doing that. Ideally, that might mean more aggression on the return, but a 1% or 5% weaker first serve is still only so playable. Instead, players should make sure not to squander the chances they're given: Make more returns, then tighten up the ground game for the inevitable rallies.

Here are three stats to illustrate deciding-set return tendencies, again expressed as ratios between how each player performs in the final set, compared to previous sets:

Player            Ret InPlay%  UFE/Pt    FH%  
Alexander Zverev         6.7%    1.1%   1.0%  
Daniil Medvedev          3.9%   -3.2%  -1.2%  
Novak Djokovic           3.0%  -10.5%   1.5%  
Hubert Hurkacz           2.9%   -1.7%   0.3%  
Carlos Alcaraz           2.7%  -10.4%  -1.9%  
-- Average --            2.5%   -2.4%  -0.3%  
Jannik Sinner           -1.2%    0.1%   0.4%

Zverev, as we might have guessed, gets a lot of deciding-set returns in play. He's exceedingly conservative by every other measure we've seen, so why not here? Behind him, heading the non-pusher category, is Medvedev, who gets nearly 4% more returns in play in the final set that he did up to that point.

Unlike Zverev, the Russian also stays in control throughout the rally. He doesn't suddenly discover the otherworldly control of Djokovic and Alcaraz, who somehow reduce their unforced error rates by 10% in the deciding set, but he leads the rest of the pack, cutting down his mistakes by more than the tour average.

The third metric shown here--forehands as a percentage of all groundstrokes--is simply a curiosity. There's no right or wrong way to choose strokes, at least not at the level of the whole tour. As we saw last week, Medvedev and Zverev go for backhands on the plus-one shot more than anyone else, because they are in the unusual position that it might really be their stronger option. If a player improves his ground game in the fifth set--and this is nothing more than a hypothesis--it might show up in the numbers as more shots from his preferred wing. None of these men show a dramatic shift in shot selection, but I can't help but notice that Medvedev hits a few more backhands in the final set than he did in the two of four sets it took to get there.

If Medvedev reaches a fifth set in tomorrow's semi-final against Zverev, he won't need this level of savvy to know what's going on. The German's tactics, whether by design or instinct, are abundantly clear. Zverev can turn a shootout into a war of attrition, with two fifth-set tiebreaks already in Melbourne and an astonishing record of 22 deciding sets won in his last 26 attempts. While it will doubtless be a grind, the Russian might just be able to use his opponent's passivity against him. Faced with the tiny margins of a grand slam fifth set, every edge is worth exploiting.

* * *

Subscribe to the blog to receive each new post by email:

 

How Coco Gauff Escaped a Trap of Her Own Making

Also today: Jannik Sinner’s near-unbreakability

Coco Gauff at the 2022 US Open. Credit: All-Pro Reels

Coco Gauff is not a pusher, but she can do an awfully good impression of one. In yesterday’s Australian Open quarter-final against Marta Kostyuk, the American coughed up 50 unforced errors against just 17 winners. The average rally lasted 4.6 strokes, a modest number that was rescued from marathon territory only by Gauff’s many unreturned serves.

Coco’s forehand, the usual culprit when things get messy, was on full display yesterday. While the stroke has shown signs of improvement–only 9% of them contributed to the unforced error tally, below both tour average and Gauff’s own standard–it remains loopy, and it gets ever-more cautious under pressure. Kostyuk was willing to go after the high-bouncing mid-court groundstrokes, often putting Gauff on the run. Fortunately for the American, her defense rarely deserts her. She eked out a three-hour, 7-6, 6-7, 6-2 victory for a place in the Melbourne semi-finals.

My impression watching the match was that Gauff put an unreasonable number of returns–especially forehand returns–in the middle of the court, not too deep, and that Kostyuk was punishing them. I was partly right: The Ukrainian forced Coco to hit forehand after forehand against the serve, more than two-thirds of her service returns all told. Gauff did indeed send more of those balls down the middle, closer to the service line than the baseline. And Kostyuk attacked… but to little avail.

Let’s get into the numbers. The Match Charting Project divides the court into thirds, both in terms of direction (forehand side, backhand side, and middle) and depth (shallow [in the service boxes], deep [closer to the service line than the baseline], and very deep). All else equal, shots deep and/or to the sides of the court are better, though of course they are riskier. Some returns will inevitably end up down the middle and shallow; the goal is only to avoid it when possible.

Here is how Gauff’s performance yesterday compared to tour average and her own typical rate of service returns that went down the middle and didn’t land close to the baseline:

RETURNS          Middle/Not Very Deep  
Tour Average                    34.0%  
Coco Average                    40.5%  
Coco vs Kostyuk                 43.7%

Indifferent return placement is nothing new for the American, and she left even more hittable plus-ones for Kostyuk than usual. It wasn’t as bad as last year’s US Open final against Aryna Sabalenka, when Gauff put more than half of her returns in the less effective zones, but Kostyuk is no Sabalenka when it comes to imposing her will with the serve.

Return placement matters. On average, tour players win 46% of points when they land a down-the-middle, not very deep return. When they put the ball anywhere else–closer to the baseline or a sideline–they win 56%. Gauff is a little better behind the weak returns, but for her career, the gap is still present: 47% versus 55%.

Except… that isn’t what happened yesterday!

RETURN OUTCOMES  Mid/NVP W%  Better W%  
Tour Average          46.2%      56.3%  
Coco Average          46.9%      54.8%  
Coco vs Kostyuk       60.0%      55.2%

When Gauff placed a return near a line, her results yesterday were typical. But Kostyuk was unable to capitalize on the rest. Among 88 matches logged by the Match Charting Project, Gauff has won 60% of those middle/not-very-deep returns only a dozen times, usually in blowouts.

Judging from the American’s performance on return, she could have made quick work of yesterday’s contest, too. The sticking point came on her own side of the ball, where her non-committal forehands didn’t work out as well.

Minus-ones

On the WTA tour, when the return lands in play, the server has nearly lost her advantage. A good first serve can give her a lingering edge, or a well-placed return can tilt the balance in the other direction, but overall, the point begins again as a neutral proposition. Servers win 52% of those points.

Gauff, on average, does a little better, converting her serve 53% of the time. There are signs she’s improving, as well. In the US Open final against Sabalenka, she won 55%, and in the Auckland final this month versus Elina Svitolina, she picked up 59%. Apart from lopsided matches, the high-50s are the best anyone can do on an ongoing basis: Iga Swiatek’s average is 57%, and Sabalenka’s is 55%.

Coco won 39% against Kostyuk.

Gauff’s lack of confidence in her forehand showed up in multiple ways. First, she didn’t use it as much as a plus-one weapon. She usually hits 57% of her plus-one shots from the forehand side, in line with tour average. Yesterday, that rate was just 51%, something that had more to do with her own choices than any return magic that Kostyuk conjured up.

Then, she didn’t do much with those forehands. The following table shows plus-one forehand rates (3F%), the percentage of plus-one forehands hit down the middle (FH Mid%), and the server’s winning percentage (FH Mid W%) behind those down-the-middle forehands:

PLUS-ONES          3F%  FH Mid%  FH Mid W%  
Tour Average     56.6%    29.9%      45.9%  
Coco Average     57.2%    35.0%      47.0%  
Coco vs Kostyuk  50.7%    39.5%      40.0%

Gauff magnified her own tendency to go back down the middle with her second-shot forehand. It didn’t work, as she won just 40% of those points, compared to her typical rate of 47%.

Even beyond the plus-one, Coco just kept pushing the forehand. She went down the middle with 46% of her forehands, compared with her usual 37% and the tour average of 28%. She won barely one-third of the points when she did so, partly because of the nine unforced errors she racked up playing an already conservative shot. Two of those missed down-the-middle forehands came on back-to-back points when she could hardly afford them, taking her from 15-all to 15-40 when trying to close out the match at 5-3 in the second set.

In the end, as we’ve seen, Gauff’s defense saved her. She won more than half of Kostyuk’s serve points despite lackluster returning. Had she served just a little better–she missed six straight first serves in that 5-3 game–she would have finished the job an hour sooner. Had she attacked a bit more effectively with her second shots, even the off-day from the line wouldn’t have amounted to much.

To state the obvious: She’ll have to play better to beat Sabalenka in tomorrow’s semi-finals. One thing, at least, will work in Coco’s favor: She’ll have many fewer choices to make. The defending champion will dictate play and give her less time to think than Kostyuk did. Gauff withstood the Belarusian barrage in New York, winning the US Open title despite a couple of detours against less aggressive players in the early rounds. The American can’t play tomorrow like she did yesterday, but thankfully, Sabalenka won’t let her.

* * *

Jannik Sinner’s near-unbreakability

Jannik Sinner has lost his serve just twice en route to the Australian Open semi-finals. He has faced 28 break points and saved 26 of them.

Since 1991, when the ATP started keeping the relevant stats, he is the 26th player to reach the final four at a major with so few breaks of his own serve:

Tournament  Semi-finalist       BP Faced  Broken  
2013 USO    Rafael Nadal               6       0  
2018 Wimb   John Isner                 7       0  
2015 Wimb   Roger Federer              3       1  
1994 Wimb   Pete Sampras               9       1  
2015 AO     Novak Djokovic            11       1  
2014 Wimb   Roger Federer             12       1  
1997 Wimb   Pete Sampras              12       1  
2010 USO    Rafael Nadal              14       1  
2012 RG     Rafael Nadal              17       1  
2004 Wimb   Roger Federer             17       1  

Tournament  Semi-finalist       BP Faced  Broken  
2014 Wimb   Milos Raonic               9       2  
2011 RG     Novak Djokovic*            9       2  
2007 USO    Roger Federer              9       2  
2006 Wimb   Roger Federer              9       2  
2006 Wimb   Rafael Nadal               9       2  
2015 USO    Roger Federer             11       2  
2014 AO     Roger Federer             11       2  
1997 USO    Greg Rusedski             11       2  
1993 AO     Pete Sampras**            12       2  
2013 Wimb   JM del Potro              13       2  
2019 AO     Rafael Nadal              15       2  
2008 Wimb   Roger Federer             15       2  
2005 AO     Andy Roddick              15       2  
1998 Wimb   Pete Sampras              17       2  
2000 AO     Yevgeny Kafelnikov        22       2  
2024 AO     Jannik Sinner             28       2

* Djokovic won one round by W/O and another by retirement
** I don't have stats for Sampras's QF, but the final score suggests that he wasn't broken

Pretty good company! As the table makes clear, though, Sinner’s 28 break points faced is not so elite. In fact, the average major semi-finalist faces exactly 28 break points in his first five matches.

The Italian’s accomplishment, then, is saving so many. 26 of 28 is a 93% clip, and that is more rarefied air:

Tounament  Player      Faced  Saved   Save%  
2013 USO   Nadal           6      6  100.0%  
2018 Wimb  Isner           7      7  100.0%  
2012 RG    Nadal          17     16   94.1%  
2004 Wimb  Federer        17     16   94.1%  
2010 USO   Nadal          14     13   92.9%  
2024 AO    Sinner         28     26   92.9%  
2014 Wimb  Federer        12     11   91.7%  
1997 Wimb  Sampras        12     11   91.7%  
2015 AO    Djokovic       11     10   90.9%  
2000 AO    Kafelnikov     22     20   90.9%

Things will get tougher on Friday, when Sinner faces all-time-great returner Novak Djokovic for a place in the final. Then again, Djokovic failed to convert his first 15 break points against Taylor Fritz yesterday–maybe he was just preparing for the matchup with Sinner.

* * *

Subscribe to the blog to receive each new post by email:

 

Anna Kalinskaya At Her Peak

Also today: Upsets, (partly) explained; January 23, 1924

Anna Kalinskaya in the 2020 Fed Cup qualifying round. Credit: Nuță Lucian

Should we have seen this coming? Of all the surprises in the top half of the 2024 Australian Open women’s draw, Anna Kalinskaya’s run to the quarter-finals stands as one of the biggest. The 25-year-old was ranked 75th entering the tournament, and she had never reached the third round of a major in 13 previous main-draw attempts.

Had we looked closely before the tournament, we wouldn’t have found a title contender, exactly, but we would have identified Kalinskaya as about as dangerous as a 75th-ranked player could possibly be. She finished 2023 on a 9-1 run, reaching the final at the WTA 125 in Tampico, then winning the title at the Midland 125, where she knocked out the up-and-coming Alycia Parks in the semi-finals. 2024 started well, too: The Russian upset top-tenner Barbora Krejcikova in Adelaide, then almost knocked out Daria Kasatkina in a two hour, 51-minute match two days later.

The only reason her official ranking is so low is that she missed nearly four months last summer to a leg injury that she picked up in the third round in Rome. Her two match wins at the Foro Italico pushed her up to 53rd in the world, just short of her career-best 51st, set in 2022. The Elo algorithm, which measures the quality of her wins rather than the number of tournaments she was healthy enough to play, reflects both her pre-injury successes and the more recent hot streak. Kalinskaya came to Melbourne as the 31st-ranked woman on the Elo list.

These alternative rankings put a different spin on her path through the Australian Open draw so far. Here are the results from her first four rounds, in which she appeared to be the underdog three times:

Don’t be fooled!

Elo has some adjustments to make:

Round  Opponent  Elo Rk  Elo vRk  
R16    Paolini       31       37  
R32    Stephens      31       50  
R64    Rus           31      107  
R128   Volynets      31      139

Kalinskaya was hardly an early favorite–Stephens did her the favor of taking out Kasatkina, and Anna Blinkova (who lost to Paolini) eliminated the third-seeded Elena Rybakina. But given how the draw worked out, seeing the Russian’s name in the quarter-finals wasn’t so unlikely after all.

More luck

Kalinskaya has a dangerous forehand and a solid backhand, but she isn’t an aggressive player by the standards of today’s circuit. Her 14 matches logged by the Match Charting Project average 4.2 strokes per point, and that skews low because it includes three meetings with Aryna Sabalenka. Yesterday’s fourth-round match against Paolini took 5.3 strokes per point, and the third-rounder with Stephens was similar.

By Aggression Score, the 25-year-old rates modestly below average, at -17 in rallies and -15 on returns. While she doesn’t have any weaknesses that prevent her from ending points earlier, she’s more comfortable letting the rally develop. When Paolini played along, the results were remarkable: 32 points reached seven shots or more yesterday, and Kalinskaya didn’t end any of them with an unforced error.

The downside of such a game style is that a lot of opponents won’t be so cooperative. Last fall, the Russian lost back-to-back-to-back matches against Ekaterina Alexandrova, Viktoria Hruncakova, and Ashlyn Krueger, three women who opt for big swings and short points. By contrast, consider the Rally Aggression Scores of the quartet Kalinskaya has faced in Melbourne:

Round  Opponent  AggScore  
R16    Paolini         -5  
R32    Stephens       -16  
R64    Rus            -59  
R128   Volynets       -38

Paolini and Stephens have roughly similar profiles to Kalinskaya’s own; Rus and Volynets are even more conservative.

This isn’t just a convenient narrative: Kalinskaya really is better against more passive players. She has played 118 career tour-level matches against women with at least 20 matches in the charting database. Sort them by Rally Aggression Score and separate them into four equal bins, and the Russian’s preferences become clear:

AggScore Range  Match Win%  
57 to 175            35.7%  
0 to 56              46.4%  
-27 to -1            50.0%  
-137 to -27          59.4%

If the whole tour were as patient as she is, the Russian would already be a household name.

Alas, it’s rare to draw four straight players as conservative as the bunch Kalinskaya has faced in Melbourne. And having reached the quarter-finals, her luck has run out. Her next opponent is Qinwen Zheng, who has a career Aggression Score of 27 and upped that number in 2023. It could be worse–fellow quarter-finalists Sabalenka and Dayana Yastremska are triple-digit aggressors–but it is a different sort of challenge than she has faced at the tournament so far.

To win tomorrow, Kalinskaya will need to play as well as she has for the last few months, only a couple of shots earlier in the rally. Otherwise, Zheng will end points on her own terms, and thousands of potential new fans will be convinced that Kalinskaya really is just the 75th best player in the world.

* * *

Why are upsets on the rise?

Only four seeds, and two of the top eight, survived to the Australian Open women’s quarter-finals. Many of the top seeds lost early. This feels like a trend, and it isn’t new.

One plausible explanation is that the field keeps getting stronger. Top-level players now develop all over the world, and coaching and training techniques continue to improve. There are few easy, guaranteed matches, even if Iga Swiatek and Aryna Sabalenka usually(!) make it look that way. I believe this is part of the story.

Another component, I suspect, is the shift in playing styles. I noted a couple of weeks ago when writing about Angelique Kerber is that WTA rally lengths have steadily declined in the last decade. In 2013, the typical point lasted 4.7 strokes; it’s now around 4.3. Shorter points are caused by more risk-taking. Risks don’t always work out, full-power shots go astray, and the better-on-paper player doesn’t always win.

In 2019, I tested a similar theory about men’s results. I split players in four quartiles based on Aggression Score and tallied the upset rate for every pair of player types. When two very aggressive players met, nearly 39% of matches resulted in upsets, compared to 25% when two very passive players met. The true gap isn’t quite that big: given the specific players involved, there should have been a few more upsets among the very aggressive group. But even after adjusting for that, it remained a substantial gap.

It stands to reason that the story would be the same for women. Instead of Aggression Score, I used average rally length. I doubt there’s much difference. I didn’t intend to change gears, I just got halfway through the project before checking what I did the first time.

The most aggressive quartile (1, in the table below) are players who average 3.6 shots per rally or less. The next group (2) ranges from 3.7 to 4.0, then (3) from 4.1 to 4.5, and finally (4) 4.6 strokes and up. The following table shows the frequency of upsets (Upset%) and how the upset rate compares to expectations (U/Exp) for each pair of groups:

Q1  Q2  Upset%  U/Exp  
1   1    40.7%   1.07  
2   1    36.2%   0.99  
2   2    35.7%   0.99  
3   1    35.1%   0.93  
3   2    35.5%   0.97  
3   3    40.9%   1.07  
4   1    37.6%   1.03  
4   2    36.6%   1.02  
4   3    34.6%   0.95  
4   4    34.7%   0.97

(If you look back to the 2019 study, you’ll notice that I did almost everything “backwards” this time — swapping 1 for 4 as the label for the most aggressive group, and calculating results as favorite winning percentages instead of upsets. Sorry about that.)

Matches between very aggressive players do, in fact, result in more upsets than expected. It’s not an overwhelming result, partly because it’s only 7% more than expected, and partly because matches between third-quartile players–those with average rally lengths between 4.1 and 4.5–are just as unexpectedly unpredictable.

I don’t know what to make of the latter finding. I can’t think of any reasonable cause for that other than chance, which casts some doubt on the top-line result as well.

If the upset rate for matches between very aggressive players is a persistent effect, it would give us more upsets on tour today than we saw a decade ago. An increasing number of players fit the hyper-aggressive mold, so there are more matchups between them. The logic seems sound to me, though it may be the case that other sources of player inconsistency outweigh a woman’s particular risk profile.

* * *

January 23, 1924: Debuts and dropshots

Men’s tennis ruled at the early Australian Championships. The tournament had been held since 1905 (as the “Australasian” Championships), but there was no women’s singles until 1922. On January 23rd, midway through the 1924 edition, the press corps was preoccupied with the severity of Gerald Patterson’s sprained ankle and the question of whether Ian McInnes had been practicing.

James O. Anderson, the 1922 singles champion who would win the 1924 edition as well, introduced what was then–at least to the Melbourne Argus–an on-court novelty:

He has developed a new stroke since he last played in Melbourne, and it has proved successful. On the back of the court he makes a pretence of sending in a hard drive, but with a delicate flick of the wrist he drops the ball just over the net, leaving his opponent helpless 30 feet away.

A veritable proto-Alcaraz, was James O.

For the few fans who weren’t solely focused on Australia’s Davis Cuppers, a superstar was emerging before their eyes. Also on the 23rd, 20-year-old Daphne Akhurst made quick work of Violet Mather, advancing to the semi-finals in her first appearance at the Championships.

Akhurst wouldn’t go any further, unable to withstand the heavy forehand of Esna Boyd in the next round. But it was nonetheless a remarkable debut: She won both the women’s and the mixed doubles titles. The correspondent for the Melbourne Age, recapping the mixed final, could hardly contain his admiration:

Miss Akhurst–an artist to her finger tips–belied her delicate mid-Victorian appearance that suggested that she had slipped out of one of Jane Austen’s books by sifting out cayenne pepper strokes from a never-failing supply.

Daphne and Jack Willard–“who ran for every ball, and continued running after he played the ball”–defeated Boyd and Gar Hone in straight sets.

The pair of championships was a harbinger of things to come. Between 1925 and 1931, Akhurst would win five singles titles (losing only in 1927 when she withdrew), four more in the women’s doubles, and another three mixed. The only thing that could stop her were the customs of the day: She married in 1930 and retired a year later. Tragically, she died from pregnancy complications in 1933, at the age of 29.

Daphne is best known these days as the name on the Australian Open women’s singles trophy. For the next several years, there will be many more Akhurst centennials to celebrate.

* * *

Subscribe to the blog to receive each new post by email:

 

Andrey Rublev, Grand Slam Quarter-finalist

Also today: Jannik Sinner’s rosy forecast; lopsided fifth sets

Andrey Rublev at Wimbledon in 2023. Credit: aarublevnews

Andrey Rublev is a known quantity. He will hit big first serves, but his second serves can be attacked. He will hit monster forehands, often venturing far into his backhand corner to play them, and his opponents will often be stuck in place, watching them go by. He’ll also miss a lot of them. His backhand isn’t the same type of offensive shot; he can be dragged into long rallies if you pepper that side.

There isn’t a lot of subtlety to his game. That isn’t a criticism: Subtlety can win you acolytes and endorsement deals, but it isn’t necessary to win championships. With yesterday’s five-set win in Australia over home hope Alex de Minaur, Rublev advanced to his tenth career grand slam quarter-final. He’s 0-9 so far in those matches, but his consistency in getting there is the bigger story. Alexander Zverev is the only other man under the age of 30 with ten major quarter-finals. Rublev will get on the board eventually.

What you might not know about the 26-year-old Russian is that he has matured into a reliably dangerous returner. He’s always been effective on that side of the ball, and his return numbers have remained steady as the strength of his competition has increased. Last year, he won nearly 39% of his return points, good for 3.2 breaks of serve per match–seventh-best on tour. At the 2023 US Open against Daniil Medvedev, his most recent attempt to reach a major semi-final, Rublev broke serve five times in his straight-set defeat. The return wasn’t the problem.

That day, Medvedev’s return was the problem. (Andrey’s second serve didn’t do him any favors either, but that’s nothing new.) Of Rublev’s 98 serve points, 65 of them lasted four shots on longer. I can’t emphasize enough how bizarre that is–or, seen from another perspective, what a performance it was from his opponent. Medvedev not only got 65 serves in play, he got 65 plus-one shots back. Rublev’s top two weapons were negated.

The standard Rublev performance, at least among the 138 matches logged by the Match Charting Project, involves 59% of his service points ending by the third shot. He wins just over three-quarters of those. (Against Medvedev, he tallied a respectable 70%, but 70% of not very many is still not very many.) Put those numbers together, and 45% of his serve points end in his favor in three shots or less.

That’s a pretty good head start! Last year, the Russian won 66% of his total serve points. The majority of the damage gets done early.

The serves and plus-ones not only account for a decent chunk of the points played–at least on a good day–but they also serve as a proxy for how the longer rallies turn out. When Rublev wins most of his short service points–even when he doesn’t play as many as he would like–he almost always comes out on top. If we sort his charted matches by winning percentage on short service points, then split them into thirds, the difference is stark:

<=3 SPW%       Matches  Match Win%  
81%+                45         87%  
75.5% - 80.9%       44         64%  
up to 75.4%         49         24%

(The buckets are slightly different sizes only because I didn't want to put nearly identical percentages into separate categories.)

When Rublev wins most of the short service points, he wins the match. When he doesn't, he usually loses. If anything, the table understates the contrast; a disproportionate number of the low-percentage victories came on clay, including several on the slow dirt of Monte Carlo.

To some extent, it's obvious that "winning more of some subset of points" correlates with "winning more of all the points" and thus winning the match. But remember, this is the success rate independent of how many points end quickly. The combination of frequency and success--"what percent of total service points end quickly and in the server's favor"--should tell us more about the overall result. But for Rublev, that metric isn't as predictive of final outcomes as the winning percentage alone.

Battling demon

Yesterday against de Minaur, Rublev won 82% of the short service points. The Australian kept it close by reducing the number of short points to just under half of Rublev's serves. But the rule I've just outlined held true, despite a pesky defense. When de Minaur put the fourth shot back in play, he won 57% of return points. That's great, but with Rublev cleaning up the overwhelming majority of the short points, it wasn't enough.

We have shot-by-shot logs for four of the six matches between these two guys:

Tournament        Result  Short%  Short W%  
2024 Australian        W   49.4%     82.0%  
2023 Rotterdam         L   60.3%     75.6%  
2022 Monte Carlo       W   42.7%     73.2%  
2018 Washington        L   53.2%     71.6%

De Minaur did his job yesterday, keeping the ball in play more often than he did in the two previous hard-court meetings. (The Monte Carlo surface presumably helped lengthen points in that match.) The Australian won both of those earlier contests, watching Rublev make more plus-one mistakes and taking care of business when the rallies lasted longer.

In Melbourne, the Russian stayed a bit more within himself. He was able to hit a forehand on barely half of his plus-one shots--below both tour average and his own typical rate. Instead of blasting away with ill-advised backhands--part of what lost him the Rotterdam match--he accepted the invitation to rally. His 43% rate of winning longer service points isn't great, but it's far superior to the 0% chance of claiming the point after smacking an unforced error.

I don't want to overstate Rublev's caution, because he didn't play a cautious match. He probably never should. But getting a few more balls in play and fighting out the ensuing rallies makes his second serve look a lot better. As we've seen, Rublev does well on return. His second-serve points aren't much better than return points... but that's okay! Yesterday he won 55% behind his second serve, a glittering result compared to the 37% and 38% he won against de Minaur in Washington and Rotterdam, respectively.

Is this the one?

Rublev can be forgiven for having a losing record in major quarter-finals; he's been the lower-ranked player in seven of the nine. He's dropped two to Novak Djokovic, one to Rafael Nadal, and three to Medvedev. He should have picked up one (or three) along the way, but as the fifth man on a tour that always seems to have a big three or four, it's an uphill struggle.

Tomorrow's opponent is Jannik Sinner, just one place above him in the ATP rankings. (Elo likes him more than that--a lot more. See below.) This will be their seventh meeting, and history doesn't bode well for the Russian. Sinner has retired twice but won the other four.

Here are the short-service-point stats for Rublev in three of those matches:

Tournament        Result  Short%  Short W%  
2023 Miami             L   62.5%     77.1%  
2022 Monte Carlo       L   41.0%     58.5%  
2021 Barcelona         L   43.8%     85.7%

(Unfortunately we don't yet have a chart of his 7-6, 7-5 loss last fall in Vienna.)

This isn't insurmountable for the Russian: He often wins matches behind 77% of his short service points, and he almost always does with a 86% win rate. He'd like more than 44% of his serve points to end quickly, but that's tougher to execute on clay.

Against Sinner, the first three shots are even more important than usual, because the Italian plays a similar game, and once a rally reaches four strokes, he plays that game better. In Miami, Sinner won two-thirds of Rublev's "long" service points. In Monte Carlo, he won 54%, in the vicinity of what de Minaur did yesterday. In Barcelona, Sinner won a whopping 70% of return points when he got the fourth shot in play--as he more often than not did.

Rublev's second serves tell the story, as they did in the de Minaur match. Those, typically, are the points he can't finish early, when he should be thinking in terms of constructing the point, not grunting and crushing. In the four completed Sinner matches, he won only 37.5% of second-serve points. That's not going to get it done.

To beat an elite opponent, Rublev needs to remember when to bash and when to think. He executed well yesterday, pulling away in the end against a man who never stops fighting. Reaching his first major semi-final, against 22-year-old who seems to get stronger every week, he'll need to play even better.

* * *

Sinner in the hands of a friendly forecast

Jannik Sinner is the favorite tomorrow: According to my Elo-based forecast, he has a 78% chance of advancing to the final four. That's a hefty margin for a match between players adjacent to one another in the official rankings. The difference is more about Sinner than Rublev: My forecast gives Sinner a nearly 30% shot at taking the title, second only to Djokovic.

While the Italian ranks fourth on the ATP computer, he's second according to the Elo algorithm, closer to Djokovic than anyone else is to him. Here is the top of the table entering the Australian Open:

Rank  Player             Elo  
1     Novak Djokovic    2217  
2     Jannik Sinner     2197  
3     Carlos Alcaraz    2149  
4     Daniil Medvedev   2104  
5     Alexander Zverev  2037  
6     Andrey Rublev     2035  
7     Grigor Dimitrov   2032

If you think in terms of major titles, official ranking points, or hype, this probably seems wrong. By those measures, Sinner is the laggard among the top four.

But Elo gives credit based on the quality of opponents beaten, and Sinner built quite a resume in the last quarter of 2023. He beat Rublev, Alcaraz, Medvedev (three times!), and most important, Djokovic twice. Nothing catapults you up the Elo list faster than knocking off the top dog.

The question, then, is whether Elo has overreacted to those two victories. My implementation of the Elo algorithm doesn't differentiate between narrow wins and blowouts. (Other versions use sets, games, or even points, though in my testing, those alternatives don't make the ratings more predictive.) The two Djokovic upsets were nail-biters. The Tour Finals round-robin match was decided in a third-set tiebreak, and each man won exactly 109 points. At the Davis Cup Finals, Sinner took the third set 7-5 despite winning fewer total points than his opponent.

While Sinner certainly deserved those victories--staring down match point against a 24-time major winner is a feat in itself--we might wonder how much they tell us about future results. If the two men keep fighting out such close matches, Djokovic is going to win some of them.

Each of the two upsets were worth a gain of 15 Elo points. Had Sinner lost them, he would've dropped 10 or 11 points instead. Call it a 25-point swing for each match. Thus, if we take the most pessimistic possible route and give both of the dead-heat results to Djokovic, Sinner's Elo rating would stand about 50 points lower, roughly tied with Alcaraz around 2,150.

(That isn't exactly right, because if Djokovic had won the Davis Cup match, Italy wouldn't have advanced to the final, and Sinner would've have beaten de Minaur. But Sinner did beat de Minaur, handily, and if we want to assess his current level, we shouldn't ignore that match.)

Handing both of the close results to Djokovic seems extreme. If we want to measure each player's current level without putting too much weight on the tiny number of points that decided those two matches, we might give one of the two victories to Djokovic. That would knock Sinner down to about 2,172, while boosting Djokovic to around 2,225.

In the Australian Open title-chances forecast, Novak would look a little better, and there would be more daylight between him and Sinner. Still, unless we make the harshest possible adjustment to Sinner's Elo rating, the Italian remains the next most likely Melbourne champion and a heavy favorite against Rublev tomorrow.

* * *

Dessert bagels

The Rublev-de Minaur match had an unusual ending: After splitting four sets, the Russian ran away with the fifth, 6-0.

Typically, if two players are so evenly matched that they reach a fifth set, neither one is going to dominate the decider. For the rare occasions that it happens, it's unique enough that I think it deserves its own name. I propose "dessert bagel."

In grand slam competition since 1968, there have been just 159 dessert bagels, including Rublev's--fewer than one per major. No one has ever recorded a dessert bagel in a final, but it has happened twice in semis. Mats Wilander polished off Andre Agassi in the 1988 Roland Garros semi-final, and Djokovic finished his 2015 Australian Open semi against Stan Wawrinka the same way. Still, second-week dessert bagels are rare: Rublev's was only the 16th in more than a half-century.

It's an oddity piled on oddities: Rublev-de Minaur was the fifth dessert bagel in Melbourne this year:

Round  Winner      Loser       Score                
R128   Mannarino   Wawrinka    6-4 3-6 5-7 6-3 6-0  
R64    van Assche  Musetti     6-3 3-6 6-7 6-3 6-0  
R64    Medvedev    Ruusuvuori  3-6 6-7 6-4 7-6 6-0  
R32    Kecmanovic  Paul        6-4 3-6 2-6 7-6 6-0  
R16    Rublev      de Minaur   6-4 6-7 6-7 6-3 6-0

Five 6-0 deciders is a record for a single slam. There haven't been as many as three since the 2007 Australian, and no major has seen more than one since 2017. If even more dessert bagels start piling up in the quarter-finals, we'll know that something bizarre is going on Down Under.

* * *

Subscribe to the blog to receive each new post by email:

 

Qinwen Zheng’s Serve Under Construction

Also today: The odds of a 42-point tiebreak; January 19, 1974

Qinwen Zheng in 2023. Credit: Hameltion

Qinwen Zheng is one of the top prospects in the women’s game, up to 14th on the WTA ranking list at age 21. She won her first tour-level title in Palermo last summer, then upset Ons Jabeur en route to a quarter-final showing at the US Open. After topping Barbora Krejcikova for a second title in Zhengzhou, she reached the final at the WTA Elite Trophy, falling in a two hour, 52-minute final to Beatriz Haddad Maia.

With yesterday’s upsets of Elena Rybakina and Jessica Pegula at the Australian Open, Zheng’s draw opened up. With only one other seed in the second quarter, she’s the heavy favorite to earn a semi-final date with Iga Swiatek. Potential is poised to become reality.

It’s never been difficult to dream big on the Chinese woman’s behalf. Her service motion–once she gets past a hitchy toss–is a photographer’s dream, and she takes advantage of her five-foot, ten-inch frame to send first serve after first serve into the corners. When she hits a target out wide, returners are lucky to get a racket on the ball, let alone put it back in play. Her forehand is equally powerful.

The results bear out the devastation wreaked by her first delivery. Here are last year’s WTA top ten in first-serve percentage:

Player               1stWon%  
Qinwen Zheng           73.7%  
Elena Rybakina         73.6%  
Aryna Sabalenka        72.8%  
Caroline Garcia        72.5%  
Liudmila Samsonova     71.5%  
Iga Swiatek            70.0%  
Petra Kvitova          69.8%  
Belinda Bencic         69.5%  
Petra Martic           69.5%  
Ekaterina Alexandrova  69.4%

Pretty good company, huh? Her forehand grades well, too. According to Match Charting Data, Zheng hits more winners, induces more forced errors, and commits fewer unforced errors with that shot than the average player on tour. Her forehand potency (FHP) per match over the last 52 weeks is 10.8, placing her in the top ten among tour regulars, just behind Haddad Maia and Madison Keys.

That’s the good news. If you’re going to have just two world-class weapons, those are the ones to pick. They’ve served her well so far: If she justifies her seed and reaches the final four in Melbourne, she could crack the top ten.

The rest of Zheng’s game is–let’s be optimistic here–a work in progress. Today I want to look specifically at her serve as a whole; we’ll save her not-as-problematic backhand for another day.

When the 21-year-old lands her first serve, as we’ve seen, good things happen. She hits more aces than almost anyone on tour, and about half of her first-serve points end with either an unreturned first serve or a plus-one winner. The problem is, she doesn’t make many first serves, and when she misses, her second serve is as erratic as her first serve is imposing.

The average top-50 player on the WTA tour makes about 62% of her first serves. In 2023, Zheng succeeded just 51.8% of the time, almost three full percentage points below anyone else.

Making matters worse, her second-serve results are nearly as bad. The average top-50 WTAer wins 47% of her second-serve points. Zheng won 45.5%, a mark that places her in the bottom third of that group. Among the current top 20, only Jelena Ostapenko and Daria Kasatkina win fewer second-serve points. It’s even worse against a strong opponent. She hung onto just 20% of second-serve points against Swiatek in the United Cup this month, 24% versus Rybakina in Beijing, and a mere 26% against Liudmila Samsonova in Montreal. Zheng’s primary weapon makes her look like an elite server, but the overall picture is more mundane. Her first serve sets her on a level with Rybakina, but she barely holds serve as often as Petra Martic.

What is to be done?

This seems like it should be fixable, especially in so young a player. It’s certainly easy to dream. Imagine the seemingly-modest scenario in which Zheng manages to land her first serves and win second-serve points at the rates of an average top-50 player while maintaining her dominance on firsts. She would then win 63.5% of her service points. Only Swiatek and Sabalenka do better.

Easier said than done, of course. A good first serve is no guarantee of a strong second. On the women’s tour, there almost zero correlation between first-serve and second-points won.

Still, this seems like partly a tactical failure, not entirely a gap in her skillset. If Zheng can win nearly 74% of her first-serve points when she misses almost half of the time, what would happen if she served a bit more conservatively? Perhaps she could make 57% of her first serves and still win 72% of them? If so, that would be a bit better. Could she make 62% of first serves–the tour average rate–and win 70% of them? That would be better still.

Once we assume that these tradeoffs are feasible, the whole thing starts to sound like less of a tactical question and more of a pure math problem. I’m not sure that it is: Players practice various types of “first serves” and “second serves,” not every theoretical delivery on the continuum between them. Maybe a thoughtful veteran could tweak things to increase or decrease her first-in percentage at will, but I’m skeptical that a young player could do th esame. At the very least, it’s a project that would take some time.

Still, it’s worth working out whether Zheng could get more bang for her serve-talent buck. In 2009, Dutch researchers Franc Klaassen and Jan R. Magnus (henceforth K&M) published a paper in the Journal of Econometrics that proposed to answer this sort of question. They worked out the usual relationship between serving risk (how many first serves in, how many double faults) and reward (rate of first- and second-serve points won). My friend Jeff McFarland converted their rather complex algorithm to a spreadsheet, which is why I’m able to publish this today, and not in March. Thanks Jeff!

The following table shows Zheng’s actual 2023 results along with her model-optimized rates:

         1stIn%  1stWon%    DF%  2ndWon%   SPW%  
Actual    51.8%    73.7%   6.0%    45.5%  60.1%  
Optimal   60.5%    70.9%   8.8%    47.5%  61.7%

K&M’s formula estimates that Zheng could get close to a tour-average level of first serves in and still win about 71% of them, a success rate that would keep her in the top five. The more surprising output is that she could do better by taking more chances on her second serve. (This is a kind of light version of the oft-discussed argument that a player should just hit two first serves. The algorithm recommends some degree of this for most pros.) By adopting the more risky second-serve approach, she would in theory win 47.5% of those points despite the increase in double faults.

Altogether, those changes would increase her total service points won from 60.1%–12th among the current top 50–to 61.7%, which would rank her fifth.

Another way of looking at the potential gain is in points per thousand. For every thousand service points played, the fully-optimized version of Zheng would win about 16 more than she does now. If her return game remained the same, that’s an improvement of eight points per thousand overall. A few years ago I stumbled on a neat rule of thumb, that an improvement of one point per thousand translates into a gain of one place in the rankings, except near the very top. If that held in this case, the re-imagined Zheng would be on the cusp on the top five.

Again, this is all theoretical. I have no idea whether a big server could consciously execute a decision to take slightly fewer chances on the first and more on the second, or whether her results would follow the model if she did.

But! This is a potential route to a jump up the rankings without reworking groundstrokes, getting fitter or stronger, or even gaining experience. It’s probably not easy, but it’s likely simpler than the alternatives. As it stands now, Zheng’s second serve–and the frequency she’s forced to hit it–is going to hold her back. Solve that problem, and much of her obvious potential is unlocked.

* * *

The odds of a 42-point tiebreak

“10-point tiebreak, my ***.” Credit: @hardpicstennis

Yesterday, Elena Rybakina and Anna Blinkova played a 42-point tiebreak. It’s the longest breaker in grand slam singles history. Blinkova won it, 22-20.

What are the odds?

Let’s start with simply getting to 9-all. We’ll assume that Rybakina and Blinkova were playing at the same level. Yes, Rybakina was a heavy favorite entering the match, and she won a few more points than Blinkova to get to 6-4, 4-6, 6-all. But the margin was narrow, and the math is simpler if we assume they are equal. They won serve points throughout the match at about a 59% rate. Since players tend to be more conservative during tiebreaks, returners fare better, so we’ll say that whoever is serving had a 55% chance of winning the point.

I ran a Monte Carlo simulation to find the odds of reaching 9-all. Here are those probabilities, along with odds at various other levels of serve dominance:

SPW   Reach 9-all  
55%         10.0%  
60%         10.3%  
65%         11.2%  
70%         12.5%  
80%         17.0%

Roughly speaking, there was a one-in-ten chance that yesterday’s breaker would reach 9-all.

From there, the math is simpler. There are two ways to get from 9-all to 10-all: both women could win their service points, or both could win their return points. Serving at 55%, the chances that one or the other occurs are 50.5%. The same logic applies to the step from 10-all to 11-all, 11-all to 12-all, and so on. So for Rybakina and Blinkova, getting from 9-all to 20-all was roughly equivalent to flipping a coin eleven times and getting heads every time–a one-in-two-thousand shot.

To reach 20-all, then, players need to get to 9-all, then trade points another eleven times. For servers at 55%, that’s a one-in-ten shot followed by a one-in-two-thousand shot, or one in twenty thousand–a 0.005% likelihood–altogether.

Here are the equivalent numbers for servers at various levels:

SPW   Reach 9-all  Reach 20-all  that's 1 in…  
55%         10.0%        0.005%         18357  
60%         10.3%        0.008%         12916  
65%         11.2%        0.014%          7086  
70%         12.5%        0.031%          3201  
80%         17.0%        0.244%           409

You might remember the 24-22 tiebreak that Reilly Opelka won against John Isner in Dallas a couple of years ago. The probabilities are dramatically different depending on how serve-dominant the players are, so the Rybakina-Blinkova result was considerably more far-fetched than what Opelka and Isner produced. Adjusting for the fact that the Dallas tiebreak was first-to-seven and assuming that both players won 80% of serve points (an estimate on the low side), this method gives us a one-in-2,192 chance of that tiebreak reaching 22-all.

There are various ways to tweak the numbers. It might be the case that players perform better facing match point; if so, it’s a bit more likely that they’d reach this sort of outrageous score. Maybe it’s appropriate to give Rybakina a modest edge over Blinkova; if we did that, the odds of drawing even so long would be lower. One-in-18,357 isn’t exactly right, but it gives us a rough idea of just how unusual yesterday’s feat truly was.

* * *

January 19, 1974: Sanctioned

Four months from its proposed opening day, things finally started to look up for World Team Tennis. On January 18th, the USLTA officially sanctioned the league in exchange for a $144,000 fee. Another chip fell the next day, when American co-number one Jimmy Connors signed with the Baltimore Banners.

WTT still had several hurdles to clear. The British LTA continued to object to the league’s attempted takeover of so many weeks of the summer calendar. The ILTF, as well, had yet to give their okay. The ATP, still a nascent players’ union, also held back. A few top men–John Newcombe, Ken Rosewall, and now Connors–had thrown in their lot with the upstarts, but until the union made its stance clearer, the WTT ranks remained dominated by women stars.

Across the country, those women were making the case that they’d be able to draw sufficient crowds on their own. Also on January 19th, the first event of the 1974 Virginia Slims circuit came to a close. 6,000 fans packed San Francisco’s Civic Auditorium to watch Chris Evert take on Billie Jean King for the title. Another 2,000 were turned away at the gates. Traffic was jammed for blocks in every direction, and ticket scalpers worked the rows of stalled motorists.

The Slims tour had been dominated by Margaret Court in 1973, with Billie Jean hampered by injury and Evert competing on a separate tour sponsored by the USLTA. This year, Court was absent, pregnant with her second child. If San Francisco was any indication, the Australian would hardly be missed. The federation had made peace with the one-time rebels of the Slims tour, and now Gladys Heldman’s women-only circuit was the only game in town. Billie Jean was healthy (and the ultimate marquee draw, after defeating Bobby Riggs), and Evert provided new blood.

Chrissie also provided fresh motivation for the Old Lady. King had hinted that she would dial back her tournament commitments in 1974, but she wasn’t one to back down from a challenge. Playing no-ad games for the San Francisco title, Billie Jean kept her younger opponent under constant pressure. Five times Evert reached sudden-death point on her serve; five times she saved it. King finally pulled ahead to take the first set, winning the tiebreak, 5-2. Evert mounted a comeback from 0-4 in the second, but Billie Jean halted her momentum when she chased down a drop shot that Chris didn’t think she could touch.

“She was very gutsy and I once thought I had no chance,” King said after the match. “And thank God for giving me a pretty good pair of wheels on that particular shot.”

Billie Jean was thrilled at both the result and the sellout crowd. Nothing pleased her more than a successful women’s tour–except, of course, for a successful women’s tour with herself on top.

* * *

Subscribe to the blog to receive each new post by email:

 

Jiri Lehecka’s Excess of Self-Restraint

Also today: January 18, 1924

Jiri Lehecka at the 2023 US Open. Credit: Hameltion

It’s been a wild 2024 so far for Jiri Lehecka. He took a set from Novak Djokovic at the United Cup, beat Jack Draper for his first career ATP title in Adelaide, and then, defending quarter-finalist points at the Australian Open, lost today in the second round to 91st-ranked Alex Michelsen.

Even before the roller-coaster January, it was clear that the Czech was someone to watch. Ranked 23rd on the ATP computer, he’s the fifth-best player on tour under 23. He scored two top-ten victories last year–over Andrey Rublev and Felix Auger-Aliassime–and outlasted Tommy Paul in a gripping third-round five-setter at Wimbledon. For a moment it seemed that Czech men’s tennis had fallen into an uncharacteristic lull; with Lehecka, Tomas Machac, and 18-year-old Jakub Mensik on the rise, the country’s fortunes are headed back in the right direction.

Lehecka’s signature skill is raw power. A feature on the ATP website last February highlighted his average forehand speed of 79.2 miles per hour, a rate that compares to the likes of Rublev, Auger-Aliassime, and Jannik Sinner. He’s so strong that he propels those rockets without even looking like he’s trying. Rublev signals that a big swing is coming with an emphatic grunt; upon ignition, Lehecka demeanor is barely distinguishable from the pre-match warmup.

Yet the eye-popping power hasn’t shown up on the statsheet. According to my forehand potency metric, FHP, Lehecka ranks near the bottom of ATP regulars. His FHP is only 1.4 per match, right behind Diego Schwartzman. Rublev’s FHP per match is ten times higher, at 14.7. Same shot–at least according to the radar gun–but very different results. Converting FHP to points won, Rublev’s forehand earns eight or nine points each match that Lehecka’s forehand does not.

The Czech’s groundstroke winners are some of the prettiest on tour: compact strokes resulting in lasers that opponents can only watch from afar. He can turn on a second serve as well as anyone. But more often, he plays like someone without those natural gifts. One of his favorite shots is the groundstroke from the middle of his court back up the middle, deep. That choice is never a liability, exactly: opponents can rarely respond with an aggressive shot of their own, due in part to Lehecka’s natural power. But it never generates winners, and it doesn’t appear to have positive follow-on effects, either. According to Match Charting Project data, after hitting a down-the-middle forehand, he wins points 47% of the time, roughly in line with tour average.

It isn’t just the forehand. Few ATPers hit so many balls down the middle. The following table shows most of the players ahead of him in the rankings, along with the rates at which they hit groundstrokes in general down the middle (All DTM), and how often they hit forehands down the middle (FH DTM):

Player              All DTM  FH DTM  
Alex de Minaur        35.8%   28.8%  
Jiri Lehecka          34.2%   27.9%  
Holger Rune           33.0%   26.9%  
Jannik Sinner         29.7%   25.7%  
Alexander Zverev      29.4%   28.8%  
Ugo Humbert           29.3%   27.2%  
Cameron Norrie        29.2%   22.8%  
Taylor Fritz          28.7%   26.8%  
Grigor Dimitrov       28.3%   20.5%  
Nicolas Jarry         27.8%   22.5% 

Player              All DTM  FH DTM  
Daniil Medvedev       27.5%   27.8%  
Karen Khachanov       27.0%   22.0%  
Adrian Mannarino      26.8%   25.0%  
Frances Tiafoe        26.7%   21.9%  
Stefanos Tsitsipas    26.4%   22.3%  
Novak Djokovic        26.0%   21.1%  
Carlos Alcaraz        26.0%   22.6%  
Tommy Paul            25.8%   20.3%  
Casper Ruud           25.5%   21.1%  
Andrey Rublev         24.3%   18.1%  
Hubert Hurkacz        21.0%   16.6%

Only de Minaur goes up the middle more often, and he is a very different kind of player. While fellow basher Sinner is near the top of the list, even he is five percentage points less likely than Lehecka to take the conservative route. Rublev earns his baseline success by going to the other extreme. The forehand-specific numbers tell a similar story, except that Zverev and Medvedev join Lehecka and de Minaur near the top.

In theory, a crush-it-deep-down-the-middle strategy could work, but there’s little evidence that it does. The typical tour player wins 46% of the points when they hit a forehand down the middle, versus 56% when they hit a forehand elsewhere. True, the direction of every shot isn’t entirely in their control: some of those down-the-middle forehands are recovery shots. But many more are in the hands of the player who hits them. Lehecka’s power should generate, on average, weaker replies, meaning that his flexibility to choose his next shot is greater than that of his peers.

Against Draper in the Adelaide final, the Czech took a few more chances. Only 30% of his groundstrokes went down the middle, and an awful lot of those were very deep. He won 54%–an unusually high rate–of points in which he hit a forehand or backhand down the middle. He also didn’t miss, committing just one unforced error in that direction for the entire match. Lehecka, similar to the tour as a whole, usually hits unforced errors on about one-tenth of their shots down the middle.

Those numbers sound unsustainable, and today’s match against Michelsen suggests that they were. The young American kept the pressure up, and Lehecka responded by reverting to form. 42% of his groundstrokes went down the middle, he missed one in ten of them, and all told, he won just 45% of those points. Trade in those numbers for his results from the Adelaide final, and the Michelsen match becomes a dead heat.

The Czech, in short, seems to be squandering his raw power. His ace rate is slightly below tour average, his first-serve win percentage even more so. There’s no guarantee that directing more groundstrokes–especially forehands–to the corners would be a net improvement, but the Rublev’s example indicates that there are immense potential gains in that direction.

It isn’t easy to achieve the proper balance between point-winning aggression and not-point-losing passivity. Lehecka has many more years to figure it out. Until he does, we can continue to marvel at the blistering forehands of a player outside the top 20.

* * *

January 18, 1924: In or Out?

One hundred years ago this week, the governing bodies of tennis were busy determining who wasn’t allowed to compete.

Regional associations in the United States were mulling a proposed USLTA rule that would revoke the amateur status of players who earned money writing about the sport. This was much more than a formality: Bill Tilden and Vinnie Richards, two of the strongest men in the game, were among those who earned their livings as journalists. Tennis was only slowly adapting to marquee names who didn’t come from money: Richards had once been suspended for working too closely with a sporting goods company, and Tilden rarely saw eye-to-eye with the men who ruled the federation.

On January 15th, the California LTA endorsed the regulation. The West Coasters tended to be a little less stodgy than the more tradition-oriented East Coast bodies, so the announcement did not bode well for Tilden’s and Richards’s chances of continuing in the amateur ranks. Tilden was ready to call the bluff: The 1925 squad for the all-important Davis Cup would look awfully fragile if the moonlighting journalists weren’t on it.

Another, more concrete decision, came down on the 18th. Molla Mallory, the Norwegian-born American star and seven-time US champion, was ruled ineligible for the Paris Olympics that summer. Tennis was still part of the Games, though 1924 would be its last appearance for decades. The USLTA had asked the International Olympic Committee for clarification: Would Mallory, would had represented Norway in 1912, be able to suit up for her adopted country?

The answer that arrived was negative–and it was worse than that. She couldn’t play for the US, because of her earlier appearances for Norway. But since she was now an American citizen, due to her 1919 marriage to businessman Franklin Mallory, she couldn’t play for Norway either!

The second flap was soon forgotten. Two weeks later, a clarification came from the IOC that Mallory was eligible to represent Norway, as she had been born there. She competed for her native country, losing in the quarter-finals to 18-year-old American sensation Helen Wills. Her chances in the doubles didn’t amount to much, since the rest of the Norwegian team was unknown abroad. With Jack Nielsen, she won a round in the mixed before falling in straight sets to the eventual silver medalists, Richards and Marion Zinderstein.

Richards had to suspend his journalistic activities to compete in Paris, since the IOC already had a policy preventing athletes from getting paid for writing about the Games. He didn’t regret it, winning gold medals in both singles and doubles. Tilden, though, honored his writing contracts and skipped the event. Besides, he said, Davis Cup was more important. He’d rather save energy for that.

Tilden would win the staredown with the USLTA, and famous tennis names would feature as newspaper bylines for years to come. Within a decade, full-time newspapermen would joke that their jobs were in danger from all the competition. In reality, those same anonymous journalists were writing the words that went under the better-known bylines. Only a few star athletes, including Tilden, cranked out their own copy.

Ghostwriting, then, was one of the early ways for “amateur” standouts to cash in on their celebrity. And in part, it was the reason that 1924 was the sport’s last full appearance at the Olympics for six decades. The IOC feared that tennis, for all its pretense to the contrary, had become too professional. As such, it didn’t belong in the Games.

Decades later, players would seize control of their own fates, even earning the right to compete in the Olympics as professionals. By then, the issues pitting athletes against federations would be different, but the movement could trace its roots to Bill Tilden and his insistence that he be allowed to write about tennis for money.

* * *

Subscribe to the blog to receive each new post by email:

 

Can Lorenzo Sonego Hunt Down Enough Forehands?

Lorenzo Sonego at Monte Carlo in 2022. Credit: si.robi

This year, Australian Open broadcasts threw a screwball into their traditional post-match statsheet. In the addition to the usual numbers–winners, unforced errors, break points won, and so forth–the graphic shows something called “Hunting 3rd Shot Forehand.” I must have missed a memo. This is the first I’ve heard of such a thing.

A puff piece for the Tennis Australia data group offers something of a definition. The new stat measures “the times the server forehands their first post-serve hit, indicating their desire to dictate the point.” Um, ok. In other words, when the service return comes back, how often does the server hit a forehand with his next shot?

The intention behind the metric is straightforward. You hear a lot these days about the “plus-one”–the server’s second shot. While the serve is the most important stroke in tennis, the plus-one shot is the next-most crucial opportunity to attack. Both because it arises often, and because it offers a chance to define the direction of the rally, even if it’s not yet possible to put the ball away.

It is easier to dictate play with a forehand than a backhand; the potential trajectories of the stroke give a player more options. Beyond that, most men have better forehands than backhands. (The stat appears on broadcasts for both men’s and women’s matches, but today I’m going to talk about the men’s game.) If the goal is to command the rally with the plus-one shot, it’s better to hit a forehand than a backhand. A higher “Hunting 3rd Shot Forehand” number, then, is better.

The post-match graphic, with new stat second from the bottom

Before we go further: I simply can’t use this name. It’s long and confusing. (Is the player hunting for the forehand? For a winner? For a silly rabbit?) I’m going to call it “3rd Shot FH%” or “3F%” for short.

(And yes, I promise to get to Sonego eventually.)

The stat is not as straightforward as the intention behind it. The implication of 3F%, I think, is something like, “How hard did the player try to hit plus-one forehands?” A possible further implication is, “How well did the returner prevent his opponent from hitting plus-one forehands?” The second question prompts yet another: “How well did the server keep the returner from sending balls to his backhand?”

It may be possible to separate some of those questions, but there’s a lot more spadework to do before we get there.

What is normal?

(You might doubt whether I am well-situated to answer. Still, we soldier on.)

Your TV screen shows you some “Hunting 3rd Shot Forehand” numbers. Are they good?

The 3F% metric can be calculated from Match Charting Project data, so we have thousands of data points to draw upon. Based on men’s matches since 2014, the average 3F% is 64.7%. The middle third of player-matches falls between 59.3% and 70.9%. Take a little liberty with rounding, and we can say that “normal” is the range from 60% to 70%. Less than 60%, and you’re doing something wrong–or you’d rather hit your backhand, or your opponent had a day. More than 70%, and you were really getting things done in the plus-one department.

Some players consistently land at the far ends of the distribution. Here are career numbers for the top ten active players by this metric, along with 15 more names of interest:

Player                3F%  
Lorenzo Sonego      78.6%  
Rafael Nadal        77.7%  
Joao Sousa          77.6%  
Denis Shapovalov    77.1%  
Albert Ramos        76.0%  
Jeremy Chardy       75.7%  
Milos Raonic        74.6%  
Stefanos Tsitsipas  74.4%  
Casper Ruud         73.9%  
Grigor Dimitrov     73.9%  
* * * * * * * * * * * * *                    
Holger Rune         69.7%  
Dominic Thiem       69.3%  
Hubert Hurkacz      67.5%  
Carlos Alcaraz      67.2%  
Andrey Rublev       64.7%  
Jannik Sinner       62.6%  
Alex de Minaur      61.2%  
Stan Wawrinka       60.9%  
Andy Murray         60.0%  
Taylor Fritz        58.3%  
Diego Schwartzman   56.9%  
Novak Djokovic      56.2%  
Frances Tiafoe      55.8%  
Alexander Zverev    51.1%  
Daniil Medvedev     50.0%

There’s Lorenzo!

The top of the list gives you an idea of what sorts of game styles result in lots of plus-one forehands. Big serves help. Left-handedness works in your favor, perhaps since everyone trains so hard to return to a right-hander’s backhand side. Some clay-courters do well, as they are less likely to think of the serve as a point-ending shot on its own, focusing instead on how it can set up the point.

It also helps to try to hit plus-one forehands. Neither Zverev nor Medvedev seem to think in those terms, so their low 3F% ratings don’t reflect any lack of execution.

Does this even matter?

Some valuable on-screen real estate–and an enormous amount of coaching time–would be wasted if 3F% didn’t correlate with points won. Fortunately for the conventional wisdom, it does: A plus-one forehand is more likely to lead to a point for the server than a plus-one backhand is.

57.5% of plus-one forehands eventually turn into a point won, compared to 50.9% of plus-one backhands. That’s a ratio of 1.13, a number that will be more useful as a reference point in a moment.

The value of a plus-one forehand depends on the player. Matteo Berrettini wins 58.5% of plus-one forehand points but only 44.6% of plus-one backhands. That’s a ratio of 1.31, one of the highest of any active player. For him, 3F% certainly matters: All else equal, more plus-one forehand points leads to better results overall.

(A word of caution, though: The marginal plus-one forehand point–that is, the next return that he might have ran around to hit a forehand, but didn’t–might not have improved his results for the better. Presumably Matteo knows his own capabilities, and he hits forehands only on those points where they improve his odds of winning. The marginal plus-one forehand, for a player like him, is a fairly desperate foray into the doubles alley.)

For others, the plus-one choice barely registers. Zverev wins 56.7% of plus-one forehands and 52.8% of backhands, a ratio of 1.07. Every other top-tenner has a wider split, but there are more extreme examples. Adrian Mannarino wins more points behind his plus-one backhand than forehand, 55.4% to 52.3%.

Sonego, our 3F% champion, gains nearly as much from his plus-one forehands as Berrettini does. He wins 57.1% of plus-one forehand points, against 44.4% for backhands. It isn’t easy to find his backhand, but it’s worth the attempt.

What about first and second serves?

As far as I know, broadcasts don’t separate the “Hunting” metric into first and second serves. But they should! Early in the rally, the effect of the serve retains plenty of influence.

The following table shows some of the tour-wide averages I’ve discussed so far–3F%, plus-one forehand points won, and plus-one backhand points won–broken down into first-serve and second-serve points:

AVERAGES     3F%  FH W%  BH W%   
1st serve  71.1%  60.3%  53.9%  
2nd serve  55.2%  52.2%  48.1%
Total      64.7%  57.5%  50.9%

Bigger serves generate weaker (and less targeted) returns, which invite more forehands. Behind second serves, ATPers only manage to hit forehands on 55% of their plus-one shots. On the other hand, the gap in points won isn’t as wide.

A fascinating outlier is Andrey Rublev. He finds the forehand on over 80% of his first-serve points, one of the highest numbers on tour. Behind the second serve, though, he hits plus-one forehands only 43% of the time–one of the lowest! It’s no secret that his second serve is a liability, but such a gap still comes as a surprise.

Sonego is a more typical case: lots more plus-one forehands on the first serve than the second (83% to 71%), and a wide gap in the results between forehands and backhands regardless of which serve it supports.

Converting from the backhand corner

With a few exceptions like Mannarino, most players want to hit as many plus-one forehands as they plausibly can. If the return goes to their forehand corner, obviously they’ll hit a forehand. If the return comes back up the middle, it’s either a no-doubt forehand or an easy decision to take a couple of steps around the ball and avoid the backhand.

The real decisions happen when the return goes to the backhand corner. Now we’ve moved into true Match Charting Project territory. I don’t know if the Australian Open has the data to drill down this far; either way, it probably won’t show up on your TV screen. In this corner of the internet, though, we’ll dive in.

About one-quarter of in-play returns go to the server’s backhand corner. Ernests Gulbis set the standard for plus-one backhanding, running around just 2% of those balls. On average, players go for the forehand 26.6% of the time. Even Zverev and Medvedev go that route sometimes: 9% for the German and 8% for the Russian.

Here again, Sonego sets the standard. He runs around 49% of those returns, winning 53% of the resulting plus-one forehands versus 47% of the backhands he can’t avoid. No other active player creates so many forehand opportunities. Of retired players in the charting dataset, only Carlos Moya and Leonardo Mayer were more extreme.

Here’s the same tour-averages table as above, now limited to points with returns to the backhand corner:

BH CORNER    3F%  FH W%  BH W%  
1st serve  34.9%  61.3%  53.0%  
2nd serve  16.9%  52.3%  47.3% 
Total      26.6%  58.7%  50.0%

It’s possible that many players–though probably not Sonego–are leaving some points on the table here. I’m surprised to see that the gap in win percentages between plus-one forehands and backhand is bigger for backhand-corner returns than returns in general. Backhand-corner returns are somewhat similar to each other–certainly more similar than returns in general. Thus I would expect that players would find an equilibrium in which they ran around enough shots that their forehand and backhand winning percentages end up closer together. Perhaps some ATPers overestimate the quality of their backhands, or maybe they don’t want to look foolish taking a chance in the doubles alley. Or they might just know what they’re doing, and the guy typing on his laptop should shut up about it.

Hunting Alcaraz

Sonego beat Dan Evans in Melbourne yesterday, earning him a date on Thursday with second-seeded Carlos Alcaraz. While there’s more to the match than Sonego’s hunt to maximize his 3F%, the battle for the Italian’s plus-one court position will play a big part.

Alcaraz is a bit better than the typical tour player at landing his returns in the server’s backhand corner, something he does 30.8% of the time, compared to the norm of 27.0%. But it doesn’t make him particularly effective at avoiding his opponents’ plus-one forehand. They find the preferred shot 64.5% of the time, almost exactly tour average. The story is the same when we look at first and second serves separately: Carlitos neither prevents nor encourages plus-one forehands.

There are, naturally, returners who consistently limit plus-one options; others don’t have the skills to avert a barrage of forehands. Jenson Brooksby allows opponents plus-one forehands on just 57.7% of his returns; Andy Murray and (surprisingly?) Rublev keep opponents’ numbers down around 59%. At the other extreme, Cam Norrie allows servers to hit plus-one forehands almost three-quarters of the time. He’s one of many lefties who struggle by this metric: Since serve returns are disproportionately backhands themselves, left-handers must often go down the line to put a return in a right-hander’s backhand corner. Norrie finds that corner only one-fifth of the time.

Well-targeted returns are good; forcing servers to hit plus-one backhands pays dividends. Alcaraz, though, is proof that you can make your money on the fourth shot and beyond. Opponents hit plenty of plus-one forehands against him, yet no matter what they choose for the second shot, they struggle to win the point. First serves, second serves, plus-one forehands, plus-one backhands … Carlitos beats tour average by multiple percentage points in every category. This table shows the difference between how players fare against Alcaraz and the average level, in percentage points:

VS CARLOS  rel FH W%  rel BH W%    
1st serve      -6.9%      -2.5%  
2nd serve      -2.2%      -5.0% 
Total          -5.7%      -3.5%

In other words, a plus-one forehand is 5.7 percentage points less likely to turn out well against Alcaraz than it is against an average ATP player. That’s a hefty margin for something that accounts for nearly half of the typical player’s service points.

It’s fun to know that Sonego occupies the unique position that he does on tour, and it’s entertaining to see some of the far-fetched places from which he’ll smack an inside-out forehand. It might even be useful to see the Italian’s “Hunting” stat at the conclusion of tomorrow’s match.

Alas, “setting up the point” and “winning the point” are two different things. Sonego might hunt down enough forehands against Alcaraz to manage the first, but the second is a considerably bigger ask.

* * *

Subscribe to the blog to receive each new post by email:

 

The Manufactured Attack of Caroline Garcia

Caroline Garcia in 2019. Credit: Peter Menzel

Last night, Caroline Garcia scored what many fans saw as an upset, straight-setting two-time Australian Open champion Naomi Osaka. While Garcia was seeded 16th and Osaka is just beginning a comeback, no one ever knows quite what to expect when the Frenchwoman takes the court. The former champ, for her part, has always been at her best on big stages.

The result was almost pedestrian. Garcia turned in a performance that exemplified the tennis of her late 20s: Serving big, returning pugnaciously, taking risks, and–on the rare occasions that Osaka left her an opening–net rushing. Osaka served well, but the 16th seed out-aced her, 13 to 11. More than three-quarters of points were decided in three shots or less, and Garcia stole a few more of those from her opponent than Osaka did from her. In a contest defined by small margins–one break of serve and a tiebreak–that was all it took.

The strange thing is, Caro didn’t use to play like this. She plays shorter points than any other tour regular, an average of 2.9 shots per point in charted matches from the last 52 weeks. It isn’t just about her powerful first serve: Her return points end even sooner than her serve points do. Back in 2018, when she first reached her career-best ranking of 4th on the WTA computer, she was averaging over four shots per point, a rally length that would put her in the range of Jessica Pegula and Maria Sakkari: in other words, a very different sort of player.

Here is the evolution of Garcia’s rally length, shown as a rolling 10-match average, for the 84 matches in the charting dataset:

Last night’s rally length was a blink-and-you’ll-miss-it 2.5 shots, the second-lowest figure I have on record for Garcia. Only a match against Donna Vekic last year comes in slightly lower, though last week’s match in Adelaide against Jelena Ostapenko may have been even more extreme. Osaka’s big game helped keep the number down, but it takes two to so comprehensively avoid the long-rally tango.

Garcia’s first serve has always been a weapon. But her tactical approach behind it has fluctuated wildly. The career trend of her Aggression Score in rallies illustrates how she has careened from one extreme to another. Aggression Score is scaled so that the most passive players rate around -100 and the most aggressive around 100, though Ostapenko and others have pushed the maximum figures further into triple digits. Here is how Garcia’s score has changed over time, again as rolling ten-match averages:

I don’t think there any other player in tennis–man or woman, past or present–who has followed a path like this. As she established herself as an elite on tour, even as she rose into the top five, she became more and more conservative. For reference, players who posted scores around zero in 2023 were Sakkari and Martina Trevisan, hardly styles that will remind you of Garcia’s. Eventually she reversed course, not only regaining her former style but surpassing it, ranking among Liudmila Samsonova and Aryna Sabalenka as one of the most aggressive players on tour, a rung below the class-of-her-own Ostapenko.

Is it working?

The oddest thing about the multiple phases of Garcia’s career is that she has reached the No. 4 ranking with two different styles. In each of her first three charted matches after achieving the peak ranking in 2018, she posted negative rally aggression scores. In two matches against Sabalenka, she averaged 3.9 and 3.7 shots per point; against Karolina Pliskova in the Tianjin final, the typical point lasted 4.3 strokes. When she returned to the No. 4 ranking at the end of 2022, after years in the wilderness, she was frequently posting triple-digit aggression scores and average rally lengths below 3.

The main effect of Garcia’s current style is that it makes the most of her serve. From 2015 to 2017, she won just over 66% of her first-serve points, a mark that is good but sub-elite. She fell all the way to 62% in 2021 before the big shift; since then, she has won more than 70% of her first-serve points. She ranked fourth in that stat heading into the Australian Open, and she converted nearly 90% of her first serves against Osaka. Her success behind the second serve hasn’t shown the same improvement, but the overall picture is a good one: She won more total serve points in 2023 than ever before.

The return game is a different story. This is where even a casual viewer can’t miss Caro’s new tactics: She’s not afraid to stand well inside the baseline to return serve, and yesterday she net-rushed one Osaka serve, SABR-style. Measured by court position, if not by winners and error stats, Garcia is even more aggressive than Ostapenko.

At her best, the Frenchwoman posted acceptable return numbers, if not great ones. Her best single-season mark, winning 42.7% of her return points in 2017, put her in the bottom third of top-50 players. As she has upped the intensity of her attack, this key number has headed south:

In the last 52 weeks, she has won just 38.3% of return points, worst among the top 50 by two full percentage points. Among the top 20, no one else is below 42%. She can get away with it because her own serve is so rarely broken, but such ineffectual return results will make it difficult to mount another assault on the top five. Breaking serve so rarely dooms her to a career of three-setters and narrow decisions. Those sorts of results can sometimes be encouraging–as in her pair of recent three-set losses to Iga Swiatek–but have a knack for halting winning streaks, too.

It doesn’t have to be this way. Players don’t sign contracts agreeing to deploy the same tactics on both sides of the ball. Garcia won return games far more often in her less aggressive days, breaking 33% of the time in 2017 compared to a dreadful 23% last year.

Some of Caro’s 2017 skills are still in evidence. She is solid enough in long rallies that she doesn’t need to so actively avoid them: In the last year, she has won a respectable 48% of points that lasted seven or more strokes, and if you remove the two Swiatek matches, she breaks even. While the Osaka match was primarily determined by short points, Garcia won 17 of 29 (59%) that went to a fourth shot.

Without any major changes, Garcia will remain the sort of player who aggravates fans and opponents alike, a dangerous lurker capable of delivering upsets, inexplicable marathons, and lame early exits in equal measure. Like any hyper-aggressive player, Caro’s results can be seemingly random, with all the frustration that entails. Unlike Ostapenko, Sabalenka, and the many ball-bashers on tour, though, Garcia has chosen to play this way, rebuilding her game into something that the 2018 version of herself would hardly recognize. If she can somehow join her late-career serve to her earlier return-game tactics, the randomness will disappear, and Caro may make yet another appearance in the top five.

* *

Subscribe to the blog to receive each new post by email:

 

The Improbable Rise of Emma Navarro

Also today: New stat leaderboards

Emma Navarro at the 2023 US Open. Credit: Hameltion

When Emma Navarro beat Elise Mertens for her first WTA title in Hobart on Saturday, it was only part of a natural progression. For more than a year now, she has shown a knack for winning, regardless of level, surface, or just about anything else. While most fans still don’t know her name, she’s up to 26th in the official rankings and 22nd on the Elo list.

The former collegiate champion–winner of the national title as a Virginia Cavalier in 2021–started her 2023 campaign just inside the top 150. She arrived at the brink of the top 100 with back-to-back ITF titles on clay in April, then cracked the top 60 with a grass-court final in Ilkley. Her first top-ten win came in September on hard courts, against Maria Sakkari in San Diego, and after a busy fall that included another two ITF titles, she broke into the top 40. She’s 8-1 so far in 2024; the only blip is a loss to Coco Gauff.

Altogether, that’s 72 victories since the beginning of last year. Not many women can boast so much success at the W25 level or higher in that span:

Player                   2023-24 Wins  
Arina Rodionova                    79  
Iga Swiatek                        73  
Emma Navarro                       72  
Oceane Dodin                       64  
Jessica Pegula                     62  
Julia Riera                        59  
Aryna Sabalenka                    59  
Martina Capurro Taborda            59  
Yafan Wang                         58  
Carlota Martinez Cirez             57

The remarkable part of Navarro’s rise is not the sheer quantity of positive results; it’s that she rose through the rankings so fast at the age she did. She first cracked the top 100 last May just before her 22nd birthday–hardly old by any rational standards, but nearly geriatric on the youth-driven WTA tour. The 25 players standing in front of Navarro in this week’s rankings broke into the top 100, on average, before their 20th birthday: The median is Aryna Sabalenka’s arrival at 19 years, 5 months. Late developers like Jessica Pegula, Barbora Krejcikova, and Navarro are exceptions to a long-standing rule.

It’s not unusual for a player to finally achieve a double-digit ranking when they are 21 or older, but it’s rare for a future star to do so–and now that Navarro is a tour-level title-holder ensconced in the top 30, she deserves that label. Since 1990, there have been 207 players who finished their age-21 season ranked between 101 and 200 without a previous appearance in the top 100. Only 25 of them reached #100 at the end of the following year; Navarro was only the fourth to crack the top 50.

Of those 200-plus players, only 35 of them ever achieved a top-40 ranking. (A few more, including Katie Boulter and Katie Volynets, could still join the group.) On average, it took them 1437 days–just short of four years–to do so. Navarro needed only 315 days, the second-fastest in the last 30-plus years. Here are the players who made the fastest move from the end of their age-21 season to the top 40:

Player                 Age 21  top 40 debut  Days  
Elise Mertens            2016    2017-08-28   245  
Emma Navarro             2022    2023-11-06   315  
Veronika Kudermetova     2018    2019-11-11   315  
Kurumi Nara              2012    2014-06-09   525  
Jamie Hampton            2011    2013-06-24   546  
Casey Dellacqua          2006    2008-07-28   581  
Tathiana Garbin          1998    2000-09-25   637  
Liudmila Samsonova       2019    2021-11-01   672  
Bethanie Mattek Sands    2006    2008-11-03   679  
Anne Kremer              1996    1999-04-12   833  
Jil Teichmann            2018    2021-04-26   847  
Zi Yan                   2005    2008-05-05   861  
Paula Badosa             2018    2021-05-24   875  
Yone Kamio               1992    1995-06-12   896  
Alison Riske Amritraj    2011    2014-06-09   896  
Johanna Konta            2012    2016-02-01  1127

It’s possible that Navarro could have been ready for the big time earlier had she not spent two years playing college tennis. Her sub-100 ranking at the end of 2022 was partly due to a limited schedule, as she played only a handful of tournaments before leaving school after the spring semester that year. But she wasn’t playing top-100 tennis when she did step on court: Elo ratings respond much more quickly to quality results (and do not reward quantity for its own sake), and her ranking by that algorithm, 148th, was virtually identical to her place on the official list.

Whatever the benefits and (temporary) costs of her stay at the University of Virginia, Navarro seemed to learn from the step up in competition–and quickly. She lost her first 11 matches against the top 50; in the last four months, she has won 5 of 6.

What works

The most memorable victory so far was Saturday’s triumph over Mertens for a debut WTA title. It was a grind, taking two hours, 50 minutes, and spanning 14 breaks of serve en route to a 6-1, 4-6, 7-5 finish. There was little first-strike tennis on display, as the average point ran to 5.5 strokes. 69 points required seven shots or more, and 37 reached double digits.

The battle for openings worked to Navarro’s advantage. In a sample of eleven previous matches logged by the Match Charting Project, she struggled in longer rallies, winning just 46% of points that reached a seventh shot compared to 49% overall. On Saturday, she reversed that trend in a big way, out-point-constructing her veteran opponent and winning a whopping 59% of the longer points. Of 84 charted Mertens matches, it was only the eighth time that she played at least 20 long points and won so few of them. Among the few players to beat her so soundly on rally tactics: Pegula and Simona Halep.

While Navarro’s results have steadily improved, her game plan is still recognizable form her days as a college champion. After defeating Miami’s Estrela Perez-Somarriba for the 2021 NCAA title, she described her approach: “I was able to dictate with my forehand and finish a lot of points with my backhand.” In Hobart, her backhand continued to populate the highlight reel, with seven clean down-the-line winners. But it was the forehand that opened the court in the first place.

She played, essentially, a clay-court match, using the forehand to create opportunities for the next ball. She hit winners with 7% of her forehand groundstrokes, slightly below tour average. But when she was able to hit a forehand, she won the point 62% of the time, an outstanding figure for a close match. One point serves as an illustration of the rest: At 2-all, 15-all in the third set, Navarro converted a return point with a down-the-line backhand winner on the 14th shot of the rally. After a deep forehand return, Navarro was forced to hit two backhands. When she was finally able to deploy the forehand on the 8th shot, she stabilized the point by going down the middle. The 10th shot took advantage of a let cord with a heavy crosscourt forehand, a weapon that worked in her favor on Saturday more than two-thirds of the time. Her next forehand went the other direction, creating the space for–finally–a backhand out of the Belgian’s reach.

While not every point was quite so tactical, point construction always lurked. Mertens frequently attempted a pattern where she would go the same direction with two consecutive groundstrokes then, having wrong-footed Navarro with the second of them, go for a winner. The sequence doesn’t work against a big swinger because the points don’t last long enough. That wasn’t a problem against the American, but Navarro’s resourcefulness nullified the tactic nonetheless. Unlike many players her age, Navarro is able to use slices off both wings to neutralize points, and she often did so on the second shot of Mertens’s would-be pattern. The Hobart champion hit 40 slices over the course of the match, ultimately winning the point on 20 of them. For a defensive shot, rescuing 50% of those situations counts as a victory.

There is little in Navarro’s game that advertises her as a world-beater: The weapons I’ve described work best as part of a carefully-managed package. She may prove to be most dangerous on clay, where aggressive opponents will have a harder time keeping points short. She might also develop yet another level. Twelve months ago, only a reckless forecaster would have predicted she could rise so high, so quickly. We still haven’t seen her peak.

* * *

Deep leaderboards

Among the cult favorites on the Tennis Abstract site are the tour leaderboard pages, which contain nearly 60 sortable stats for the top 50 players on each circuit. Many of those stats aren’t available anywhere else, including things like average opponent ranking and time per match. It’s also possible to filter the matches for each calculation to determine things like the best hold percentages on clay.

Last week I introduced three new pages that extend the same concept:

Here’s just one example of what’s possible, the best WTA players outside the top 50 by ace percentage:

These are a great way to identify standout skills of lesser-known players. All of the leaderboards update every Monday.

* * *

Subscribe to the blog to receive each new post by email: