Italian translation at settesei.it
Wimbledon is here, so it’s time for another cycle of media commentary about the demise of net play, especially the serve-and-volley. The New York Times published a piece by Joel Drucker last week that covered this familiar territory, cataloguing various reasons why the game has changed. Racket and string technology, along with tweaks to the All England Club playing surface, are rightfully on the list.
But the first reason Drucker gives is the rise of the two-handed backhand and, by extension, the threat posed by players with weapons on both sides:
In May 1999, 43 of the top 100 male players in the world hit their backhands with one hand. As of June 2019, there were 15. According to Mark Kovacs, a sports science consultant and tennis coach, “Most players used to have a weaker side, usually the backhand. And the two-handed backhand changed that completely. It doesn’t give you a spot you can hit to.”
I’m more interested in the “weaker side” argument than the fortunes of the one-handed and two-handed backhands. Many players who still use one-handers, such as Stan Wawrinka, would rightly bristle at a claim that their shots are weak. In terms of effectiveness, the contemporary one-handed shot might have more in common with a two-hander of old than the all-slice, only-defensive backhand favored by many pros in the 1970s and 1980s.
Both sides, now
The “weaker side” argument can be slightly rephrased into a research question: For contemporary players, is there a smaller gap between forehand effectiveness and backhand effectiveness than there used to be?
To answer that, we need a working definition of “effectiveness.” Long-time readers may recall a stat of mine called “potency,” as in “backhand potency” (BHP) or “forehand potency” (FHP). It’s a simple stat, using data derived from the shot-by-shot records of the Match Charting Project, calculated as follows:
BHP approximates the number of points whose outcomes were affected by the backhand: add one point for a winner or an opponent’s forced error, subtract one for an unforced error, add a half-point for a backhand that set up a winner or opponent’s error on the following shot, and subtract a half-point for a backhand that set up a winning shot from the opponent.
The same procedure applies to forehand potency and slice potency. The weights–plus one for some shots, plus a half point for others, and so on–are not precise. But the results generally jibe with intuition. Across 3,000 charted ATP matches, an average player’s results from a single match are:
- Forehand potency (FHP): +6.5
- Backhand potency (BHP): +0.8
- Slice potency (SLP): -1.3
- Backhand side potency (BSP): -0.5
The first three stats isolate single shots, while the final one combines BHP and SLP into a single “backhand side” metric. All of these exclude net shots, and since forehand slices are so rare, I’ve left those out of today’s discussion as well.
The forehand reigns
The numbers above shouldn’t come as a surprise. The average ATP player has a stronger forehand than backhand, regardless of how many hands are on the racket for the latter shot. Novak Djokovic possesses one of the best backhands in the history of sport, but the gap between his FHP and BSP numbers is greater than average: +11.3 per match for the forehand, and +2.5 for the backhand, resulting in a difference of 8.8. Even a backhand master reaps more rewards on his other side.
The Match Charting Project has at least three matches worth of data for 299 different men across several generations, spanning from Vitas Gerulaitis to Jannik Sinner. Only 30 of them–about one in ten–gain more points on their backhand than on their forehands, and for half of that minority, the difference is less than a single point. It’s a diverse group, including Pat Cash, Jimmy Connors, Guillermo Coria, Ernests Gulbis, Daniil Medvedev, and Benoit Paire. This mixed-bag minority doesn’t provide much evidence to settle the question.
Proponents of the “weaker side” argument often point to the arrival of Lleyton Hewitt as a turning point between the net-play-was-feasible era and the approach-at-your-peril era. Others might point to Andre Agassi. As it turns out, both of these figures are surprisingly average.
The Match Charting Project has extensive records on both men. Hewitt’s forehand was worth +10.0 per match, while his backhand and slice combined for +2.9. That’s a difference of 7.1, a bit greater than average, though less than Djokovic’s. Agassi’s FHP was good for +13.0 per match, compared to a BSP of +6.8. That’s a difference of 6.2, even closer to the mean than Hewitt. Ironically, that gap is almost identical to that of Pete Sampras, whose FHP of +6.3 and BSP of -0.1 were equally spaced, even though his groundstrokes were considerably less effective.
Comparing eras
We can’t answer a general question about trends over time simply by calculating shot potencies for individual players, no matter how pivotal. Instead, we need to look at the whole population.
First, a quick note about our data: The Match Charting Project is extremely heavily weighted toward current players. Our sample of 300 players consists of only 40 whose careers were mostly or entirely in the 20th century, and 30 more whose matches mostly took place in the first decade of this century. Thus, the averages mentioned above are skewed toward the 2010s. That said, the 70 “older” players in the sample are the most prominent–the guys who played in major finals and semi-finals, and Masters finals. If there has been a marked trend across decades, those players should help us reveal it.
The earlier players in our sample are, in fact, quite similar to the contemporary ones. I ranked the 299 players by the absolute difference between their FHP and their BSP, with the most balanced player ranked 1, and the least balanced ranked 299. I looked at two subgroups: the 52 oldest players in the sample, most of whose careers were fading out when Hewitt arrived; and the 78 players with the most recent matches in the sample.
- Oldest — Average rank: 143, Average (FHP – BSP): 5.7
- Most recent — Average rank: 155, Average (FHP – BSP): 6.5
These numbers do not indicate that players used to have a weak side, and now they don’t. They don’t really reflect any trend at all. The difference between forehand effectiveness and backhand side effectiveness has barely changed over several decades.
As further evidence, here is a selection of players who are both well-represented in the Match Charting Project data and noteworthy representatives of their eras. They’re listed in approximate chronological order. Each of the shot-potency numbers is given on a per-match basis, and the final column (“Diff”) is the difference between FHP and BSP–the gap between each player’s forehand and backhand sides.
Player FHP BHP SLP BSP Diff
Bjorn Borg 12.9 11.5 -0.5 11.0 2.0
Jimmy Connors 6.5 9.1 -0.3 8.9 -2.4
John McEnroe 2.0 -0.4 -2.1 -2.4 4.4
Mats Wilander 7.2 6.8 -0.5 6.3 0.9
Ivan Lendl 10.3 4.0 0.6 4.6 5.7
Stefan Edberg 1.9 1.8 -1.1 0.7 1.1
Boris Becker 5.9 2.1 -1.5 0.7 5.2
Jim Courier 13.3 4.2 -0.3 3.9 9.4
Michael Stich 2.0 2.0 -3.4 -1.4 3.4
Michael Chang 9.7 5.0 -0.6 4.4 5.3
Player FHP BHP SLP BSP Diff
Thomas Muster 18.4 2.2 -1.1 1.1 17.3
Pete Sampras 6.3 0.7 -0.7 -0.1 6.4
Andre Agassi 13.0 7.2 -0.5 6.8 6.3
Patrick Rafter 3.5 0.5 -1.6 -1.1 4.6
Carlos Moya 9.8 -0.9 -1.4 -2.3 12.1
Lleyton Hewitt 10.0 3.5 -0.6 2.9 7.1
Guillermo Coria 4.7 6.3 -1.2 5.2 -0.5
David Nalbandian 8.8 5.6 -1.7 3.9 4.9
Nikolay Davydenko 7.2 4.4 -1.2 3.2 4.0
Roger Federer 10.0 0.2 -0.4 -0.3 10.2
Player FHP BHP SLP BSP Diff
Rafael Nadal 15.3 2.6 -1.0 1.6 13.7
Andy Murray 7.2 2.9 -1.8 1.1 6.1
Novak Djokovic 11.3 3.4 -0.8 2.5 8.8
Richard Gasquet 1.9 1.4 -1.4 0.0 1.9
Stan Wawrinka 6.2 0.5 -1.7 -1.2 7.3
Kei Nishikori 5.4 3.8 -1.1 2.7 2.8
Dominic Thiem 9.3 -0.1 -1.6 -1.7 11.0
Alexander Zverev 3.6 4.2 -1.1 3.0 0.6
Stefanos Tsitsipas 8.3 -0.9 -2.2 -3.0 11.4
Daniil Medvedev 1.6 3.3 -1.4 1.9 -0.3
Not weaker, but weak
These numbers cast a lot of doubt on the “weaker side” hypothesis, that it used to be easier to move forward by approaching an opponent’s less dangerous wing.
Instead, what has probably happened is that for the typical player, both sides got stronger. As a result, the weaker side was no longer flimsy enough to make approaching the net a profitable strategy. Even players with weaker-than-average backhands are now able to hit powerful topspin passing shots. This is essentially the racket-and-string-technology argument, and it seems to me to be the most valid.
There’s no question that tennis has drastically changed in the last few decades. But the conventional explanations for those trends don’t always hold up under scrutiny. In this case, while volleys have been reduced to a vestigial part of the singles game, groundstrokes–on both sides–have only gotten better.