The first serve clips the tape and drops long. It is 30-40, late in the set, and there is one ball left. The arm slows down a fraction, the ball floats into the middle of the box at a comfortable height, and the returner, who has been waiting for exactly that ball, steps in and hits it hard.
It is tempting to call that serve a choke. The match data tell a slightly different and more useful story. Under pressure, the second serve mostly shows how careful a player becomes.
What a missed first serve costs
The gap between the two serves is large, and it has been measured many times. In 508 men’s and 508 women’s matches at Wimbledon from 1992 to 1995, Klaassen and Magnus found that 59.5% of first serves went in for men and 61.6% for women, against 86.4% of second serves. When the first serve landed, men won 74.0% of the points and women 63.1%. When the second serve landed, those figures fell to 59.4% and 52.6%.
Hawk-Eye data show the same gap. Across 262,596 points from 1729 matches at ATP, WTA and Grand Slam events between 2003 and 2008, men won 71.0% of points on their first serve and 54.2% on their second, and women 62.9% and 49.4%. In 2024 the ATP’s own analysis put the benchmark for the men’s tour plainly: “The ATP Tour average is 51 per cent points won behind second serves.” The studies count slightly differently, for example in whether double faults are included, so the exact figures vary, but every one of them shows a clear gap.
The same ATP analysis shows that the best second serves do not all look alike. Jannik Sinner serves at the body, so he “does not have to worry about the singles line or centre line on the sides of the service box.” Carlos Alcaraz “prefers to hit a heavy kick that pulls his opponent outside the singles court.” Neither is trying to hit a second first serve.
What pressure does to the second serve
If pressure made players fall apart, double faults would rise on the biggest points. In men’s matches at the 2012 Grand Slams, Jeff Sackmann found the opposite. Compared with each server’s usual rate for the season, double faults on break point were “19.1% less than average.” His conclusion: “Servers demonstrate the ability to limit double faults, and do so as the importance of the point increases.”
That control has a price. Klaassen and Magnus concluded that players “can achieve a larger efficiency gain on their second service than on their first service, possibly because of a misguided (inefficient) fear for a double fault.” In other words, even top players’ second serves were, if anything, too safe.
A study of seven Grand Slams from 2006 and 2007 found a similar pattern across all strokes, not only the serve: “on more important points, players hit strokes that are closer to the center of the court, and clear the net by a higher margin.” Its author estimated that a player’s chance of beating an equal opponent “would increase from 0.5 up to as much as 0.75-0.80 if they could play as aggressively on the more important points as on the less important ones.” That is a model estimate, not a promise, but it shows how much caution can cost.
Do players choke?
On the wider question, the research is split. An analysis of more than 650,000 Grand Slam points from 2016 to 2019, counting double faults and unforced errors, found that “an increase in the pressure index (a 1-5 score) significantly increased the probability of a performance error”, “as did an error on the preceding point”. A study of 8,280 service games in the first sets of the 2010 Grand Slams concluded that “men consistently choke under competitive pressure, but with regard to women the results are mixed.”
Other studies found little effect of the score. A 2001 analysis of 43 Grand Slam singles matches reported that “the proportion of points won by the server was not significantly influenced by score”, and Sackmann later found that “most players win break points about as often as they do non-break points.”
The clearest pattern across all of this is caution rather than collapse. Players protect themselves against the double fault, and the second serve is where the protection is most visible.
Spin is the margin
A good second serve is not a slow first serve. Researchers who measured elite male players found average spin rates of 127.4 rad/s for the flat serve, 232.1 for the slice and 336.5 for the kick, roughly 1,200, 2,200 and 3,200 revolutions per minute. Horizontal ball speed moved the other way, from 52.0 to 46.4 to 40.8 m/s, about 187, 167 and 147 km/h. In the authors’ words, “a clear tradeoff existed between the development of ball spin rate and horizontal ball velocity.”
The professionals use that tradeoff on purpose. Hawk-Eye data show that second serves carry significantly more spin than first serves, and a notational study of 116 men’s matches at the French Open found that the topspin serve “was essentially used on the second serve.” At the top of the game, the safety margin on a second serve comes largely from spin.
That is also why “two first serves” is rarely the answer. After analysing about 1,300 ATP matches from 2015, Sackmann concluded that “there are almost no situations where it would be sensible for a player to hit two first serves.” The data point toward a bolder second serve, not a second first serve.
Practising the serve that matters
If pressure changes the second serve, it makes sense to practise it under pressure. A meta-analysis of 14 studies with 394 participants found that pressure training had a medium effect on performance (g = 0.67), and its authors concluded that “coaches and instructors should create pressurized training environments rather than relying on greater amounts of training to help performers adjust to pressure.”
Tennis has a small laboratory example of why. In a study of 23 players hitting second serves before and after an anxiety manipulation, the rise in the stress hormone cortisol went together with the drop in serving performance: “cortisol response and a drop in serving performance are positively correlated (r=.68, p<.001).” Pressure is a physical state as well as a thought, and another basket of serves on a quiet court does not recreate it.
A few changes to practice can bring some of that pressure onto the court:
- Play practice games in which the server has only one serve, so every point starts with a second serve.
- Give double faults a cost in practice sets, such as losing two points instead of one, so every second serve is hit with something at stake.
- Choose a target before every second serve. As the ATP analysis of Sinner shows, the body is a real option.
- Build margin with spin and height over the net rather than by slowing the swing.
- Keep a simple record in matches: second-serve points won and double faults, on big points and on the rest.
Almost all of the data above come from professional tennis. At club and junior level the numbers may differ, but the question is the same, and a record of your own matches will answer it: on the points that matter, does your second serve get safer, and does it cost you?
None of this means a player should stop caring about double faults. It means the second serve deserves its own practice, with its own targets, and some of that practice should feel like 30-40 late in a set. The serve a player hits with one ball left shows how they handle pressure. The data say most players handle it by becoming careful. The work is to become confident instead.
Sources
- Klaassen and Magnus (2006). Are economic agents successful optimizers? An analysis through service strategy in tennis. Tinbergen Institute Discussion Paper TI 2006-048/2. A later version appeared in the Journal of Econometrics, 148(1), 72-85 (2009). tinbergen.nl
- Rioult, Mecheri, Mantel, Kauffmann and Benguigui (2015). What can Hawk-Eye data reveal about serve performance in tennis? Workshop on Machine Learning and Data Mining for Sports Analytics (MLSA 2015), ECML/PKDD. kuleuven.be
- O’Shannessy (2024). Here’s why Alcaraz & Sinner are second-serve standouts. ATP Tour, 7 August 2024. atptour.com
- Sackmann (2013). Avoiding double faults when it matters. Tennis Abstract, 24 April 2013. tennisabstract.com
- Sackmann (2015). Are two first serves ever better than one? Tennis Abstract, 2 November 2015. tennisabstract.com
- Sackmann (2016). First and second serves: another ATP info-miss. Tennis Abstract. tennisabstract.com
- Paserman (2023). Gender differences in performance in competitive environments? Evidence from professional tennis players. Journal of Economic Behavior and Organization, 212, 590-609. doi.org/10.1016/j.jebo.2023.05.042
- Harris, Vine, Eysenck and Wilson (2021). Psychological pressure and compounded errors during elite-level tennis. Psychology of Sport and Exercise, 56, 101987. doi.org/10.1016/j.psychsport.2021.101987
- Cohen-Zada, Krumer, Rosenboim and Shapir (2017). Choking under pressure and gender: evidence from professional tennis. Journal of Economic Psychology, 61, 176-190. doi.org/10.1016/j.joep.2017.04.005
- O’Donoghue (2001). The most important points in Grand Slam singles tennis. Research Quarterly for Exercise and Sport, 72(2), 125-131. PubMed
- Sakurai, Reid and Elliott (2013). Ball spin in the tennis serve: spin rate and axis of rotation. Sports Biomechanics, 12(1), 23-29. doi.org/10.1080/14763141.2012.671355
- Gillet, Leroy, Thouvarecq and Stein (2009). A notational analysis of elite tennis serve and serve-return strategies on slow surface. Journal of Strength and Conditioning Research, 23(2), 532-539. PubMed
- Low, Sandercock, Freeman, Winter, Butt and Maynard (2021). Pressure training for performance domains: a meta-analysis. Sport, Exercise, and Performance Psychology, 10(1), 149-163. doi.org/10.1037/spy0000202
- Lautenbach, Laborde, Achtzehn and Raab (2014). Preliminary evidence of salivary cortisol predicting performance in a controlled setting. Psychoneuroendocrinology, 42, 218-224. PubMed
