The server bounces the ball three times, looks across the net and starts the motion. On the other side, the returner stands a step behind the baseline with their weight forward. The ball goes up. For a moment, before the racket has touched anything, the returner already has a guess about where the serve is going.
That guess has to come from somewhere, and professional returners are clear about where they look first. In a study that interviewed eight current or former top-250 professionals, one player put it simply: “I suppose I would look mainly at the toss. The action generally from player to player isn’t going to change that much. The toss obviously changes, so you’re trying to get little cues out of the toss.”
The toss does carry information. But it is one clue among several, it varies less than many players assume, and the research has a few surprises about how reading it can be learned.
Why returners read anything at all
The return has to start early because there is very little time after contact. A 2014 study, summarising earlier work, put the time the ball takes to reach the returner at “somewhere between 0.5-1.2 s, depending on serve quality (first and second serve), initial velocity and court surface.” Figures it quotes from an earlier study give average times on clay of 913 milliseconds for first serves and 1158 for second serves, and on a fast carpet 720 and 868 milliseconds.
In less than a second, the returner has to judge direction, spin and height, move and swing. Skilled returners therefore use whatever they can see before contact to prepare. The same 2014 study found that players returning balls from a machine, with no server and no toss to watch, timed their movement differently from when they faced a real server. Reading the server is part of the return.
What skilled returners watch
Laboratory studies support what the professionals say, with one important addition: the toss is never the only cue. When 13 skilled, 13 recreational and 11 novice players judged serve direction from video, “information from the ball toss and the arm + racquet region underpinned players’ anticipation skill”. An earlier study of how players search the server’s body for information found that during the routine before the serve, “experts focus on the shoulder/trunk areas whereas novices concentrate their search around the head of the server.” Once the motion begins, “experts concentrate on the racquet whereas novices use more cues.”
The expert picture is broad: shoulders and trunk early, then the toss, the arm and the racket. A returner who watches only the ball in the air may be ignoring much of what the server shows.
What the toss gives away
If every toss were identical, reading it would be pointless, and professional tosses are not identical. In a study of 15 professional men with ATP rankings between 180 and 1431, the kick serve’s toss peaked 18 to 21 cm to the side of the other serve types, and its contact point 30 to 35 cm away, from both sides of the court. Second serves to the T and out wide also differed, by 11 cm at the top of the toss and 18 cm at contact, but only from the deuce court. The authors note that whether returners can actually use differences of that size has not yet been tested.
Direction on the first serve is subtler. A study of professional servers found that for first serves aimed at different targets, “The positions of the front foot, ball zenith, and ball impact were significantly different”, while on second serves the kinematics “were consistent”. The conference report from the same research group, based on six right-handed professionals serving to the deuce court in a laboratory, put it more plainly: the players hit second serves to different parts of the court “‘off the same toss’, yet require significant but subtle variation in toss kinematics to do the same with their FSs”, meaning their first serves. The differences in the first-serve toss were a matter of a few centimetres.
The two research groups do not fully agree about second serves, and both studies are small. Taken together, they suggest that even at professional level the toss can reveal the type of serve, and sometimes its direction, but the differences are small and most visible when the serve type changes.
A lean, not a commitment
The professionals interviewed in the 2018 study were careful about how much they took from the toss. They were “trying to pick up any cues possible that you can to try to get a slight lean on a serve.” One line in the study sums up what happens next: “Once the server makes contact with ball, clearly the ball flight information overrides all previous contextual and kinematic information.”
Disguise shows why the lean should stay light. In a study of groundstrokes, not serves, “The experts’ anticipatory advantage was removed by disguise at 40 ms before contact.” Novices did worse still: shown disguised shots cut off at contact, their predictions were “significantly below chance”. A returner who commits fully to an early read can be sent the wrong way. One who leans and then follows the ball loses much less when the read is wrong.
Can reading be taught?
Here the research has an uncomfortable message for coaches who like to explain. In a study with junior players, one group was told which features of the server’s action predict direction. That group “did not demonstrate any significant performance improvements after the intervention.” Another group simply practised predicting the serve from video that stopped at different points of the action. They “significantly improved their prediction accuracy”, but the training effect “dissipated after a 32 day unfilled retention interval.”
One study is not the last word, but the pattern is worth noting. Telling a player to watch the toss may do little on its own. Practising the read, often and against real serves, is more promising, and it has to continue if the gains are to last.
Using it on court
For returners and their coaches, the evidence points to a few habits:
- Watch the whole server, not only the ball: shoulders and trunk early, then the toss, arm and racket.
- Use the toss for a lean, never a commitment. The flight of the ball after contact decides.
- Return against real servers as well as ball machines, so the reading part of the return gets trained too.
- Make prediction a game in practice: the returner calls the serve type or direction as early as possible and keeps score.
- Keep it going. In the video study above, the gains faded within about a month once the training stopped.
Servers can use the same research from the other side of the net. Filming the toss from the returner’s side shows what it gives away. Professionals’ kick tosses differ from their other serves by about 20 cm, so complete disguise is not a realistic goal for most players. A toss that looks the same for the flat and slice serves, and second serves that go to different targets “off the same toss”, are worth working toward.
The toss is the first clue in a very short story. Skilled returners read it along with everything else the server shows, and then let the ball have the final word.
Sources
- Vernon, Farrow and Reid (2018). Returning serve in tennis: a qualitative examination of the interaction of anticipatory information sources used by professional tennis players. Frontiers in Psychology, 9, 895. doi.org/10.3389/fpsyg.2018.00895
- Carboch, Süss and Kocib (2014). Ball machine usage in tennis: movement initiation and swing timing while returning balls from a ball machine and from a real server. Journal of Sports Science and Medicine, 13(2), 304-308. PMC
- Jackson and Mogan (2007). Advance visual information, awareness, and anticipation skill. Journal of Motor Behavior, 39(5), 341-351. PubMed
- Goulet, Bard and Fleury (1989). Expertise differences in preparing to return a tennis serve: a visual information processing approach. Journal of Sport and Exercise Psychology, 11(4), 382-398. doi.org/10.1123/jsep.11.4.382
- Carboch, Tufano and Süss (2018). Ball toss kinematics of different service types in professional tennis players. International Journal of Performance Analysis in Sport, 18(6), 881-891. doi.org/10.1080/24748668.2018.1519750
- Reid, Whiteside and Elliott (2011). Serving to different locations: set-up, toss, and racket kinematics of the professional tennis serve. Sports Biomechanics, 10(4), 407-414. PubMed
- Reid, Whiteside and Elliott (2011). Hitting to different spots on the court: the ball kinematics of the professional tennis service. ISBS 2011 conference, Porto. Portuguese Journal of Sport Sciences, 11(Suppl. 2), 373-376. uni-konstanz.de
- Rowe, Horswill, Kronvall-Parkinson, Poulter and McKenna (2009). The effect of disguise on novice and expert tennis players’ anticipation ability. Journal of Applied Sport Psychology, 21(2), 178-185. doi.org/10.1080/10413200902785811
- Farrow and Abernethy (2002). Can anticipatory skills be learned through implicit video-based perceptual training? Journal of Sports Sciences, 20(6), 471-485. doi.org/10.1080/02640410252925143
