The two most important shots in tennis are the serve and return, yet I would wager they are the two least practiced shots across the board; from grassroots to the pros most time on-court is spent rallying and working on the fundamentals of forehands, backhands, and volleys. Serves and returns are often ‘mixed in’—they are the necessary chores of tennis training for a lot of players, akin to stretching and fitness—yet it is clear that mastering either can take you a long way. One only has to look at the stark difference between the game styles of John Isner and Diego Schwartzman to understand this.
I think the serve and return are neglected for two main reasons:
They aren’t as ‘fun’ to practice. There is no rhythm and continuity. They must be trained like dull repetitions, and especially at a grassroots level where fun and enjoyment are paramount to maintaining squad numbers, what you practice is somewhat financially motivated.
Coaches and players confuse performance with learning when players rally. A distinction (from a prior piece on contextual interference):
It’s very easy to assume learning has occurred when you watch someone groove a stroke and hit it better and better over the course of a closed drill, but that’s not learning, that’s performance. Learning is about retention and transfer.”
Players and coaches see the rallies improve as they hit for longer and assume they are ‘getting better’ rather than just adjusting to the conditions of the day. You won’t know if you’ve learned something until some time is passed and you’re tested on it. You hit the ball flawlessly all week and then crumble in the Saturday fixture, sound familiar? You’ve performed well all week, but haven’t necessarily learned anything. This is why matches are the gold standard; they are simultaneously the lesson and the exam, and they force players to start every point with the serve and the return.
Returns
As I have touched on in prior pieces, the best exponents of forehands and backhands generally use gravity to start the racquet acceleration whilst reducing movement at the distal joints (reduced wrist action). The return is an abbreviated groundstroke, especially if standing up in the court where you have less time. The best exponent of this abbreviation is—not surprisingly—Novak Djokovic. See below his forehand return at ~0:25 seconds in. The take back is short, the follow-through is non-existent, and his strings face the target well beyond contact. As I have written about previously, improving your game is often about removing things: swing length, wrist action, head movement, etc., to gain control of the racquet face.
Nadal’s is very similar. An extended wrist and upright racquet head and short takeback make timing the ball easier because the racquet head is relatively quiet.
Players that have simple, compact strokes excel at returning, and this includes short players whose strokes are compact by virtue of their height (or lack thereof), such as Diego Schwartzman. In Brad Gilbert’s book, Winning Ugly, he explains how he used to choke up on his grip to gain control as it made the racquet shorter, and we’ve seen Nadal adapt this to his own game when slicing and volleying.
Serves
While we live in an era of great returners and baseliners, the serve is still the most important shot in the game. It’s allowed players like Isner to remain at the top of the game for decades despite limited movement and groundstrokes (Isner has been in the top 40 since 2009). In contrast to the return, the serve can often be improved by increasing swing length. This is why taller players tend to have better serves; their racquet head moves through a longer arc due to having longer levers. One player who has struggled with serving despite his height is Alexander Zverev. If we look at his serve, the main reason I think he struggles with double faulting is his super high ball toss that is more pronounced on second serves. Below are two serves from his third-round match against Brandon Nakashima at Roland Garros.
Zverev’s high ball toss means the ball is dropping as he tries to make contact, which isn’t ideal, but a bigger problem for me is that such a high toss invites a stall in the swing speed. To hit powerful serves you need racquet head speed, but to hit consistent second serves you still need huge racquet head speed; you want all that speed translated into spin. The low racquet head speed is why Zverev hits double faults.
If I had to summarize the mechanics of great servers:
They tend to move the racquet quickly through a lot of space. If we look at great servers—not just tall guys, but brilliant servers relative to their height—we tend to see the racquet head maintain or build speed through the backswing (where the racquet head moves from right-to-left behind the players head).
The one-minute video below demonstrates how Federer improved this aspect of his serve over his career.
My favourite example of this is Kyrgios, but we’ve seen similar mechanics from Karlovic, Sampras, Fritz, Federer, Djokovic, Lopez, and Roddick1. These guys swing from the hip essentially. The racquet is going to move quickly through a lot of space. This is not dissimilar to how the high 12 o’clock take-back of huge forehands uses the initial drop of gravity before the body drives and rotates the racquet. Kyrgios gets such a fast racquet by doing these two things:
He keeps the racquet tip down as he tosses.
He tosses low.
This ‘forces’ Kyrgios to start swinging from the hip; if he doesn’t the ball will be too low. It’s the most instinctive way to make a player swing fast. Feliciano Lopez is another great example.
Here is Lopez:
Let’s break it down:
0:13 - starts the weight shifting to the back foot. This will allow him to drive forward into the court, using his weight/momentum to generate pace.
0:21 - as he starts to move forward and toss the ball, his racquet is still moving back with the tip down.
0:25 - ball is released, yet the racquet tip is down and still at knee level. The racquet never slows down from here. It basically accelerates from his knee to contact. if you drew a line following his racquet tip from here to contact it would be a pretty long line!
For reference, the video below is an example of what kills a player’s serve. A stall in the backswing/behind the head. Despite being one of the greatest ball strikers off the ground, Nalbandian’s serve was a liability throughout his career.2 Note the stall when the racquet gets to the trophy position.
Going back to Zverev’s serving woes, it’s clear he is capable of serving huge. His first serve is a rocket. When the toss is low and he accelerates his 6’6’’ frame it’s a great serve. But in big moments he falls into the slow speed trap like Nalbandian. How could he ensure he keeps the same speed on the second serve?
Two ways:
Toss lower/in line with his first serve.
Delay the racquet head/toss first. By tossing with the racqet arm relaxed at his side (like Fritz, Sampras, Medvedev) Zverev won’t be allowed to slow down the racquet.
Until he makes a change, the serve will continue to haunt Zverev on big points. The reason Djokovic and Federer and Nadal remained at the top for so long was their continual evolution as players—Federer and Djokovic both added more speed in their serve backswings throughout their careers, and they served great in their 30s as a result.
People tend to think of Roddick’s serve as abbreviated, but if you look at it carefully here note how Roddick drops the racquet head low to his right side before driving his legs. The racquet passes back over his head with a lot of momentum.
Nalbandian averaged 3.7 aces/match and 3.3 double faults/match.
was thinking about the NextGen YEC physical tests (https://www.reddit.com/r/tennis/s/Ljjj3dOUBE) along with this article about the jump as a key part of the serve (https://www.tacticaltennis.com/serving-mechanics-the-jump/).
is it possible that servers like Sampras and Federer compensate for their relative lack of height (with respect to the usual servebot suspects) with greater vertical explosiveness to jump relatively higher? it seems plausible that this could even be a generalizable balancing mechanism, since we usually expect players above 198 cm to be less explosive in their lateral and forwards movement.
on a related note, do you think it's a real phenomenon that lefty servers have an inherent advantage at some level (separate from the ad-court advantage conferred by tennis's scoring system)? from what i've read, lefty servers are more commonly described as having optimal disguise (e.g. Ivanisevic, McEnroe, Arthurs). is this different from the sort of advantage that a lefty gets in that they make righties think more on court because of less familiar rally patterns and thus become more reactive than proactive?
and finally, do you know if training various sorts of jumps is done frequently in tennis-focused athletic training, compared to, e.g., balance, flexibility, and first step explosiveness drills? would that training directly help with serves, and could serving help someone passively improve their jumps?