Tennis Serve Speed: Pace Is a Measurement, Not a Rating

Tennis serve speed measures how fast the ball travels, not how good the serve is. A useful comparison needs three labels: first or second serve, maximum or average speed, and the match or tournament being measured.

For scale, 120 mph is approximately 193 km/h; 100 mph is approximately 161 km/h. Those numbers describe pace. They do not tell you whether the ball landed in, stretched the returner or helped the server win the point.

What does a radar reading measure?

Sports radar uses the Doppler effect: it measures the frequency change in radio waves reflected from a moving object. In tennis mode, the Stalker Pro II measures both the ball’s peak serve speed and its declining speed during flight. Those are different measurements, even for the same serve. The manufacturer’s manual explains the distinction; it should not be assumed that every tracking system reports the same measurement.

The ball does not maintain its launch speed all the way to the receiver. Aerodynamic drag acts against its motion, while spin influences its trajectory. Consequently, dividing the distance travelled along the ball’s flight path by the time from racquet contact to bounce gives an average over that flight—not its initial or peak speed. Using only the horizontal court distance produces a different measure again. Tennis Warehouse University’s free-flight research describes these forces and measures changes in ball speed downrange.

Radar alignment also matters. An uncompensated radar measures a lower speed when the ball travels at an angle to its aim. Stalker’s manual gives a clear example: a ball actually travelling at 100 mph registers 96.6 mph at a 15-degree angle. That is a measurement effect, not evidence that the player hit more softly.

Keep maximum, first-serve average and second-serve average separate

A maximum is one delivery. An average describes a sample, whose inclusion rules need to be clear: does it contain faults, only legal serves, first serves, second serves or both?

The distinction can be substantial. In Jeff Sackmann’s analysis of the 2024 US Open, Arthur Fils averaged 115 mph on first serves and 92 mph on seconds. Jannik Sinner averaged 91 mph on seconds. These are event-specific observations, not universal ATP benchmarks. Sackmann’s analysis of Fils’s second serve uses the comparison to show that raw second-serve pace alone did not explain Fils’s difficulties.

For broader professional comparisons, see how fast a professional tennis serve is. For the separate question of single-delivery extremes, use the serve-speed record comparison—not a typical-speed benchmark.

A faster maximum does not establish better serving

The Australian Open’s published statistics for its four men’s semifinalists provide a useful example. These figures cover their 2024 tournament runs through the quarterfinals, as reported on 26 January, rather than the completed event.

Player	Fastest serve	First-serve points won	Unreturned serves
Novak Djokovic	208 km/h (129.2 mph)	79%	36%
Daniil Medvedev	211 km/h (131.1 mph)	75%	31%
Jannik Sinner	210 km/h (130.5 mph)	77%	32%
Alexander Zverev	218 km/h (135.5 mph)	77%	33%

Source: Australian Open serving analysis. Mph values are converted from the published km/h figures and rounded to one decimal place. Unreturned-serve percentages are reproduced as published; the article does not specify their denominator.

Zverev had the fastest maximum; Djokovic led these players in first-serve points won and the published unreturned-serve percentage. That does not prove slower serving is better, or identify placement as the cause: they faced different opponents, and first-serve points won includes points settled after the return.

It does show why a speed leaderboard cannot substitute for an effectiveness leaderboard. Unreturned serves describe points ending without a successful return, while points won also reflects what happens next. For comparisons across providers, check the definitions and denominators, as explained in first-serve percentage versus unreturned rate.

Does extra speed help?

There is no dependable conversion from additional mph to additional wins.

Sackmann’s analysis of Serena Williams’s first serves across 23 Wimbledon matches in 2014, 2015, 2018 and 2019 found an association of roughly one additional percentage point in first-serve points won for every extra two mph. Yet the relationship varied markedly among the other women studied. Placement, spin and tactical choices were not fully isolated. The original analysis is evidence of a player-specific relationship, not a guaranteed training payoff.

Spin also changes what the returner encounters. A kick serve’s bounce depends on speed, its angle into the court and topspin; raw pace alone does not determine bounce height. Rod Cross’s kick-serve analysis explains that interaction.

Comparing your own serve

For club or college players, a repeatable measurement is more useful than an isolated comparison with a professional maximum:

• Keep the device, settings and radar position consistent, following its instructions.
• Record first and second serves separately, and identify serve direction.
• Keep faults in a separate count; report the average speed of legal serves alongside the percentage landing in.
• State the number of attempts and the session date, rather than reporting only the fastest ball.
• In match play, pair pace with unreturned serves, points won and double faults; record the surface and opponent too.

If your legal-serve average rises but fewer serves land in, you have demonstrated more pace—not yet a more effective serve.