How is average RPM and shot speed measured?

[socallefty]

When you play on a Playsight court or have a racquet sensor that detects and measures speed and topspin of shots, I am assuming that it measures the average topspin and speed of FH/BH etc. on all your shots as it has no way of knowing if the shot landed in or not. However, is this the correct way to measure average topspin and do pro tournaments measure it similarly when they flash the stats of the pros? Also, is there a way for the device/software to know which are topspin shots and which are slice shots and calculate average topspin only on topspin drives and under spin average only on slice shots?

A 3.5 hacker could miss half of his groundstrokes hitting too hard and if the software just measured all his shots to calculate average RPM, it would show a high shot speed. However, as an opponent who need to play only the shots that land in, it is likely his slower shots that land in and so, the average speed of shots in play might be much less. The same with average serve speed of those with flat serves that land in at a 20% rate in matches.

So, should average speed/spin be measured only for shots that land in the court and exclude errors? Do pro stats reflect this although it is less of an issue for them as they make few errors and serve a high %?

 

[Deleted member 776614]

It's a good question about which shots to count, or which shots are counted in the pros. I only know about how RPM is measured from doing my own little bit of research (and being an engineer and having some experience with golf.)

Basically the two ways to get RPM are to measure it with radar, or calculate it using ball velocity, trajectory, and distance. At tournaments they've typically used multiple view angles to accurately calculate (not measure) RPM, but I don't know if lately they've started using radar or not (something like Trackman, which I heard is starting to be used in tennis.)

To calculate RPM, you use basic physics and take the ball trajectory and speed, apply gravity and drag, and predict where it should land. The difference between that and where the ball actually landed is due to spin, and the spin is calculated using aerodynamic properties of a tennis ball. This is what most/all of the apps are doing.

 

[curly_2350]

I’d assume they are calculating off trajectory as Joe days, but that does the beg the question of what the margin of error is and how variables like wind and ball condition could impact the result.