likely non-linear deceleration over the net

[petegrif]

The Advanced Tennis research team has uncovered the intriguing fact that a Sampras serve, say, loses approximately 50% of its space by the time it has crossed the net and bounced.

This raises a further interesting topic for research.

We know that this deceleration is caused by air resistance to the ball.

And we know that air resistance increases as the square of the velocity.

In other words the faster the serve as it leaves the racket the more air resistance it will encounter.

So it seems logical that the faster the serve the higher the percentage of speed it will lose. The server is working harder and nature is taking more away.

It would be intriguing to graph serves from say 60 to 120 and see what they slow down to by the time the ball bounces.

It would also be interesting (and is probably important) to understand how spin affects air resistance. If spin does indeed affect air resistance then it will affect velocity.

Just my 2c.

[maverick1]

Something related:
I used the numbers from Advanced Tennis Research to calibrate a formula for calculating serve speed from video footage(if you can count the number of frames of video it takes the serve to land).

I didn't plot any graphs, but you can use this calculator (http://mavericks.cc/tennis/serve_speed.html) to generate lots of points for the graph you mention.

This assumes that the type of spin is not a factor. I dont understand how spin affects the flight of the ball, but it doesn't seem to matter in speed calculations. A Tennis Warehouse forum member, MikeCottrill, tested this formula on a recording of a 137 mph Roddick serve in the recent Australian open; the formula agreed to within 1mph of the official radar, and that is very good considering the official radar rounds speeds to 1 mph.