Serve Analysis
Chas Tennis, you are correct.
What kind of serve is that?
The pictures were copied from public tennisplayer.net video - A New Teaching System
http://www.tennisplayer.net/public/site_tour/teaching_system_serve/index.html
I don't know the type of serve that was hit when I see serve videos.
For your video, the camera was hand held or the frames have been cropped differently. You can see this by looking at the signs on the stands, they move in the frame. Use objects behind the racket as references since the camera pointing angle is changing or the cropping is different. Did you make your composite image by moving things around to overlay frames?
It looks as if the ball is struck and moves probably first in frame #6 and certainly by frame #7. If you look at the top edge of the racket frame, it reaches a step in the stands at the highest point. The head of the racket seems to be at that same peak height for frames 5 & 6, when the ball is struck, and it is definitely down by frame #7. I'd say that the racket face was probably not rising when it impacted the ball.
In your composite picture, after the ball is struck there are smudge marks that repeat every 6 frames as the ball rotates. It looks like two kinds of marks, a larger one and a smaller one. Those marks don't seem like any standard ball markings- was the ball marked with two different marks? If I compare ball marks from two frames, side-by-side, they appear to mostly be rotating horizontally. If the ball rotates one complete rotation in about 6 frames it is rotating about 60° between frames.
But we don't know the camera's frame rate. To get an idea of the frame rate, let's look at how a 2000 rpm ball serve would look to a 240 fps camera. For a typical serve rotation rate of 2000 rpm, the ball rotates 2000 rpm /60 sec/min or 33 revolutions per second (rps).
If 240 fps high speed video the ball would rotate 33 rps / 240 fps or 0.14 rotation per frame. In about 7 frames, the ball would rotate one revolution. That is not far off the 6 frames that we see in your composite picture of the high speed video with unknown frame rate.
I'd have to look up Cross's information on the spin rates and the angle of the spin axis for each type of serve.
I'd say the spin is mostly horizontal and it has a higher spin rate, so it's probably a
slice serve. Also, the general path of the racket is at a large angle to the ball's trajectory and I believe that also is the path of a slice serve. I can't tell what serve has been struck looking at videos and would always like some additional information to identify the serve, the bounce or curved trajectory, etc.
I believe that this is a slice serve because in the original video I can see that the spin on the ball is horizontal and the ball curves and bounces as a slice should.
https://vimeo.com/27528347
Each serve type just depends on a few milliseconds of racket string impact over a few centimeters of contact and the racket string direction and other details. With 240 fps the frame rate is not enough for observing as impact can only appear on one frame, occasionally possibly two. 420 fps is better. But >1000 fps is needed. See some replies on frame rates in the 11 to 5 Rafter thread. There is also the issue of the strings stretching and applying spin to the ball. To see this string action requires frame rates of a few thousand frames per second.
Your composite picture might be easier to view if just the frame before impact, at impact and after impact were displayed, 3 frames only. Or maybe 5 frames. ? This would show the direction of the racket strings at impact. I'd use something in the background to do any aligning.