WildVolley
Legend
Since I am a complete layman when it comes to cameras, physics, and all those technical details, I enjoy testing the app using pro tennis tv footage. I recently used it to measure a fan-recorded video of Roddick playing at SAP (indoor event) open a few years ago. Almost every single serve Roddick hit where the contact and landing point was caught on an individual frame, I got very accurate results (+/- 1 mph from the radar speed). I used for the matchpoint of Miami from last week. Nadal struck a 121 mph serve on matchpoint, I got 123 mph.
I'm quite fond of this method to calculate speed. If you work through every step with meticulous detail, you never get numbers that are "too good to be true".
This is what we want to hear.
If we can avoid having to construct a more accurate physics model with a bunch of differential equations, that's what we want to hear. Far easier to just have a model with a bunch of simplifying assumptions that can be tossed into an excel spreadsheet.
What people need to understand at this point is that the potential error is mostly a factor of frame estimating when using low speed video (30fps) and incorrectly judging distance from ball contact to court bounce, rather than the simplifying assumptions of the physics model.