> PITCHf/x is a video-based tracking system that is permanently installed in every MLB stadium and has been used since the start of the 2007 season to track every pitch in every MLB game. The system consists of two 60 Hz cameras mounted high above the playing field, with fields of view that cover most of the region between the pitching rubber and home plate
> Depending on the specifics of each installation, the pitch is typically tracked in the approximate range y=5-50 ft, resulting in about 20 images per camera for each pitch
So the raw data is from two 60 Hz cameras build some time before 2007, each taking 20 pictures of a small, fast moving object tens/hundreds of feet away? The error bars on that raw data must be significant, especially when a "knuckleball wobble amplitude" of just a few inches would already be very effective.
If I'd design that PITCHf/x system today, I'd add a handful of 80 GHz radar modules, and cameras on two more axis.