AcrossTheCourt wrote:Touch time might be important for catch-and-shoot shots (no dribbles.) Taking a shot right away in rhythm is different than holding for five seconds.
This is a good idea- I think I'll include a binary Catch and Shoot vs. Off dribble and multiply that by touch time. I would think this would capture that effect.
italia13calcio wrote:
I've played around with the data in the past and noticed that distance from defender matters a lot more near the rim. Might help with your problem with shots near the rim. I've done a bit of work with the same dataset in the past, would be happy to talk over stuff.
It's funny because one of my first findings into "shot defense data" was with the contesting shots near the rim being much lower than pressuring shots at the rim (from a Vantage article I wrote a long time ago) but with the SportVU data you can't really account for that. The big issue with modeling shots near the rim is accounting for the hand up/hand down and where the defender is positioned. When I get the Vantage shot by shot data, I might look to see if I can model shots near the rim better with their data. But as I mentioned earlier, they bin their shot defense which is problematic too I think (or maybe not).