APBRmetrics

What is a quick release 3 in elapsed time for catch n shoot, or from start of last dribble? How much time variance is there between "quick", "average" and "slow"? I assume some insiders know but I have never seen / heard public estimates from anyone. This seems like an omission to rectify.

Same for elapsed seconds from first forward movement for drives from behind 3 point line near top to ball in cylinder or on backboard.

People give subjective assessments but how accurate and consistent are they? Do any stylistic variances influence eye test error?

What qualifies as a quick jump to typical rebound height? First or second, offense or defense?

When do these things become quantified for the public discussion?

Michael Beuoy (@inpredict) recently published estimation of "three point windups" based on large quantities of tracking data.

The string of tweets is here:

Form is worth studying (not by me). There is no time data provided though, even after being asked in that thread.

Is the difference between fast and average, tenths of second or less?

It says 'at 1/5 speed', so I assume real time is the GIF length x5?

The interesting part for me is that Steph seems to release faster because he releases lower than other people. His chart is right next to Russell Westbrook and, at least at 1/5 speed and just using my eyeball, they both get to 7 or 7.5 feet at the same time. It's just that Steph lets the ball go there while Russell goes up another foot or so. Makes you wonder why Steph's shot isn't blocked more often.

The GIF is about 1.5 secs. It would be 1.5 / 5= about .3 second in real time. The differences are slight. Seems to be in hundreds of a second difference. I'd guess no more than .05 of a second in this sample. Variance for same player might be hundreds of a second. I question the quality of people's ability to judge between quick and average. May be right in same cases but probably hard and there may be mistakes and exaggerated statements based of processing / memory of shot speed for few shots chosen for evaluation or a large sample. Length of shot windup and elapsed time of windup may be generally correlated but there might be contra-data in some cases. Speed of movement may vary.

Next step would be to measure / see how much closer / better positioned a defender can get it in an extra .01 - .05 of a second and the value of that vs. the timing of and speed of their own movement. What are the average release times of blocked and not blocked shots? In general, for specific offensive players and defensive players? Who much can you speed up either without damaging effects to the shot or the decision to start defending the shot instead of the court? How much can you gain by better disguising the start of a shot or occupying the defender longer with court defense thinking over shot defense thinking? Plenty that could be done. Not sure how much can be done well for net improvement. But one could try. And if so, probably need real fine time assessment of video.

If you really got into this, one should measure elapsed time and movement of player on dribbles and average defender reactions. One could investigate the optimal desired and achieved net change of dribbler and defender position per dribble at every location and by strategic goal by type of dribble and associated movements.

Speed of passes, moment of windup & release, efficacy of exact location of delivery.

Probably bring in some baseball pitch / hit analysts or at least some of their work.

What has already been done in private? I dunno. I haven't seen any solid public hints of it.

Lets see if I can get any reaction.

https://twitter.com/bballstrategy/statu ... 4744838144

xkonk wrote: ↑Wed Apr 29, 2020 10:06 pm The interesting part for me is that Steph seems to release faster because he releases lower than other people. His chart is right next to Russell Westbrook and, at least at 1/5 speed and just using my eyeball, they both get to 7 or 7.5 feet at the same time. It's just that Steph lets the ball go there while Russell goes up another foot or so. Makes you wonder why Steph's shot isn't blocked more often.

Note that Curry has very little horizontal component to his motion, so he's not bringing the ball toward the defender but releases it at a high loft angle. Surely it's traveling faster in flight than in hand, so carrying the ball higher to release it does not particularly reduce the chance it gets blocked. (It might increase the chance that you get fouled.)

Chris Paul is the other player who releases well below 8'; and his motions have perhaps the greatest horizontal component of all these.
Paul had 9 of his 803 FGA blocked this year, 15 of 720 last year. Curry 40 of 1420 in that interval.
Both guys (and most of these shown) have myriad options that dissuade defenders from trying to block their treys.

Nobody responded with meaningful information on Twitter.