APBRmetrics

I did a little analysis on play by play data this morning to find if there are differences between the 1st free throw and the 2nd free throw.
The analysis was done on the 2014 season.

It turned out that shooting percentages were 4% higher league-wide on the 2nd free throw.

Also I pulled out the 5 players that had the largest improvements between 1st and 2nd free throw:

Player 1st shot % 2nd shot %
JJ Hickson 43% 61%
Nene 50% 66%
Roy Hibbert 68% 84%
Ramon Sessions 76% 90% !!
Joakim Noah 69% 80%

Posted the full thing on my blog: http://dunksallday.com/#free-throw

Interesting. Something new to me.

For the guys with the greatest differentials would you try to change that with lots of one (first) shot practice intervals? Or two shot sets? Do some players practice / depend too much on high frequency sets and developed non-game like rhythm? Too much on visual adjustment?



This is new too, I guess? https://www.reddit.com/r/bballanalytics/

There's some numerical work about free throw shooting here:

http://www.inpredictable.com/search/label/ShArc

I'm not entirely sure where he got his data.

Thats an interesting question about how to fix it... 1 shot sets could work I think, like do a short sprint, shoot a free throw, sprint, shoot a free throw.

But a lot of it too is just muscle memory , not sure how to train that other than getting reps in

None of these guys averaged more than 3 FTA/G last year. Does that matter?

Hibbert missed 32% of first shots and just 16% of 2nd shots. He's twice as likely to miss the first.
Sessions is indeed the best in this ratio (of the 5), 2.4 times as likely to miss his first.
Hickson is least volatile of the 5, in this perspective; about 46% more likely to miss #1.

If Steph Curry (say) hits 90% on first attempts and 94% on 2nd shots, that's 67% (10/6) more misses on first shots.
But his 'improvement' would be just the league average at +4% (10-6).

Who are the top players in the ratio (1-1st%)/(1-2nd%) ?
What's the league average?
I'm guessing Sessions is #1.

Instead of jumping around to miss% to try and get to a ratio, just use the odds ratio. That's what it's for. And it makes your multiplicative statements ('twice as likely') actually meaningful.

Sessions has the best odds ratio in increasing from first to second free throw; the odds are ~2.8 times as likely. But then Hibbert had the biggest increase at ~2.5, Hickson ~2.1, Nene ~1.9, Noah ~1.8. Your hypothetical Curry improvement is a bit less than Noah's; for someone to increase their odds as much as Sessions but starting from 90%, they'd have to hit a bit over 96% on their second free throws.