A different look at shooting percentages

Home for all your discussion of basketball statistical analysis.
Post Reply
cemitten
Posts: 16
Joined: Wed Oct 01, 2014 12:02 am

A different look at shooting percentages

Post by cemitten »

The post about a simple formula for floor spacing got me thinking about different ways to approach the idea. My first thought was looking at the variance across the shooting zones provided on basketball reference. (0-3ft, 3-10ft, 10-16, 16 - <3pt, 3pt) My thought process fell apart but turned into looking at looking at if players choose their most effective spots to shoot. For each zone i multiplied a players percentage of shots from that zone by their shooting percentage subtracted from the league average from that zone. I included freethrows as well as factors for number of ft's 2pt and 3pt shots taken.

The results below are an average of 12-13 13-14 and current season to date.
Top 25 >100 minutes per season on average

Code: Select all

Kevin Durant	2.283906393
Al Horford	2.213471067
LeBron James	1.793313504
Dirk Nowitzki	1.705065678
Brook Lopez	1.640762341
Kyle Korver	1.541182333
Hassan Whiteside	1.4733116
Dwyane Wade	1.388999911
Chris Paul	1.386115633
Stephen Curry	1.385574178
Al Jefferson	1.326321022
Serge Ibaka	1.299102326
Chris Bosh	1.245757807
Marc Gasol	1.240908696
David West	1.233605637
Blake Griffin	1.113849289
Anthony Davis	1.093200985
LaMarcus Aldridge	1.086980126
Nikola Vucevic	1.013114611
Brandan Wright	1.002167392
Tyrus Thomas	0.9833482
Amir Johnson	0.955913067
Tony Parker	0.76422023
Jonas Valanciunas	0.762637319
Marcin Gortat	0.742751063
Bottom 25 >100 minutes/season on average

Code: Select all

Jusuf Nurkic	-0.8688797
Elijah Millsap	-0.8916892
Reggie Evans	-0.892757474
Victor Oladipo	-0.899674033
Dion Waiters	-0.901122421
Dominique Jones	-0.914759467
Marcus Camby	-0.922458267
Andris Biedrins	-0.93820295
Lance Thomas	-0.966729422
Omer Asik	-0.973455215
Gal Mekel	-0.9838718
Joakim Noah	-1.039745374
Larry Sanders	-1.06441723
K.J. McDaniels	-1.084219825
Ramon Sessions	-1.185830016
Xavier Henry	-1.235189881
Tyreke Evans	-1.246001067
Andrei Kirilenko	-1.288529252
Derrick Rose	-1.301139225
Andrew Bynum	-1.3542185
Elfrid Payton	-1.354463633
Ricky Rubio	-1.357005244
Tony Wroten	-1.506507044
Tim Frazier	-1.7251705
Joffrey Lauvergne	-1.8710256
The mean rating was -0.099, the median was -0.117 and the standard deviation was 0.499.

Just found an issue with my previous results. I had a duplicated record for Dennis Schroder and Dennis Schröder one one part. This caused the calculations to be messed up for every player who's first name would come after Dennis. Updated results are displayed now. Egg firmly on face :oops:
Last edited by cemitten on Wed Mar 04, 2015 2:39 pm, edited 1 time in total.
Crow
Posts: 10533
Joined: Thu Apr 14, 2011 11:10 pm

Re: A different look at shooting percentages

Post by Crow »

This is interesting. The spread is huge. All pros but very very different on this product.

Would you be willing to provide the data for more players? The all-stars from this year? Dion Waiters? Enes Kanter? Anybody else you thought was surprisingly high or low?
cemitten
Posts: 16
Joined: Wed Oct 01, 2014 12:02 am

Re: A different look at shooting percentages

Post by cemitten »

I would be happy to share the whole list. I just didn't know of a good way to display the information. I graphed the results vs things such as TS% PER and OBPM. Some interesting looking shapes but nothing very revealing. I'm at work and don't have access to image sharing sites or google drive so i can not do a better job making everything available.

Ranks and Rating and TS% for this years all star participants.

Code: Select all

Carmelo Anthony	0.408189467	55	0.550666667
LeBron James	1.793313504	3	0.622333333
Pau Gasol	0.615552785	34	0.527666667
John Wall	-0.19284663	324	0.524
Kyle Lowry	-0.153681389	290	0.544333333
Chris Bosh	1.245757807	13	0.579
Jimmy Butler	-0.204946404	331	0.559333333
Al Horford	2.213471067	2	0.572666667
Kyrie Irving	0.2872672	77	0.554333333
Kyle Korver	1.541182333	6	0.666
Paul Millsap	0.033457644	190	0.554
Jeff Teague	-0.072876989	247	0.552
Dwyane Wade	1.388999911	8	0.562666667
Blake Griffin	1.113849289	16	0.568333333
Marc Gasol	1.240908696	14	0.549666667
Kobe Bryant	0.2328184	100	0.517333333
Anthony Davis	1.093200985	17	0.581333333
Stephen Curry	1.385574178	10	0.609333333
LaMarcus Aldridge	1.086980126	18	0.519
DeMarcus Cousins	-0.186418704	318	0.541
Tim Duncan	0.706314193	28	0.543333333
Kevin Durant	2.283906393	1	0.638333333
James Harden	0.50285793	44	0.609
Damian Lillard	-0.096982463	258	0.56
Dirk Nowitzki	1.705065678	4	0.575666667
Chris Paul	1.386115633	9	0.585666667
Klay Thompson	0.582686426	38	0.558666667
Russell Westbrook	-0.543262133	473	0.537333333
Crow
Posts: 10533
Joined: Thu Apr 14, 2011 11:10 pm

Re: A different look at shooting percentages

Post by Crow »

Thanks. Minor spreadsheet glitches are easy to creep in. You caught it, so back to good.

TS% is the important value, but this product gives an indication of where a player is on skill curve. The guys with the biggest negatives are high on usage, perhaps too high for what they are doing with it compared to others on similar shots (Wall, Butler, Cousins, Westbrook, etc.)
Crow
Posts: 10533
Joined: Thu Apr 14, 2011 11:10 pm

Re: A different look at shooting percentages

Post by Crow »

Should Atlanta try to get Horford's usage up to 25-29% in playoffs? Never above 22.6% then but by these numbers I would think he could use more well. Just let what happens happen? That might not be good enough to get to Finals. But need to look at all players, tape, defensive strategy, etc.

Should Wall, Westbrook, others with negative products cut back or be forced to cut back? Key questions to consider.
cemitten
Posts: 16
Joined: Wed Oct 01, 2014 12:02 am

Re: A different look at shooting percentages

Post by cemitten »

This really murders deandre jordan for his ft shooting. If he could shoot league average he would go from ranked 180 to 13th. That feels pretty extreme. Like ft shooting shouldn't have that much affect. It doesn't kill his TS% that badly.
bchaikin
Posts: 307
Joined: Thu May 12, 2011 2:09 am

Re: A different look at shooting percentages

Post by bchaikin »

This really murders deandre jordan for his ft shooting.

go one step further and look at offensive efficiency (combine shooting with turnovers) and it's even worse - he's just a bit better than average for a C despite the very high 2pt FG% because of so many missed FTs and turnovers...

since 2010-11 jordan (12.3 pts/40min) shot 67% on 2s, but just 43% on FTs (734 missed FTs) with 10% turnovers per touch, such that his rate of offensive efficiency of 2.11 pts/0ptposs (points scored per zero point team possession personally responsible for) was just a bit better than league average for a C (2.01 pts/0ptposs)...

for comparison tyson chandler since 2010-11 (13.3 pts/40min) shot 65% on 2s but 70% on FTs with just 9% turnovers per touch, and had offensive efficiency of 2.71 pts/0ptposs - big difference compared to jordan...
cemitten
Posts: 16
Joined: Wed Oct 01, 2014 12:02 am

Re: A different look at shooting percentages

Post by cemitten »

http://goo.gl/vrT2od

Google Doc with 2.5 year aggregate.
Crow
Posts: 10533
Joined: Thu Apr 14, 2011 11:10 pm

Re: A different look at shooting percentages

Post by Crow »

Thanks.
boooeee
Posts: 88
Joined: Sun Jan 22, 2012 5:32 am
Contact:

Re: A different look at shooting percentages

Post by boooeee »

cemitten wrote:This really murders deandre jordan for his ft shooting. If he could shoot league average he would go from ranked 180 to 13th. That feels pretty extreme. Like ft shooting shouldn't have that much affect. It doesn't kill his TS% that badly.
I summarize win probability added due to free throw shooting at my site. Jordan's free throw WPA is the worst in the league at -2.29. Since it takes +0.50 win probability to win a game, that means that if Jordan were an average free throw shooter, Clippers could conceivably have five more wins. Granted, if he was average from the stripe, he would probably get fouled less.

Here's the link: http://stats.inpredictable.com/nba/ssnP ... &order=ASC
Mike G
Posts: 6144
Joined: Fri Apr 15, 2011 12:02 am
Location: Asheville, NC

Re: A different look at shooting percentages

Post by Mike G »

boooeee wrote: Granted, if he was average from the stripe, he would probably get fouled less...
But if he made 70% of his FT, he could be productive on more than the occasional ally-oop and putback. He could actually post up and learn other offensive moves. He might get a lot more FGA, and while his FTA/FGA might be less, he could also get more FTA.
cemitten
Posts: 16
Joined: Wed Oct 01, 2014 12:02 am

Re: A different look at shooting percentages

Post by cemitten »

I updated this to include last season's data. Decided I wanted to take a different look at it and figured adding to this thread would be better than starting a new one.

As a quick refresher I used the shooting data from BBR and built a metric based on % of shots from different zones on the floor and shooting % in those zones compared to league average). This also includes FTs.

I presented this data to some people and it was pointed out to me that most of the players ranked highest are expected. Great shooters will be at the top of a shooting metric, so what does this really show us? Based on this feedback I decided to compare the player rankings from my metric to a ranking of TS%.

Code: Select all

SmartShoot	SSRank	D SS TS Rank
Anthony Davis	1	-42
Nikola Vucevic	2	-140
Kevin Durant	3	-8
Stephen Curry	4	-4
Kyle Korver	5	3
LaMarcus Aldridge	6	-202
Al Horford	7	-88
Blake Griffin	8	-119
Brook Lopez	9	-98
Chris Paul	10	-26
Marc Gasol	11	-97
Dwyane Wade	12	-173
Klay Thompson	13	-31
J.J. Redick	14	-2
Dirk Nowitzki	15	-88
Pau Gasol	16	-115
Chris Bosh	17	-126
Wesley Matthews	18	-31
Hassan Whiteside	19	1
Rudy Gay	20	-94
Al Jefferson	21	-289
James Harden	22	-4
LeBron James	23	-44
Marcin Gortat	24	-24
Carmelo Anthony	25	-173
David West	26	-255
Kevin Martin	27	-109
Kawhi Leonard	28	-56
Andrea Bargnani	29	-183
Derrick Favors	30	-79
The second column shows the rank based on my metric while the third column shows the difference from the player's TS% rank. (for clarification's sake negative numbers mean TS% is however many spots lower)

(I sure wish i could figure out how Mike G makes his tables look so clean and ordered)
Post Reply