Search found 294 matches

by xkonk
Wed Jul 08, 2015 9:56 pm
Forum: APBRmetrics
Topic: Worst a Player Can Be
Replies: 27
Views: 17436

Re: Worst a Player Can Be

If you think it's a gamma, can't you pull some random samples and compare them to actual NBA ratings to see what looks close? Then you'll have the distribution and will know the asymptote.
by xkonk
Wed Jul 08, 2015 12:05 am
Forum: APBRmetrics
Topic: Worst a Player Can Be
Replies: 27
Views: 17436

Re: Worst a Player Can Be

Are we assuming hypothetically, or 'in real life'? In real life, if I got on the court in an NBA game, I could only muck up the offense so much because my teammates would never pass me the ball. Hypothetically, if they treated me like I could actually play, I imagine I could actually perform worse.
by xkonk
Fri Jun 12, 2015 1:11 am
Forum: APBRmetrics
Topic: 538: Will Cavs' Defense Persist?
Replies: 6
Views: 4804

Re: 538: Will Cavs' Defense Persist?

Maybe related to Mike's calculation difficulties, but a question I have anyway... why is Neil doing so much weighting by leverage in his 538 articles? I understand in some circumstances it might make sense, like if you want to say something about clutch/playoff performances. But why weight games by ...
by xkonk
Sat Jun 06, 2015 5:55 pm
Forum: APBRmetrics
Topic: Hot Hand
Replies: 20
Views: 17562

Re: Hot Hand

And if I'm right that some players have a sweet spot they shoot well from, that could help explain the pattern too -- but only if they don't have an offsetting "cold spot". A sweet spot is an interesting idea. The ones that always leap to mind for me probably fall into the category of what you desc...
by xkonk
Sat Jun 06, 2015 2:18 am
Forum: APBRmetrics
Topic: Hot Hand
Replies: 20
Views: 17562

Re: Hot Hand

The Vox article has a mention at the end of a lack of a 'cold hand'. It's hard for me to think of a mechanism that would (convincingly) create a hot hand but not a cold hand. Any ideas?
by xkonk
Wed May 27, 2015 12:06 am
Forum: APBRmetrics
Topic: New stat - NBA Passer Rating
Replies: 6
Views: 5070

Re: New stat - NBA Passer Rating

I don't know that it would matter too much, but you do some mixing of per possession and per game. For example, in the first part of step 2 you have Pts created by assists equal to 18.2 (pts per game) - 1.056 (points per possession) * 7.8 (assists per game). Shouldn't that second part have some kind...
by xkonk
Wed Mar 11, 2015 11:39 pm
Forum: APBRmetrics
Topic: skill curve numbers are wrong
Replies: 19
Views: 11975

Re: skill curve numbers are wrong

It's a little more directly interpretable in your first regression. In the last chunk of regulation (coded 0 by your description), there's about an 7.97% chance of Duncan being mentioned if the game is tied (i.e., that's just the intercept). If the Spurs were up 10, there's still a 7.91ish %chance. ...
by xkonk
Tue Mar 10, 2015 10:04 pm
Forum: APBRmetrics
Topic: skill curve numbers are wrong
Replies: 19
Views: 11975

Re: skill curve numbers are wrong

What's the DV in that regression? Are those coefficients "big"?
by xkonk
Sun Mar 01, 2015 6:14 pm
Forum: APBRmetrics
Topic: SSAC research paper posters posted
Replies: 17
Views: 6138

Re: SSAC research paper posters posted

Crow wrote:Anyone here do or know who did the Move or Die Ball Movement paper? No link provided yet at Sloan.
Looks like another Goldsberry joint: http://fivethirtyeight.com/datalab/what ... onference/ (Sunday 12:17 AM post)
by xkonk
Sat Feb 28, 2015 4:49 pm
Forum: APBRmetrics
Topic: Other statistical techniques applicable to NBA analysis?
Replies: 10
Views: 4513

Re: Other statistical techniques applicable to NBA analysis?

Zero inflated models are pretty much just what the name says. Poisson and binomial would be examples; if you had a process that you would expect to be Poisson but think there might be more 0s than expected due to some additional factor, you would model it with a zero inflated Poisson (ZIP) instead o...
by xkonk
Wed Feb 25, 2015 3:58 am
Forum: APBRmetrics
Topic: skill curve numbers are wrong
Replies: 19
Views: 11975

Re: skill curve numbers are wrong

It might help to figure out if you've made a mistake or missed something if you posted an example calculation or two.
by xkonk
Mon Feb 23, 2015 11:47 pm
Forum: APBRmetrics
Topic: Four Factors Importance Using Win Probability Added
Replies: 13
Views: 9346

Re: Four Factors Importance Using Win Probability Added

One thing I've never seen regarding WPA-like metrics is whether it is actually more predictive than non-WPA versions. IOW, how much does it actually matter whether a player hits a shot or gets a rebound when the game is "on the line" vs. the same team being up by 20 points with 10 minutes left? Has...
by xkonk
Thu Feb 12, 2015 11:38 pm
Forum: APBRmetrics
Topic: Predictive test using lineups (Updated with results)
Replies: 12
Views: 6477

Re: Predictive test using lineups (RAPM, WP, WS, BPM, etc.)

As others have made clear in previous threads on predictions, the tests should be completely out of sample, which (hopefully) includes the sample on which the metric was calculated. This applies primarily to BPM and any RAPM with a prior, since they'll carry information over from one season to anoth...
by xkonk
Wed Feb 11, 2015 12:46 pm
Forum: APBRmetrics
Topic: Proof of Diminishing returns on Rebounds, USG-EFF, and more
Replies: 28
Views: 21288

Re: Proof of Diminishing returns on Rebounds, USG-EFF, and m

Performance between three groups isn't necessarily linearly transitive. It is likely but not required. Age 25-29 lineups are +4.6 against 20-24 players and -4.42 vs Age 30+ lineups. Age 30+ lineups are +4.3 vs Age 20-24 lineups This can't be right, right? If age 25-29 is +4.6 against 20-24, and age...
by xkonk
Wed Feb 11, 2015 12:08 am
Forum: APBRmetrics
Topic: Proof of Diminishing returns on Rebounds, USG-EFF, and more
Replies: 28
Views: 21288

Re: Proof of Diminishing returns on Rebounds, USG-EFF, and m

Age 25-29 lineups are +4.6 against 20-24 players and -4.42 vs Age 30+ lineups. Age 30+ lineups are +4.3 vs Age 20-24 lineups This can't be right, right? If age 25-29 is +4.6 against 20-24, and age 30+ is +4.3 against 20-24, then 25-29 and 30+ should be about the same. How is age 25-29 -4.42 against...