APBRmetrics

Using tool for George with Westbrook: "When shooting, Paul George's estimated points per 100 shots is -6.59 better than the average player for his position." He is negative? His efg% is exactly league average. This is just efg% right? Didn't want to go to ts% or offensive rating? Any way ...

My brother helped me make a site!

https://zanderchase.github.io/nba-player-compatibility/

This is really impressive work! The results in the LeBron case study certainly make sense intuitively. My biggest question on a model like this, with a huge number of features, is how were the estimates were cross-validated--how was random noise was accounted for? Thanks! Not entirely sure what you...

Hi all! Wanted to share some work I've been working on, which I think is best described as attempting to model player compatibility. I had a poster at CASSIS but wanted to make it more widely available. Here is a link to a github page which has a link to both the poster and a bunch of the code used ...

Thanks for the reply. However, I am looking more for error metrics on possession basis, as opposed to on a game basis (which seems to be all that is reported in the paper, although perhaps I am missing something...)

Thanks for pointing me in the right direction tho

Doing some benchmarking of models. Wanted to validate results. Does anyone either know if error metrics associated with RAPM (r2, mse, rmse, etc) have been reported here or elsewhere, or does anyone have the time/setup to run on their own and confirm w me? Thnx

I have seen a few threads talking about scraping play-by-play with lineups, for use in calculating APM derivatives. I created a script last year for doing so from stats.nba.com, and have uploaded it to GitHub. https://github.com/hwchase17/Play-By-Play-from-NBA.com-with-Lineups This gets all PBP w/ l...

My PAA has Curry with a sizable lead, then Westbrook, Draymond, Paul George, Blake Griffin and Tim Duncan.

Team WinProjFinal
------
ATL 50
BOS 45
BKN 28
CHA 25
CHI 52
CLE 57
DAL 37
DEN 26
DET 39
GSW 59
HOU 55
IND 44
LAC 53
LAL 25
MEM 47
MIA 44
MIL 43
MIN 29
NOP 44
NYK 32
OKC 54
ORL 31
PHI 25
PHX 37
POR 30
SAC 34
SAS 56
TOR 42
UTA 44
WAS 46

I interpret the first article (carefully done) as indicating the impact of post season and vet teammates is small. Agree? The second article ranks but does not reveal how much estimated difference separates top from middle and low for the impact of coaches and teams. The impacts are at the end are ...

I wrote this, looking at the affect of postseason experience and veterans on a players progression in his early years: http://harvardsportsanalysis.org/2015/05/player-progression-in-the-nba/ Also a follow up looking at which coaches have been associated with the most player progression: http://harva...

How part time is this? (weekly hour commitment?)

There's been a lot of studies showing that hot hand may in fact exist... not that shocking.

http://www.sloansportsconference.com/wp ... proach.pdf

Kathoro wrote: You might be dealing with one of the highest living IQs on the planet here. My IQ is so high that I can make arguments without quoting statistics and they will probably be correct.

Argument won.

Have you thought about using PPP on a team basis?