BPM 2.0

Moderator: Doctor MJ

User avatar
Dr Positivity
RealGM
Posts: 59,845
And1: 15,538
Joined: Apr 29, 2009
       

BPM 2.0 

Post#1 » by Dr Positivity » Fri Feb 28, 2020 6:45 pm

After Westbrook broke the old BPM, Daniel updated it which is reflected on basketball reference. (Russ is now still league leading in 2017, but a more human 11.1)

Some major changes

- Stockton numbers are now very strong with many seasons in the 8-9 range
- Nash looks more like a superstar than previous versions with 4.7, 5 and 5.9 from 05-07
- Durant peaks at 10.2 in 2014
- Rodman’s BPM is now pretty mediocre peaking at 2.8 earlier, but only having 0, 0.5 and -0.2 from 96-98.
- Iverson and Rose weirdly seem to have benefitted (Iverson is now at 6.1 in 2001, Rose at 6.8 in 2011)

https://www.sports-reference.com/blog/2020/02/introducing-bpm-2-0/
Arman_tanzarian
Veteran
Posts: 2,577
And1: 2,712
Joined: Dec 27, 2012
     

Re: BPM 2.0 

Post#2 » by Arman_tanzarian » Fri Feb 28, 2020 9:45 pm

It looks weak to me. This is overvaluing steals for whatever reason and undervaluing players who do both Score and Create well. That's what these metrics should be rewarding as we are seeing teams led by these players flirt with the all-time ORtg record annually these days.
Image
User avatar
Dr Positivity
RealGM
Posts: 59,845
And1: 15,538
Joined: Apr 29, 2009
       

Re: BPM 2.0 

Post#3 » by Dr Positivity » Fri Feb 28, 2020 11:26 pm

Arman_tanzarian wrote:It looks weak to me. This is overvaluing steals for whatever reason and undervaluing players who do both Score and Create well. That's what these metrics should be rewarding as we are seeing teams led by these players flirt with the all-time ORtg record annually these days.


A lot of skilled creators like Nash, Stockton, Paul, Durant did well with the update, it's players like Ben Wallace and Rodman who suffer
dhsilv2
RealGM
Posts: 42,799
And1: 22,534
Joined: Oct 04, 2015

Re: BPM 2.0 

Post#4 » by dhsilv2 » Tue Mar 3, 2020 7:36 pm

Basically looks to have nerfed defense along the lines we're seeing with RPM of recent years. Basically, historical RAPM had at least a enough top tier defensive seasons where you saw some balance, more recent seasons seem to have less of those (which isn't that crazy given how much spacing has limited the value of rim protection, by no means is it not still critical). But without seeing the timeline they used to do the regression and the final formula...I'm struggling with what to think of it. The old math explained why the westbrook thing happened without much difficulty in doing an analysis. Without that, it's hard to judge this.

Return to Statistical Analysis