APBRmetrics

The discussion of the analysis of basketball through objective evidence, especially basketball statistics.
It is currently Fri Jul 20, 2018 1:01 am

All times are UTC




Post new topic Reply to topic  [ 31 posts ]  Go to page 1, 2, 3  Next
Author Message
 Post subject: BPM and VORP
PostPosted: Mon Apr 16, 2018 2:12 pm 
Offline

Joined: Sat Apr 14, 2018 2:16 pm
Posts: 1
I love BPM and VORP as an advanced statistic; it's probably the one I rely on most in determining value. It values Michael Jordan and LeBron James incredibly highly, it confirms that Kobe is overrated, and it's absolutely shitting on Carmelo this season. It also seems to be a reliable statistic for determining the MVP, and correctly names the MVP in some seasons when it was awarded incorrectly.

However, I was curious about the interaction term at the end that includes assist and rebound percentage. I was curious if that term hurt certain players who have a very low assist or rebounding percentage. The 04-05 and 05-06 seasons seem to be the best examples. While I don't think Steve Nash was the MVP either of those years, I also don't think it's reasonable to say that he was the 51st best (04-05) and 22nd best (05-06) player in the league those two years (according to VORP). Is it that specific term that causes him to be ranked so low or some other facet of his game those seasons?


Top
 Profile  
 
 Post subject: Re: BPM and VORP
PostPosted: Mon Apr 16, 2018 4:42 pm 
Offline

Joined: Thu Apr 14, 2011 11:10 pm
Posts: 5231
That factor affects a lot of extreme cases (on rebounds and / or assists) and distorts them imo. It is one of the main reasons I try to avoid using BPM. It may help on average but it hurts too.

It hurts Nash but so did his defense and the shot defense attributed to him by BPM. He might have been worse on shot defense than BPM attributed with its simple, imprecise team level method. Probably should dig up his DRAPM estimate to check it.


Top
 Profile  
 
 Post subject: Re: BPM and VORP
PostPosted: Mon Apr 16, 2018 4:46 pm 
Offline

Joined: Thu Apr 14, 2011 11:18 pm
Posts: 819
Location: Maine
jmkeele wrote:
I love BPM and VORP as an advanced statistic; it's probably the one I rely on most in determining value. It values Michael Jordan and LeBron James incredibly highly, it confirms that Kobe is overrated, and it's absolutely shitting on Carmelo this season. It also seems to be a reliable statistic for determining the MVP, and correctly names the MVP in some seasons when it was awarded incorrectly.

However, I was curious about the interaction term at the end that includes assist and rebound percentage. I was curious if that term hurt certain players who have a very low assist or rebounding percentage. The 04-05 and 05-06 seasons seem to be the best examples. While I don't think Steve Nash was the MVP either of those years, I also don't think it's reasonable to say that he was the 51st best (04-05) and 22nd best (05-06) player in the league those two years (according to VORP). Is it that specific term that causes him to be ranked so low or some other facet of his game those seasons?


As the creator of BPM and VORP, I agree with your questions and concerns.

Steve Nash is an interesting subject. His offense was spectacular, but he supplied it in ways that the box scores struggle to see. He had an extremely high assist percentage--but so have others who have supplied less value in their passing. Coupled with his lack of rebounding and other positive indicators, the box scores simply don't yield the information that shows his spectacular offensive contributions. The long-term RAPM (which includes a prior based on playing time and team quality) that I am using to investigate changes to BPM has his career offensive rating as +6.4, third only to LeBron James and Chris Paul--who still have the downside of their careers to bring that number down. Dirk, Ray Allen, and James Harden also have similar offensive ratings.

_________________
Developer of Box Plus/Minus
APBRmetrics Forum Administrator
GodismyJudgeOK.com/DStats/
Twitter.com/DSMok1


Top
 Profile  
 
 Post subject: Re: BPM and VORP
PostPosted: Mon Apr 16, 2018 5:24 pm 
Offline

Joined: Thu Apr 14, 2011 11:10 pm
Posts: 5231
What was Nash's long-run DRAPM and how did it (and / or single season DRAPM) compare to his main teammates or total team in the MVP years?

What was the correlation between the AST * REB term and ORAPM or RAPM or today's RPM?

Is a BPM 2.0 (with or without play by play defensive data) likely or unlikely?

What do you like or not with Jacob Goldstein's PIPM? Any comments about its stated superior performance (without such a "versatility" power term)?


Top
 Profile  
 
 Post subject: Re: BPM and VORP
PostPosted: Mon Apr 16, 2018 7:35 pm 
Offline

Joined: Thu Apr 14, 2011 11:18 pm
Posts: 819
Location: Maine
Crow wrote:
What was Nash's long-run DRAPM and how did it (and / or single season DRAPM) compare to his main teammates or total team in the MVP years?

What was the correlation between the AST * REB term and ORAPM or RAPM or today's RPM?

Is a BPM 2.0 (with or without play by play defensive data) likely or unlikely?

What do you like or not with Jacob Goldstein's PIPM? Any comments about its stated superior performance (without such a "versatility" power term)?


1. DRAPM was -2.3, worse than any of the other star offensive players (close to Ray Allen and Kobe Bryant)
2. High, but I'd have to do some digging.
3. Likely, but I'm not moving fast.
4. PIPM is a nice statistic. The stated superior performance was an apples/oranges comparison (Jacob didn't actually run the numbers on the same dataset to compare).

_________________
Developer of Box Plus/Minus
APBRmetrics Forum Administrator
GodismyJudgeOK.com/DStats/
Twitter.com/DSMok1


Top
 Profile  
 
 Post subject: Re: BPM and VORP
PostPosted: Mon Apr 16, 2018 7:41 pm 
Offline

Joined: Thu Apr 14, 2011 11:10 pm
Posts: 5231
Thanks for the replies. Look forward to seeing more from you as you desire / have time.


Top
 Profile  
 
 Post subject: Re: BPM and VORP
PostPosted: Wed Apr 18, 2018 12:06 am 
Offline

Joined: Thu May 12, 2011 2:09 am
Posts: 285
Steve Nash is an interesting subject. His offense was spectacular...

was john stockton's offense just as spectacular? giving nash the benefit of the doubt, and looking only at his seasons of 2000+ minutes played (11 of them, ages 26-36), compared to stockton ages 26-36:

- overall they shot similar (60.2% vs. 60.3% ScFG%)...
- nash scored 16.6 pts/g, stockton 15.2 pts/g (19.5 vs. 17.2 pts/40min)...
- nash committed more turnovers (3.8 vs. 3.6 TO/40min)...
- their rates of offensive efficiency were close (2.14 pts/0ptposs for nash, 2.06 pts/0ptposs for stockton)...
- but stockton pass for 16% more assists (13.7 vs. 11.5 ast/40min)...
- and stockton was the better offensive rebounder by 20%...

stockton in this age range was clearly equal to or better than nash offensively...

he supplied it in ways that the box scores struggle to see... box scores simply don't yield the information that shows his spectacular offensive contributions...

care to expound on this? specifically?...

and can the same be said of stockton? you have nash's offensive rating as 3rd to only lebron james and chris paul. how is he better than stockton, and why?...

and not just offensively but overall how does BPM/VORP rank nash vs. stockton?...


Top
 Profile  
 
 Post subject: Re: BPM and VORP
PostPosted: Wed Apr 18, 2018 4:46 am 
Offline

Joined: Sun Sep 10, 2017 8:20 pm
Posts: 37
Ehh, Nash's offense was more high leverage than Stockton's, very very hard to pick up on in stats.


Top
 Profile  
 
 Post subject: Re: BPM and VORP
PostPosted: Wed Apr 18, 2018 6:31 am 
Offline

Joined: Thu May 12, 2011 2:09 am
Posts: 285
Ehh, Nash's offense was more high leverage than Stockton's, very very hard to pick up on in stats.

howso? if it can't be picked up by stats, just what are you referring to?...

over the long 7 year stretch of 0405-1011 nash (ages 30-36) and the suns had the league's top rate of offensive efficiency at 112.0 pts/100poss scored...

over a similar 7 year stretch from 9192-9798 stockton (ages 29-35) and the jazz had the league's top rate of offensive efficiency at 110.3 pts/100poss scored...

similar individual stats, similar team stats over significant lengths of time and sample sizes...


Top
 Profile  
 
 Post subject: Re: BPM and VORP
PostPosted: Wed Apr 18, 2018 9:09 am 
Offline

Joined: Fri Apr 15, 2011 12:02 am
Posts: 4128
Location: Asheville, NC
The '05-11 Suns also didn't award assists as readily as most franchises. Check out Nash's home-and-away Ast/TmFG rates, and it suggests you could add 5 to 10% to his annual Assist totals.
Has anyone tried this to see how it affects his BPM, his WS/48, or his PER?

Nash moved around a bit, and he improved first the Mavs and then, more spectacularly, the Suns.
Stockton stayed put. Had he moved, he could well have been "MVP".


Top
 Profile  
 
 Post subject: Re: BPM and VORP
PostPosted: Wed Apr 18, 2018 12:55 pm 
Offline

Joined: Sun Sep 10, 2017 8:20 pm
Posts: 37
bchaikin wrote:
Ehh, Nash's offense was more high leverage than Stockton's, very very hard to pick up on in stats.

howso? if it can't be picked up by stats, just what are you referring to?...

over the long 7 year stretch of 0405-1011 nash (ages 30-36) and the suns had the league's top rate of offensive efficiency at 112.0 pts/100poss scored...

over a similar 7 year stretch from 9192-9798 stockton (ages 29-35) and the jazz had the league's top rate of offensive efficiency at 110.3 pts/100poss scored...

similar individual stats, similar team stats over significant lengths of time and sample sizes...


Referring to the way they get their stats. Stockton for the most part got his as part of the system (where he winds up deserving ~50% of the credit for an assist), while Nash more often created opportunities from nothing (deserving closer to 100% of the credit for an assist)


Top
 Profile  
 
 Post subject: Re: BPM and VORP
PostPosted: Wed Apr 18, 2018 4:26 pm 
Offline

Joined: Thu May 12, 2011 2:09 am
Posts: 285
Referring to the way they get their stats. Stockton for the most part got his as part of the system (where he winds up deserving ~50% of the credit for an assist), while Nash more often created opportunities from nothing (deserving closer to 100% of the credit for an assist)

anyone care to translate this to colloquial basketball english?...


Top
 Profile  
 
 Post subject: Re: BPM and VORP
PostPosted: Wed Apr 18, 2018 4:47 pm 
Offline

Joined: Thu Apr 14, 2011 11:18 pm
Posts: 819
Location: Maine
eminence wrote:
bchaikin wrote:
Ehh, Nash's offense was more high leverage than Stockton's, very very hard to pick up on in stats.

howso? if it can't be picked up by stats, just what are you referring to?...

over the long 7 year stretch of 0405-1011 nash (ages 30-36) and the suns had the league's top rate of offensive efficiency at 112.0 pts/100poss scored...

over a similar 7 year stretch from 9192-9798 stockton (ages 29-35) and the jazz had the league's top rate of offensive efficiency at 110.3 pts/100poss scored...

similar individual stats, similar team stats over significant lengths of time and sample sizes...


Referring to the way they get their stats. Stockton for the most part got his as part of the system (where he winds up deserving ~50% of the credit for an assist), while Nash more often created opportunities from nothing (deserving closer to 100% of the credit for an assist)



That was my point exactly. Stockton's assists were not as high value--in other words, his passes did not create the high value shots to the same level Nash did. Nash's skill created very high value shots for the players he passed to.

At the other end of the spectrum would probably be Rondo--who would gather very high assist totals in Boston where most of the made baskets were relatively low value midrange shots or shots where his efforts did not provide much positive value.

_________________
Developer of Box Plus/Minus
APBRmetrics Forum Administrator
GodismyJudgeOK.com/DStats/
Twitter.com/DSMok1


Top
 Profile  
 
 Post subject: Re: BPM and VORP
PostPosted: Wed Apr 18, 2018 5:02 pm 
Offline

Joined: Thu Apr 14, 2011 11:10 pm
Posts: 5231
With shot charts from play by play logs or player tracking video data and expected point calculations, it would be possible to take assist credits to a variable, more accurate assessment. For today's data and in the past (with pbp) in anyone gave it the time. Folks have given the time to calculate the shooter data, woukdn't be that big a leap to run the assist data, though still have to divide the surplus value by general rule or by shot location.


Top
 Profile  
 
 Post subject: Re: BPM and VORP
PostPosted: Wed Apr 18, 2018 5:38 pm 
Offline

Joined: Thu May 12, 2011 2:09 am
Posts: 285
Stockton's assists were not as high value--in other words, his passes did not create the high value shots to the same level Nash did. Nash's skill created very high value shots for the players he passed to.

a nonsensical statement unsupported by facts...

over the 7 year stretches mentioned above both players threw for 1000+ more assists than any other player in the league during those times, and both PGs' teams lead the league in eFG% over those long 7 year periods...

yet somehow one player's assists were of higher value than the other's?...

you also sidestepped the question of what was the offensive BPM/VORP of nash versus stockton...

i'm also curious as to the defensive BPM/VORP of both players, and the overall ratings - who does BPM/VORP say was the better player?...


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 31 posts ]  Go to page 1, 2, 3  Next

All times are UTC


Who is online

Users browsing this forum: No registered users and 23 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum

Search for:
Jump to:  
Powered by phpBB® Forum Software © phpBB Group