This is Google's cache of viewtopic.php?p=2491&sid=f82190a04e8527141ebcb32d3b97007b. It is a snapshot of the page as it appeared on Mar 26, 2011 23:54:09 GMT. The current page could have changed in the meantime. Learn more

Text-only version
These search terms are highlighted: adjusted plus minus  
APBRmetrics :: View topic - Linear weights calculated from adjusted plus/minus
APBRmetrics Forum Index APBRmetrics
The statistical revolution will not be televised.
 
 FAQFAQ   SearchSearch   MemberlistMemberlist   UsergroupsUsergroups   RegisterRegister 
 ProfileProfile   Log in to check your private messagesLog in to check your private messages   Log inLog in 

Linear weights calculated from adjusted plus/minus

 
Post new topic   Reply to topic    APBRmetrics Forum Index -> General discussion
View previous topic :: View next topic  
Author Message
Eli W



Joined: 01 Feb 2005
Posts: 402

PostPosted: Tue May 17, 2005 11:57 am    Post subject: Linear weights calculated from adjusted plus/minus Reply with quote

I know that after Dan Rosenbaum developed his adjusted plus/minus statistic, he worked backwards to calculate the linear weights that correlated to adjusted plus/minus. I've been trying to find these weights in order to estimate adjusted plus/minus for this season.

I've tried to use the coefficient estimate numbers from Table 2 on this page - http://www.uncg.edu/bae/people/rosenbaum/NBA/winval2.htm , but the numbers don't seem to be coming out quite right (for instance, should I use eFGA/40 or the square of that?). Can anyone get them to come out correctly?
Back to top
View user's profile Send private message Send e-mail Visit poster's website
Dan Rosenbaum



Joined: 03 Jan 2005
Posts: 541
Location: Greensboro, North Carolina

PostPosted: Tue May 17, 2005 3:29 pm    Post subject: Re: Linear weights calculated from adjusted plus/minus Reply with quote

John Quincy wrote:
I know that after Dan Rosenbaum developed his adjusted plus/minus statistic, he worked backwards to calculate the linear weights that correlated to adjusted plus/minus. I've been trying to find these weights in order to estimate adjusted plus/minus for this season.

I've tried to use the coefficient estimate numbers from Table 2 on this page - http://www.uncg.edu/bae/people/rosenbaum/NBA/winval2.htm , but the numbers don't seem to be coming out quite right (for instance, should I use eFGA/40 or the square of that?). Can anyone get them to come out correctly?

There are weights for both eFGA/40 and eFGA/40 squared. To replicate my results, you need use both of them. Also, all of my statistics are pace-adjusted, so if you are very close to what I am getting, then that might be the difference.
Back to top
View user's profile Send private message Send e-mail Visit poster's website Yahoo Messenger
Eli W



Joined: 01 Feb 2005
Posts: 402

PostPosted: Tue May 17, 2005 4:50 pm    Post subject: Reply with quote

Thanks, that seemed to do the trick. I've managed to use the weights to calculate statistical plus/minus ratings for 2004-05. These are based on the weights in Dan's article, but they are not adjusted for pace (what exactly is the pace adjustment you use, Dan?).

Here are the top ten overall:

1. Andrei Kirilenko
2. Kevin Garnett
3. LeBron James
4. Shaquille O'Neal
5. Dirk Nowitzki
6. Vince Carter (with the Nets)
7. Tim Duncan
8. Brad Miller
9. Larry Hughes
10. Amare Stoudemire

Steve Nash ranks around 70th.

Here are the top rookies, which I found interesting:

1. Andris Biedrins
2. Andre Iguodala
3. Anderson Varejao
4. Devin Harris
5. Delonte West
6. Josh Childress
7. Josh Smith
8. Tony Allen
9. Damien Wilkins
10. Dwight Howard

Ben Gordon and Emeka Okafor rank much lower.
Back to top
View user's profile Send private message Send e-mail Visit poster's website
Ed Küpfer



Joined: 30 Dec 2004
Posts: 785
Location: Toronto

PostPosted: Wed May 18, 2005 12:29 am    Post subject: Reply with quote

I had my custom pace adjustments at hand when I read this, so I added them to your list. "102" means that during that player's time on the floor, the team played at 102% of the league average possessions/minute (90.3). I'm not sure how the pace adjustment would work, so I'll leave it up to you if you want.

John Quincy wrote:
These are based on the weights in Dan's article, but they are not adjusted for pace (what exactly is the pace adjustment you use, Dan?).

Code:

Here are the top ten overall:

1. Andrei Kirilenko    96
2. Kevin Garnett       99
3. LeBron James        99
4. Shaquille O'Neal   100
5. Dirk Nowitzki      102
6. Vince Carter        99
7. Tim Duncan          97
8. Brad Miller        103
9. Larry Hughes       104
10. Amare Stoudemire  105



Here are the top rookies, which I found interesting:



1. Andris Biedrins    100
2. Andre Iguodala     105
3. Anderson Varejao    96
4. Devin Harris       104
5. Delonte West       102
6. Josh Childress     101
7. Josh Smith         101
8. Tony Allen         103
9. Damien Wilkins      96
10. Dwight Howard     104

_________________
ed
Back to top
View user's profile Send private message Send e-mail
Display posts from previous:   
Post new topic   Reply to topic    APBRmetrics Forum Index -> General discussion All times are GMT - 5 Hours
Page 1 of 1

 
Jump to:  
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
You cannot vote in polls in this forum


Powered by phpBB © 2001, 2005 phpBB Group