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 

Basic questions about Pythagorean winning percentage

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



Joined: 03 Dec 2006
Posts: 34

PostPosted: Tue Jan 22, 2008 4:53 am    Post subject: Basic questions about Pythagorean winning percentage Reply with quote

I'm hoping these are easy questions anybody here can answer. You can skip the ones that might be controversial or require opinion, as I really only need the simple stuff.

At any given point within a season, when attempting to predict subsequent games during the same season, which is better? Actual winning percentage or Pythagorean winning percentage? Are both useless for this purpose because of luck, strength of schedule, streakiness, etc.?

When looking at end-of-season records, what is the relationship between actual winning percentage and Pythagorean winning percentage? How well does the latter explain the former? When there is a large disparity between a team's actual and Pythagorean winning percentages, would we do better to attribute that difference to luck or to over/underachieving?

Is there any substantial difference when figuring Pythagorean winning percentage between using points for/against and using offensive/defensive efficiency? If so, which is better?

I know, I know, I need to sit down and read BOP.
Back to top
View user's profile Send private message
Kevin Pelton
Site Admin


Joined: 30 Dec 2004
Posts: 709
Location: Seattle

PostPosted: Tue Jan 22, 2008 10:54 pm    Post subject: Re: Basic questions about Pythagorean winning percentage Reply with quote

Tim Lehrbach wrote:
At any given point within a season, when attempting to predict subsequent games during the same season, which is better? Actual winning percentage or Pythagorean winning percentage? Are both useless for this purpose because of luck, strength of schedule, streakiness, etc.?

The difference between the two isn't huge, but differential tends to be a better predictor after the first month or so of the season. By midseason, both have correlations of about 0.7 with the team's record in the second half of the season.

Quote:
Is there any substantial difference when figuring Pythagorean winning percentage between using points for/against and using offensive/defensive efficiency? If so, which is better?

If you're averaging team and opponent possessions and using the Pythagorean method, there is no difference. Offensive and Defensive Ratings can be a little misleading if possessions aren't equal. If you're using Expected Wins or pure point differential as opposed to the Pythagorean method, then pace can play a factor.
Back to top
View user's profile Send private message Send e-mail Visit poster's website AIM Address
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