Subject: | |
From: | |
Reply To: | Wayne T. Smith |
Date: | Tue, 3 Mar 1998 09:36:52 -0500 |
Content-Type: | text/plain |
Parts/Attachments: |
|
|
Bill wrote, in part..
> ... and yet is still unable to climb past the 28th in the Automated
> Ratings Percentage Index. I realize that the Bears got off to a rocky
> start (2-10) back in November-December, but doesn't current performance
> count for more than past misfortune? When you're hot, you're hot!
It's very rare for a rating to include recent performance. The only
major hockey rating scheme using (or to have used) recent performance
(that I recall!) is PWR.
If you were designing a rating scheme, how would you favor recent
performance? How would you decide the strength of your recent
performance rating effect?
If I were to consider "recent performance", I'd like to use historical
data to try to calculate the recent performance effect. Would using
that factor directly be "fair" or "best"?
cheers,
Wayne T. Smith mailto:[log in to unmask]
Systems Group - UNET (formerly CAPS) University of Maine System
Co-owner of the College Hockey mailing lists
HOCKEY-L is for discussion of college ice hockey; send information to
[log in to unmask], The College Hockey Information List.
|
|
|