HOCKEY-L Archives

- Hockey-L - The College Hockey Discussion List

Hockey-L@LISTS.MAINE.EDU

Options: Use Forum View

Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Peter Kester <[log in to unmask]>
Reply To:
Peter Kester <[log in to unmask]>
Date:
Thu, 2 Dec 1993 16:40:04 -0500
Content-Type:
text/plain
Parts/Attachments:
text/plain (37 lines)
Tim Harrison writes:
> Rec   = the team's winning percentage x 1000
> OpRec = the average of the team's opponents' record rating
> Eff   = 1000 x GF/(GF+GA) : What this represents is, of all
>         the points scored in a game between this team and
>         an opponent, this team scored this % of them.
> OpEff = the average of the team's opponents' efficiency ratings
>
> *The x 1000 is included only to eliminate decimals past tenths.
>  Ex.  Rec = 900 yields a winning percentage of 0.900
>       Eff = 600 means 60.0% of goals scored between this team
>             and its opponents are scored by this team.
>
> TOTAL = Rec + OpRec + Eff + OpEff
>
 
I'm not sure that I like this new rating system.  It gives strength of
schedule too much (IMHO) emphasis. Take this contrived example:
 
2 games played:    Team A - 3     Team A - 1
                   Team B - 9     Team C - 4
 
stats are:
Rank Team    Record   Rec  OpRec  Eff  OpEff  TOTAL
 1   Team C   1-0    1000      0  800   235   2035
 2   Team A   0-2       0   1000  235   775   2010
 3   Team B   1-0    1000      0  750   235   1985
 
So undefeated team B is ranked lower than winless team A.  Perhaps
this anomaly goes away when more games are played.  If every team
played every other team this system might work, but of course they don't.
 
Comments?
 
Pete Kester
[log in to unmask]

ATOM RSS1 RSS2