Sender: |
|
Subject: |
|
From: |
|
Date: |
Wed, 28 Apr 1993 12:29:21 CDT |
Reply-To: |
|
Parts/Attachments: |
|
|
Brian Cavanaugh write:
>Erik Beaver informs me that the RPICH rating that he sends out
>and the RPI rating that the NC$$ selection committee use to determine
>a ranking for playoff selection are vertually the same. OK - that's
>what I was waiting to hear. WHY ????
>
>When I was at the meetings this past week the NC$$ committee told
>the D-I coaches that they could (or would not) send out the
>RPI ratings that they use as the season comes to a close. I sat
>there almost in disbelief - knowing that I followed the ratings
>thru hockey-l all year long (presuming that they may be close
>in statistical accuracy).
>
>Some of the coaches asked if the NC$$ would give out these
>ratings toward the time of selection - and the committee -
>hedged quite a bit and then the NC$$ rep said that did'nt
>seem likely.
Yup. That's why I decided to generate my own numbers. The formula isn't a
secret. I wrote a computer program to crank out the numbers, but a person
could do it with a calculator given the time and inclination.
>It seems that some coaches really don't know who's rated where
>as season ends - MAYBE THEY SHOULD GET ON HOCKEY-L. The RPI
>index has a lot of bearing on the selection and if its "to
>close to call" the committee uses several other tiebreakers
>to determine both selection and seeding. It's these
>"tiebreakers" (head to head, last 20 games, playoff results,
>etc.) and there importance and priority that took up a lot
>of discussion time.
Yes, indeed. I offer the weekly RPICH numbers to the college hockey
community in hope that understanding of the seeding and selection
process will be increased. I certainly hope that coaches, administrators,
and fans find the numbers useful.
-- Erik
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Erik Biever Ski-U-Mah!
[log in to unmask] / [log in to unmask] / [log in to unmask]
******** MINNESOTA GOLDEN GOPHERS - 1993 WCHA PLAYOFF CHAMPIONS ********
|
|
|