Content-Transfer-Encoding: |
7bit |
Sender: |
|
Subject: |
|
From: |
|
Date: |
Thu, 25 Jun 1998 13:01:13 -0400 |
Content-Type: |
text/plain; charset=us-ascii |
MIME-Version: |
1.0 |
Reply-To: |
|
Parts/Attachments: |
|
|
A couple observations about how poll numbers add up:
(a) Teams ranked 11-15 in a "Top 10" poll aren't the same as the result would be
in a "Top 15" poll where all voters submit ballots for 15 (or more) places.
(b) The results for teams ranked, say, 7-10 in a "Top 10" poll using a 10-team
ballot are different than if a 15-team ballot was used to select the top 10.
(c) The team receiving the 18th highest vote in a poll using 10-place ballots
got there because someone thought that team should be 10th, not because all
voters considered the top 18 places.
If all voters submit, say, 15-place ballots, that 18-place team might drop to
23rd. When ALL voters submit points for places 11-15, the results for anything
below 7th place is affected. I think a poll selecting the "Top 10" should use a
ballot ordering each voter's choice of the top 15 teams.
Similarly, a "Top 15" poll ought to use about a 20-place ballot. Ordering 20
teams from a national field of 54 is a strain for any hockey mind. That's
probably the deciding factor in what makes results meaningful. I don't think
anything beyond 15 places could be meaningful, but the "also receiving votes"
list would still be interesting and useful.
I vote for a "Top 15" decided by a 20-place ballot.
Bob Griebel
HOCKEY-L is for discussion of college ice hockey; send information to
[log in to unmask], The College Hockey Information List.
|
|
|