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:
Charlie Shub <[log in to unmask]>
Reply To:
Date:
Wed, 22 Jan 1997 10:23:45 -0700
Content-Type:
text/plain
Parts/Attachments:
text/plain (71 lines)
Tim describes how his software computes the pair-wise comparisons.  It
is not clear if the computation (as done by the selection committee) is
done exactly like tim describes or is done by not giving multiple points
for H2H.  For the case cited, giving BU 2 points in H2H makes no
difference to the results.  However, It could for other scenarios.  To
use tim's terminology, a comparison involves five criteria. The
question really being asked is whether each of the criteria counts
equally in determining the comparison result.  Tim's program gives
excess weight to the H2H crteria.  Can anybody confirm how the
selection committee does it?  Consider a scenario where team A plays
team B 5 times and gets swept, but team A has done better in every
other criteria.
        Team A          TEAM B
RPI     0.6000  1       0.5555  0
L20     15-5-0  1       10-9-1  0
TUC      6-5-0  1        5-6-0  0
COP      6-0-0  1        0-6-0  0
H2H             0               ?  (either 1 or 5)
======================================
PTS             4          (1 or 5)
 
my understanding of the criteria is that Team A would win the Pair
Wise Comparison result because it has won 4 of the 5 criteria.  Tim
seems to be representing that team B would win the pair wise
comparison point.
 
 
> On Wed, 22 Jan 1997 11:16:24 -0500 Tim Brule <[log in to unmask]> wrote:
> My comment on this is a bit ambiguous. The best way to understand the PWR
> is to look at a particular comparison. I display all comparisons between
> Teams Under Considerations (TUC) on USCHO. Since BU vs UNH was mentioned
> here it is
>
>     Boston University   vs New Hampshire
> RPI      0.5757  0           0.6052  1
> L20    12- 3- 5  0         16- 4- 0  1
> TUC     3- 3- 1  0          6- 6- 0  0
> H2H              2                   0
> COP     7- 3- 3  0         10- 2- 0  1
> ============================================
> PTS              2                   3
>
> Key:
> RPI -- Ratings Percentage Index
> L20 -- Record in Last 20 Games
> TUC -- Record vs Teams Under Consideration
> H2H -- Head to Head
> COP -- Record vs Common Opponents
>
> Notice that BU got 2 points for its two wins vs UNH, but UNH
> won the overall comparison 3 to 2 and those gets "1 OVERALL comparison
> win". Thus a "comparison" consists of the 5 criteria and each comparison
> won by a team is worth one point.
>
> Tim Brule
> [log in to unmask]
>
> HOCKEY-L is for discussion of college ice hockey;  send information to
> [log in to unmask], The College Hockey Information List.
>
 
 
 
      charlie shub   University of Colorado at Colorado Springs
[log in to unmask]  -or-  [log in to unmask]  -or-  [log in to unmask]
                    http://www.cs.uccs.edu/~cdash
            (719) 262-3492               (fax) 262-3369
 
HOCKEY-L is for discussion of college ice hockey;  send information to
[log in to unmask], The College Hockey Information List.

ATOM RSS1 RSS2