>>>No muss, no fuss.  Is this not reality?
>
>>That's about it ... more or less.
>
>        Er, except the part about assigning bids based on the PWR.
>Wasn't the whole lesson of your interview with Joe Marsh
><http://www.uscollegehockey.com/tournament/032097.html> that they're
>now using the individual pairwise comparisons rather than the total
>PWR?
 
That's why I said "more or less" --- Joe didn't even know what PWR
was --- but once it was explained, he understood that it was "more or
less" what they were doing.
 
Instead of, however, absolutely adding up PWR wins as a finite number
... they basically take clusters of teams with similar PWR, and check
out their individial PWR against each other.
 
Since RPI and PWR are usually similar, you will get the same clusters
.... then they head to the comparisons once you are looking at those
clusters.  That's why last year, PWR had I think Miami ahead of Cornell
... but in head to head, Cornell won, so they got the higher seed (or
something like that ... if not those two teams, then replace their names
with others - the concept is the same).
 
It is of my belief that -- which this really doesn't alter things all
that much -- it would be wiser to use the overall wins ... and I think
if it was rationally explained to the committee, they would so how much
better a concept it is.  I just don't think they ever really saw it
explained, and when PWR was first created by the forerunners of this
list, it was because that's how they interpreted what the committee was
doing.
 
And again, the essential concept of PWR is exactly what the committee
does ... again, the only difference is, they don't total up the wins.
 
The reason why they should is the same reason why, for example:
   If Yale has more wins than Clarkson, but Clarkson beat Yale twice in
the season --- that doesn't mean Clarkson wins the ECAC regular season
title.  That would be silly.
 
Same thing for seedings in the tournament, in my opinion .... I think
the team with the most total wins should get the nod -- then if there's
a tie, break it by RPI or head-to-head comparison - whatever.
 
Bottom line --- the result is more or less identical ... but the
potential is there for a difference.
 
AW
 
HOCKEY-L is for discussion of college ice hockey;  send information to
[log in to unmask], The College Hockey Information List.