This has certainly been an interesting discussion.  I do include a second
rating with CCHP that only takes into account wins and losses instead of
scores and it correlates very closely with KRACH and RHEAL.  I believe very
strongly that it is just as worthwhile as the normal CCHP rating (as KRACH
and RHEAL are as well), but it provides a different point of view on the
same data.  I haven't tried because I've been busy lately, but I suspect if
I take out home ice adjustments, my second rating might match exactly with
one of either KRACH or RHEAL.  But I think the home ice adjustments are very
important to take into account.
 
In any case if I were seeding tournament teams or even casting a poll ballot
I almost certainly would rely more heavily on the second CCHP rating, KRACH,
or RHEAL because wins do matter.  RPI however just cannot accurately rate
teams like Quinnipiac because of the extreme overall weakness of the MAAC.
Much of the reason for that is because an extended iterative process that
does converge changes things a lot from how the RPI is calculated.  Teams
that have outstanding records against poor competition slide down the scale
fairly quickly during the process.
 
There are clearly several alternatives besides RPI that would do a much
better job of ranking the teams.
 
Eric Carlson
CCHP ratings
http://www.ptialaska.net/~carlsonj/uafhockey/hratings.html
 
HOCKEY-L is for discussion of college ice hockey;  send information to
[log in to unmask], The College Hockey Information List.