Time to engage in more equine sado-necro-bestiality...
 
Date sent:              Thu, 30 Mar 2000 23:22:03 -0500
From:                   De eP <[log in to unmask]>
Subject:                Re: NCAA seeding comments
 
[ Could you please try to keep your line length to under 80 columns
(preferably, under 72) in the future?  Thank you.]
 
> Eric Burton wrote:
>
> > Actually if you ask the Sioux Hockey Players I think you will see that they
> > respected What Niagara accomplished this season and made mention of it in
> > the papers.
>
> Well, I still come out on the side that Niagara's victory should result in:
>
> 1) Respect for Niagara (and saying that other team's could have beat
> UNH is an argument that appears to be disrespectful)
 
???
 
It's not Niagara's fault that UNH played close games all year and
wasn't playing all that well going into the tournament.  Saying that
other teams could have beaten UNH says no more about Niagara relative
to those other teams than does the fact that Niagara did beat UNH
(despite attempts to the contrary by some people on this list).
 
> 2) Belief that the current team selection system works
 
If needing to resort to ad-hoccery to get even a nearly correct field
qualifies as "working," I guess.  IMO, any supposedly objective system
that requires the committee to make a subjective judgement that a team
"isn't as good as they look" (Quinnipiac) is broken.
 
> Several people have posted that the current system does not have
> enough strength of schedule component to it.
 
That would be because it doesn't.
 
> I don't care.  All of the other systems require math majors or
> engineers to understand them.
 
And the RPI and PWR don't?
 
Have you ever tried calculating the RPI and PWR?  It's theoretically
possible to do without a computer, but it would require a huge amount
of time.  They're not as simple as they look, and it's possible to make
(IMO) vast improvements without significantly changing the structure of
the rankings (e.g. KRACH/KPWR -- I would argue that KRACH is
conceptually easier to understand than RPI).
 
> It is hard enough to get people to understand PWR.  With Niagara's
> victory it just goes to show there is no need to switch from PWR b/c
> it seemed to do an admirable job at picking the top 12 teams.
 
Whatever.  We've been over this so many times it's not worth correcting
you.
 
> And based on the non-conference schedules which seem to be appearing
> for Alabama-Huntsville, Niagara, Quinnipiac, UConn, Holy Cross,
> Canisius, Mercyhurst, etc. (from preliminary reports, all these
> schedules seem to have quite a few big 4 teams on their schedules),
> it seems that RPI should start ranking all of these squads
> appropriately again.
 
> So my vote is for keeping the current system.
 
It might seem that way, but in my opinion, the deficiencies of the RPI
will just be masked by numbers which are not obviously incorrect.  This
is a golden opportunity to make improvements, because there are obvious
problems that need fixing.  In five years when those problems are no
longer as apparent, it may be impossible to correct the subtle
deficiencies which can hurt teams.
 
Remember CC in 1994 - they were victimized by the weaknesses of the
selection system then in effect, and a brief analysis on my part shows
that they would still be victimized by the selection system today, but
for the "CC Rule."  Because the source of this error was not obvious,
the committee fixed the symptom rather than the underlying problem.
 
> Yes, if a squad like Niagara has a slight edge b/c their last 16
> games are against a little weaker group than another squads, so be
> it.  You could say that an ECAC squad of the past may have had some
> similar advantage b/c the ECAC may have been considered weaker.
 
Definitely.  This is further argument for modifying the system --
Niagara and QC are not the only beneficiaries of the ideosyncracies of
the current system, only the most flagrant.
 
> Let the current system do its job.  RPI will be statistically correct
> again from here on out
 
I'll let the Statistics PhD's on the list refute this one.
 
> and it plays a big part in the current system.  And having both a
> MAAC team and a CHA team in next year's tourney will only help all of
> college hockey.
 
I'd love to see the MAAC and CHA in the tournament, much as I was happy
to see Niagara (or MSU-Mankato, but they didn't make it) in the
tournament this year... but it's important that the process be fair.
At this point, I'm uncomfortable with bumping at-large bids before we
expand the tournament, though it looks as though that will be
happening.  And I think the benefit for college hockey is debatable.
 
 
--
 Craig Powers                   NU ChE class of '98
 [log in to unmask]       http://lynx.neu.edu/home/httpd/c/cpowers
 [log in to unmask]              http://www.hal-pc.org/~enigma
 
"Good..bad....I'm the guy with the gun." -- "Ash" in *Army of Darkness*
 
HOCKEY-L is for discussion of college ice hockey;  send information to
[log in to unmask], The College Hockey Information List.