Hi all -
>On Sat, 12 Aug 2000, Garrett Lanzy wrote:
> John Whelan <[log in to unmask]> posted:
>> Details of last month's NCAA Division I Ice Hockey Selection Committee
>> Meeting are in an article in the latest NCAA News, now available
>> online at: <http://www.ncaa.org/news/20000814/active/3717n10.html>
>
> "The committee also discussed the rating-percentage index (RPI), which
> is part of the selection criteria. The group looked at different scen-
> arios from last season and determined that, by and large, the current
> system is efficient."
>
> Interesting choice of words. I find it hard to buy the "efficient"
> description of the RPI when the committee had to implement extra
> rules this past year to account for the fact that it doesn't sufficiently
> consider strength of schedule, especially for "disconnected" scenarios.
Ain't that the truth ....
> Those of us who have been following HOCKEY-L (for longer than we'd like
> to remember in some cases :-) know that there are far better rating
> systems that have already been developed -- by participants on this
> list! I know that change can be difficult to accept, but it's *really*
> time for the committee (and the NCAA) to step up to using a statisti-
> cally sound methodology -- and "PWR using RPI + exceptions" isn't.
I've suspected (with nothing more than a 'hunch' to back it up :-) for
some time that the NC$$ selection committee is loath to devalue the use
of RPI in determining Tournament teams because either/both (1) its use in
football and bouncy-ball tournament selection is well established and has
never (AFAIK) engendered the same level of 'scrutiny' that's happened when
it's been applied to hockey (i.e., if it works for them, the complaining
hockey fans can't possibly have a legitimate beef) and (2) no one has yet
been able to describe to them IN LANGUAGE THEY CAN UNDERSTAND the serious
deficiencies the present RPI brings to the table when used to evaluate the
insular schedules of the CHA and MAAC. The underlying root of RPI is fun-
damentally easy for even coaches to comprehend - my team should benefit by
playing/beating teams with winning records, the more the better. Weight-
tweaking worked fine when only the 'Big 4' conferences were involved, but
I'm convinced that the committee just doesn't realize that these simple
adjustments are wholly inadequate now that the potential Tournament Field
has expanded to include teams from conferences who compete largely amongst
themselves ....
I'll admit that the technical details so often debated here w/r to KRACH
vs. MASSIE vs. CHODR vs. etc. are often lost on me, but I've learned
enough to know that KRACH employs by far the most statistically 'robust'
alogrithm(s) to more fairly evaluate teams with largely insular schedules.
What's needed, IMHO, is for KRACH/Ken Butler to get a fair hearing before
the committee. But they'll listen ONLY if the statistical jargon is kept
to a minimum, and they can be convinced in SIMPLE LANGUAGE that there IS
a better way than RPI, and here's *WHY* it's better. Ken has taken a stab
at it, but there's room for considerable improvement:
<http://www.mscs.dal.ca/~butler/krachexp.htm>
Do I expect this to happen anytime soon ?? Of course not - resistance
to change is hard to overcome, especially when it will be perceived as
"going against the grain" of their Big Time football/basketball brethren.
Maybe it's time to ring up the same folks in the NCAA hierarchy who were
able to push through the BCS (football) ratings despite the loud wailing
and knashing of teeth from Lincoln to Tallahassee. That's a precedent
worth exploiting, if only the committee has the balls to embrace it ....
>> There's also an article on the awarding of the 2004-2006 Frozen
>> Fours, at <http://www.ncaa.org/news/20000814/active/3717n11.html>
>
> I would be interested in hearing more info about plans for how the new
> ticket allocation scheme will work.
So would I :-) I can send 'em xerox copies of my ticket stubs from the
12 Frozen Fours I've attended in person (the first in 1982); will that get
me "priority" over the 'slackers' who've only attended the last 2 or 3
Tournaments ?? <Big Grin> Seriously - how do they expect to identify this
pool of "loyal" fans who supported the Tournament in the past ?? It's a
nice thought, but I'll cynically wager this "rewarding loyal fans" concept
will never amount to anything more than lip service ....
> For example, if the number of applications for tickets from the "repeat"
> pool exceeds the number of tickets available in it, will those then
> automatically get put into the "regular" lottery? (I'd hope so....)
As AOLers would say, "me too !!" ....
> And the big question: is the increase in ticket allotments for the
> teams (which I have *no* argument with) coming out of the "existing"
> NCAA allotment, or are they taking an additional 400 seats out of the
> "remaining" pool (which will then be split 50%-50% between the repeat
> and general "lotteries")? If the answer is that these are coming from
> the existing allotment, I think it's a good thing; if they're keeping
> all of the "VIP" allocations (for such things as travel packages from
> their "preferred" travel agent"), then I'll have to go back to the NC$$
> nomenclature!
I think you know where the odds lay ....
Of greater interest to me are some of the other, less-discussed decisions
made by the committee. Quoting from the NCAA press release:
"In the past, if one team won the regular-season and postseason title,
that team would have received a bye in the first round of the NCAA
championship [the so-called "Clarkson Rule"]. This no longer will be the
case. The top four seeds will be based on the selection criteria. If the
top two seeds nationally are from the same region, the second seed will
be sent to the opposite region and become that region's top seed."
Well, isn't that special .... While it sounds very egalitarian on the
surface (who wouldn't be in favor of rewarding the Top Seeds with the most
favorable preliminary match-ups), I'm less than thrilled about shifting
the top teams out of their home regions. In the scenario given above,
let's say for the sake of argument that BC and Maine earn respective
national ranks of #1 and #2 on Selection Sunday. Would Maine *really*
prefer to fly to Colorado Springs as the #1W (bye) seed vs. playing in
Worcester before a BIG crowd of Maniacs as the #2E (bye) seed ?? Is there
really THAT much of a difference between the #4W/#5W winner vs. the #3E/
#6E winner to justify flying half-way 'cross the country away from your
rabid fans ?? I know that cynics will say that Walsh would MUCH rather
play in the West to avoid #3 in the East, especially if #3E has barely
broken a sweat pasting the MAAC qualifier seeded #6E, but I'm not quite
that cynical :-) I liked knowing that if my team performed well enough
to earn a bye it would stay in its own region; it seemed an appropriate
reward for both the team AND its loyal fans. But now ....
Not only that, there was no mention that avoidance of 1st-round intra-
conference match-ups was any less important, so I presume this "unwritten"
policy will continue under the new "flopping byes" (TM :-) scenario. I
forsee that the combination of these two criteria will result in 8-10
teams switching regions, much as is now the norm for the "regional" B-ball
match-ups as well. When the Tournament (eventually) expands to 16 teams
and 4, 4-team regionals, only the Top 4 seeds may wind up in their "home"
regions, and the make-over of our little hockey tournament to the "demo-
cratic" ways of the squeak-ball Tournament will be complete. Call me an
old fogey, but I think there was a lot to like - especially from a fan's
perspective - in the old system, and I'm sad to see it go ....
Just as a curiosity, has anyone (OK, that's you John :-) computed the
seeds from LAST year's tournament had this new "flopping byes" policy
existed last season ?? Would SLU have been shipped to St. Paul as the #2W
seed ?? How else would the seeds have shifted ?? It's an academic exer-
cise at best, but I'm curious nonetheless - how about it John; pretty
please ?? :-) (Since John's scripts rely on the USCHO data-base that now
reflects each team's *post-tournament* record, I can't do it myself from
here) ....
Before I close, I can't resist:
"Overall, the committee was pleased with the officiating, and thought the
best officials advanced to the Frozen Four .... regardless of conference
affiliation."
I presume this means they wisely kept Jim Fitzgerald as far away from
Providence as possible, as it was his premature whistle in the Colgate-
Michigan opening round contest that denied the Red Raiders their winning
ot goal. All things considered, I'll admit that neutral officials looks
ON PAPER to be the way to go in Tournament games, but there's something
to be said for assigning officials familiar with a team's (and conference)
style of play to the Big Games, especially if both teams are from the
SAME conference. It's unlikely to have changed the outcome, but I think
it would've been a more smoothly flowing game (especially through that
interminable 1st period :-) had a Hockey East official worked the 1999
Maine-UNH final rather than a neutral referee. Just one man's opinion ...
Counting down the days to the IceBreaker - Jim
|
