John Edwards writes:
>On Mon, 9 Mar 1998, John T. Whelan wrote:
>> It sounds to me that they identify the "lock" teams, and hence
>> the "bubble" ones by using the PWR--or more precisely the comparisons
>> among teams not receiving automatic bids--and *not* by the RPI.
>This sounds like splitting hairs to me. There is little difference
>between RPI and PWR, in terms of generally where a team sits.
Well, sure, it doesn't make a difference ... until it makes a
difference. Might as well be working with the correct info just in case.
>Here's what the current rankings give us:
> R.P.I. PWR
>*1 Boston University *1 North Dakota
>*2 North Dakota *2 Boston University
>*3 Michigan State *3 Michigan State
> 4 Michigan 4 Michigan
> 5 New Hampshire 5 New Hampshire
> 6 Boston College 6 Clarkson
> 7 Clarkson *7 Yale
>*8 Yale 8 Boston College
> 9 Ohio State 9 Ohio State
>10 Colorado College 10 Wisconsin
>11 Wisconsin 11 Colorado College
>12 St. Cloud 12 St. Cloud
>13 Miami 13 Miami
>14 Northeastern 14 Northeastern
>15 Northern Michigan 15 Minnesota-Duluth
>16 Minnesota-Duluth 16 Northern Michigan
>*-automatic berths
Remember, if you just look at the order of teams on USCHO's
PWR page, that they break all ties by looking at the RPI, which makes
their PWR ordering closer to the RPI ordering than if they used the
individual comparisons to break the ties (which is a lot closer to how
the committee uses the PWCs). Using the individual comparisons as
tiebreakers switches Clarkson with UNH and OSU with BC in the PWR
ordering. Anyway, with today's comparisons, at-large bids are a
no-brainer, since the top twelve teams in the PWR win every comparison
with the other 8 TUCs, which means it doesn't matter.
>I'm going to ask the Clarkson Question again. Suppose:
>a) Clarkson loses in the ECAC first round.
>b) The conference tournaments are all won by teams who didn't
>win the regular season and currently are lower than 10th. (Suppose
>Northeastern, Princeton, Miami and Minnesota-Duluth.)
>That puts only four at-large berths in play. If we assume that Michigan
>and UNH will get berths, is it possible for Ohio State and B.C. to
>advance high enough to beat out Clarkson?
It comes down to how the comparisons in question look, and for
that there's nothing to do but bite the bullet and look at
<http://uscollegehockey.com/pwr/comparisons.txt>:
Clarkson vs New Hampshire
RPI 0.5699 0 0.5762 1
L20 15- 4- 1 1 13- 6- 1 0
TUC 5- 6- 2 1 5- 6- 1 0
H2H 0 0
COP 4- 2- 0 1 4- 3- 1 0
============================================
PTS 3 1
This looks tougher than it is; since Vermont is a common
opponent, two-and-out for Clarkson against the Catamounts in the ECAC
quintafinals would flip the common opponents criterion, plus a split
against teams under consideration in the Hockey East tournament (and
remember, *any* tournament champion is a TUC, regardless of their
overall record) would give UNH that criterion a well. As long as they
maintain the lead in RPI, New Hampshire only need to flip one of those
two criteria, so yes, they're likely to overtake Clarkson if the
Knights crash and burn.
Clarkson vs Ohio State
RPI 0.5699 1 0.5560 0
L20 15- 4- 1 1 14- 5- 1 0
TUC 5- 5- 1 1 7- 8- 0 0
H2H 0 1
COP 2- 0- 0 0 3- 0- 0 0
============================================
PTS 3 1
Unless they pass Clarkson in the RPI, Ohio State needs to
reverse two of the criteria. Fortunately for them, Clarkson's 19th
and 20th games into the pat were both wins, so being swept by Vermont
would drop the Knights to 13-6-1 in Last 20. Unfortunately, since
Lake State is right at .500 and will cease to be a TUC if they lose to
Ohio State, OSU can't tie Clarkson on TUC record without winning the
CCHA tournament, so this comparison can't go anywhere without a big
change in the RPI.
Boston College vs Clarkson
RPI 0.5714 1 0.5699 0
L20 11- 4- 5 0 15- 4- 1 1
TUC 5- 4- 2 1 4- 6- 2 0
H2H 0 1
COP 5- 2- 2 0 8- 1- 0 1
============================================
PTS 2 3
BC can easily reverse the comparison with Clarkson, since a
3-1 playoff run would put them 14-3-3 in the last 20 games, which
would trump Clarkson's projected 13-6-1 and give them a 3-2 win on
criteria.
So we have BC, UNH and Michigan able to pass Clarkson; we need
one more. Wisconsin and Colorado College are the only obvious bubble
teams to consider. Clarkson currently wins the CC comparison 3-0;
Wisconsin looks good
Clarkson vs Wisconsin
RPI 0.5699 1 0.5456 0
L20 15- 4- 1 1 12- 7- 1 0
TUC 5- 6- 2 0 8- 6- 1 1
H2H 0 0
COP 2- 1- 0 0 4- 0- 1 1
============================================
PTS 2 2
============================================
PTS 14 0
but with Wisconsin looking at a 12-8-0 last 20 record even if they go
4-1 in the WCHA playoffs there's no way for this comparison to turn
over.
So I think Clarkson is basically secure unless Ohio State
overtakes them in the RPI (*and* we have four darkhourse tournament
champions, *and* Clarkson gets swept by UVM, *and* UNH and BC perform
admirably).
>Oh, and if I flunk my Physics test tomorrow, I'm blaming you, Whelan. Any
>advice on subatomic particles and the Theory of Relativity would be
>appreciated. :-P
Great, can I blame you if I'm unemployed when my postdoc ends
in the fall? :-/ Just remember, it's turtles all the way down.
John Whelan, Cornell '91
Official Scorer/PA Announcer
U of Utah Ice Hockey Club
<[log in to unmask]>
<http://www.cc.utah.edu/~jtw16960/joe.html>
HOCKEY-L is for discussion of college ice hockey; send information to
[log in to unmask], The College Hockey Information List.
|
