URL for this frameset: http://elynah.com/tbrw/tbrw.cgi?2002/pairwise.020305.shtml
Game results taken from US College Hockey Online's Division I composite schedule
The regular season has come to an end for all of Division I outside of College Hockey America, and the other five conferences are about to start their playoffs. Vermont, Union, Fairfield, Bentley, American International, and UMass-Amherst have all finished their seasons with losing records, and cannot make the NCAAs. To get a peek at how the rest of the teams are doing, let's see see how the NCAA selection procedure would play out if the season ended today.
First, the pairwise comparisons among the 28 tournament eligible teams with non-losing records (against other eligible teams):
The top eleven teams win all of their comparisons with all the teams below them. If we assume that four of the five conference champions receiving automatic bids come from that group (which includes the top seeds in the CCHA, WCHA, ECAC, and Hockey East tournaments), and that the fifth goes to the top seed in the MAAC tournament, Mercyhurst, we get a tournament field consisting of:
Team | lPWR | RPI | Comparisons Won | ||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|
New Hampshire (H) | 11 | .6174 | Mn | DU | Mi | BU | MS | Me | Cr | SC | CC | Ak | Mh |
Minnesota (W) | 10 | .6225 | DU | Mi | BU | MS | Me | Cr | SC | CC | Ak | Mh | |
Denver U (W) | 8 | .6202 | Mi | BU | MS | Me | SC | CC | Ak | Mh | |||
Michigan (C) | 7 | .5879 | BU | MS | Me | Cr | CC | Ak | Mh | ||||
Boston Univ (H) | 6 | .6041 | MS | Me | Cr | CC | Ak | Mh | |||||
Mich State (C) | 6 | .5945 | Me | Cr | SC | CC | Ak | Mh | |||||
Maine (H) | 5 | .5828 | Cr | SC | CC | Ak | Mh | ||||||
Cornell (E) | 5 | .5784 | DU | SC | CC | Ak | Mh | ||||||
St Cloud (W) | 5 | .6068 | Mi | BU | CC | Ak | Mh | ||||||
CO College (W) | 2 | .5751 | Ak | Mh | |||||||||
AK-Fairbanks (C) | 1 | .5665 | Mh | ||||||||||
Mercyhurst (M) | 0 | .5431 |
The top four teams in the country get first-round byes. That group obviously includes UNH and Minnesota, and also pretty clearly Denver and Michigan. These teams should be seeded so that #1 UNH would play #4 Michigan in the NCAA semifinals, which can be accomplished by making UNH the top seed in the East and Michigan the second seed in the West (the Wolverines have to play in the Western Regional, which they host). Given their superiority according to the selection criteria, Minnesota should be the top seed in the West, with Denver getting the number two seed in the East.
Of the remaining eight tournament teams, four come from the East and four from the West. In past years, the NCAA would have switched some of these teams into the other regionals, but in the wake of the September 11, 2001 terrorist attacks, they have decided to seed postseason tournaments so as to minimize air travel. Ordinarily, this would mean keeping all eight teams in their own region, but in this case, Mercyhurst would actually have a shorter trip from Erie, PA to the West Regional in Ann Arbor, MI than to the East Regional in Worcester, MA, and Alaska-Fairbanks would have to fly into the same time zone for either regional. So let's assume the committee would switch these teams after all. This sets up the following regionals:
Western Regional | Eastern Regional | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
The East Regional can be seeded straight-up according to the pairwise comparisons; in the West we need to change things around to avoid a SCSU-CC game in the first round. Switching St. Cloud with Michigan State also gets rid of the potential second-round all-CCHA game (it's not such a big deal any more, but we get this one "for free"), so let's do that, which gives us NCAA tournament brackets of:
5W CO College (W) 6E AK-Fairbanks (C) 4W Mich State (C) 3E Boston Univ (H) 1W Minnesota (W) --+--2E Denver U (W) | 2W Michigan (C) --+--1E New Hampshire (H) 3W St Cloud (W) 4E Maine (H) 6W Mercyhurst (M) 5E Cornell (E)
If you want to have a look at why each pairwise comparison turned out the way it did, you can click on the individual comparisons in the table at the top of this article for a breakdown of criteria. You can also go though the process yourself using today's pairwise comparisons with the "You Are The Committee" script.