If the season ended today, 2002 January 29

© 2000-2002, Joe Schlobotnik (archives)

URL for this frameset: http://www.elynah.com/tbrw/tbrw.cgi?2002/pairwise.020129.shtml

Game results taken from US College Hockey Online's Division I composite schedule

In honor of today's second Division I advisory committee teleconference (according to the schedule in the NCAA's Men's Division I Championship Handbook) let's take a look at how the NCAA selection procedure might play out if the season ended today.

Much of the seeding is based on pairwise comparisons carried out according to the selection criteria. Here are how those comparisons look for the 28 tournament-eligible teams with records at or above .500:

Pairwise Comparisons (including games of 2002 January 27)

Pairwise Comparisons
Rk Team PWR RPI Comparisons Won
1 Denver U  (W) 27 .6530 NH SC Mn MS Me NM ML CC Mi BU Cr NO NE Ak WM OS Pv Mh Ha BC Ni Ck Da Un SH Qn HC
2 New Hampshire (H) 26 .6269 SC Mn MS Me NM ML CC Mi BU Cr NO NE Ak WM OS Pv Mh Ha BC Ni Ck Da Un SH Qn HC
3 SCloud  (W) 25 .6097   Mn MS Me NM ML CC Mi BU Cr NO NE Ak WM OS Pv Mh Ha BC Ni Ck Da Un SH Qn HC
4 Minnesota  (W) 24 .6188     MS Me NM ML CC Mi BU Cr NO NE Ak WM OS Pv Mh Ha BC Ni Ck Da Un SH Qn HC
5 Mich State  (C) 23 .6097       Me NM ML CC Mi BU Cr NO NE Ak WM OS Pv Mh Ha BC Ni Ck Da Un SH Qn HC
6 Maine  (H) 22 .5896         NM ML CC Mi BU Cr NO NE Ak WM OS Pv Mh Ha BC Ni Ck Da Un SH Qn HC
7 Northern Mich (C) 20 .5778           ML   Mi BU Cr NO NE Ak WM OS Pv Mh Ha BC Ni Ck Da Un SH Qn HC
8 Mass-Lowell  (H) 20 .5931             CC Mi BU Cr NO NE Ak WM OS Pv Mh Ha BC Ni Ck Da Un SH Qn HC
9 CCollege  (W) 19 .5815           NM   Mi   Cr NO NE Ak WM OS Pv Mh Ha BC Ni Ck Da Un SH Qn HC
10 Michigan  (C) 18 .5695                 BU Cr NO NE Ak WM OS Pv Mh Ha BC Ni Ck Da Un SH Qn HC
11 Boston Univ  (H) 18 .5923               CC   Cr NO NE Ak WM OS Pv Mh Ha BC Ni Ck Da Un SH Qn HC
12 Cornell  (E) 16 .5689                     NO NE Ak WM OS Pv Mh Ha BC Ni Ck Da Un SH Qn HC
13 NE-Omaha  (C) 15 .5623                       NE Ak WM OS Pv Mh Ha BC Ni Ck Da Un SH Qn HC
14 NorthEastern  (H) 14 .5460                         Ak WM OS Pv Mh Ha BC Ni Ck Da Un SH Qn HC
15 AK-Fairbanks  (C) 12 .5441                           WM OS Pv Mh   BC Ni Ck Da Un SH Qn HC
16 Western Mich  (C) 11 .5390                             OS Pv Mh   BC Ni Ck Da Un SH Qn HC
17 Ohio State  (C) 11 .5379                               Pv Mh Ha BC Ni Ck Da Un SH Qn HC
18 Providence  (H) 10 .5238                                 Mh Ha BC Ni Ck Da Un SH Qn HC
19 Mercyhurst  (M) 9 .5209                                   Ha BC Ni Ck Da Un SH Qn HC
20 Harvard  (E) 9 .5166                           Ak WM         Ni Ck Da Un SH Qn HC
21 Boston Coll  (H) 6 .5213                                     Ha Ni Ck Da   SH Qn  
22 Niagara  (A) 6 .5138                                         Ck Da Un SH Qn HC
23 Clarkson  (E) 5 .5149                                           Da Un SH Qn HC
24 Dartmouth  (E) 4 .4799                                             Un SH Qn HC
25 Union  (E) 4 .4953                                       BC       SH Qn HC
26 Sacred Heart  (M) 2 .4406                                                 Qn HC
27 Quinnipiac  (M) 1 .4578                                                   HC
28 Holy Cross  (M) 1 .4404                                       BC            

The tournament champions from five conferences receive automatic bids. For this exercise we award these to the current leaders in those leagues; of those, Denver (WCHA), Michigan State (CCHA) and New Hampshire (Hockey East) are all currently in the top five according to the pairwise; Cornell (ECAC) is #12, but would still need the auto bid to get in because Mercyhurst (MAAC) would also receive an automatic bid. In awarding the remaining seven bids, three are obvious (St. Cloud, Minnesota, and Maine), since they win all the comparisons with all the other teams in the running for at-large bids. The tricky part comes in awarding the last four bids. Five teams win their comparisons with everyone "below" them; here is that five-team "bubble":

TeamlPWRRPI Comps Won
Northern Mich (C) 3 .5778 ML   Mi BU
Mass-Lowell (H) 3 .5931   CC Mi BU
CCollege (W) 2 .5815 NM   Mi  
Michigan (C) 1 .5695       BU
Boston Univ (H) 1 .5923     CC  

We need to choose four of those five teams. Presumably NMU and UML are in because they each win three out of a possible four comparisons with other bubble teams. But after that there are two possibilties:

  1. Colorado College are also in because they win two out of four bubble comparisons, and the last spot is between Michigan and BU, with Michigan getting the nod for winning the head-to-head comparison.
  2. We consider CC along with Michigan and BU for the last two spots. Looking at the comparisons among those three teams, each wins one and loses one, so we go to the Ratings Percentage Index, with the two bids thus going to BU and CC.

Given our current understanding about the selection process, this is actually one of the situations where we can't say with certainty who makes the tournament. A similar situation arose in 1999, but other issues were at play and we never got a full explanation of it. But for the sake of guessing, let's say they go with the second choice and give the last two spots to Boston University and Colorado College. That makes the tournament field:

TeamlPWRRPI Comparisons Won
Denver U (W) 11 .6530 NH SC Mn MS Me NM ML BU CC Cr Mh
New Hampshire (H) 10 .6269 SC Mn MS Me NM ML BU CC Cr Mh
SCloud (W) 9 .6097   Mn MS Me NM ML BU CC Cr Mh
Minnesota (W) 8 .6188     MS Me NM ML BU CC Cr Mh
Mich State (C) 7 .6097       Me NM ML BU CC Cr Mh
Maine (H) 6 .5896         NM ML BU CC Cr Mh
Northern Mich (C) 4 .5778           ML BU   Cr Mh
Mass-Lowell (H) 4 .5931             BU CC Cr Mh
Boston Univ (H) 3 .5923               CC Cr Mh
CCollege (W) 3 .5815           NM     Cr Mh
Cornell (E) 1 .5689                   Mh
Mercyhurst (M) 0 .5209                    

The top four teams get byes, and #1 Denver should be bracketed to play #4 Minnesota in the national semifinals. This can be accomplished by making Denver the top seed in the West and Minnesota the second seed in the East. The remaining two bye teams can be left in their natural reasons, with #2 New Hampshire taking the top seed in the East and #3 St. Cloud the second seed in the West.

Now the remaning eight teams need to be placed in the regionals. This may be different than in past years, if the NCAA opts to extend the adjusted policies put in place for last fall's championships to reduce air travel in the current climate. Five of the non-bye teams are from the East and three from the West, so we do need to send one of the Eastern teams to the West Regional in Ann Arbor, MI rather than the East regional in Worcester, MA. The logical choice would be Mercyhurst, which is actually located closer to the West Regional site anyway. However, that would leave three non-bye teams from Hockey East in the same regional, and even under the aforementioned guidelines, teams can be moved to a different region to avoid first-round intraconference games. Since Boston University must play in the East Regional as host, Maine or UMass-Lowell would need to be shipped out, and Maine wins the pairwise comparison, so that gives us the following regionals:

Western Regional
(Grand Rapids, MI)

Eastern Regional
(Worcester, MA)

TeamlPWRRPI Comps Won
Denver U (W) 1 .6530 SC
SCloud (W) 0 .6097
Mich State (C) 3 .6097 ML CC NM
Mass-Lowell (H) 1 .5931 CC  
CCollege (W) 1 .5815   NM
Northern Mich (C) 1 .5778 ML  
TeamlPWRRPI Comps Won
New Hampshire (H) 1 .6269 Mn
Minnesota (W) 0 .6188
Maine (H) 3 .5896 BU Cr Mh
Boston Univ (H) 2 .5923 Cr Mh
Cornell (E) 1 .5689   Mh
Mercyhurst (M) 0 .5209    

We can't quite seed the regionals according to the pairwise comparisons, since Michigan and Northern Michigan can't meet in the first round. We can fix this by giving the sixth seed in the West to Colorado College instead; in this case the pairings are set, although there's some question as to whether UMass-Lowell would be the fourth seed because they have the best RPI among the three bottom teams, which each win one of two possible comparisons with the others, or whether NMU would be seeded higher than UML (and thus be the official home team for their first-round game) because they win the comparison with UML. Assuming the latter gives the following brackets:

5W Mass-Lowell (H)                 6E Mercyhurst (M)    
4W Northern Mich (C)               3E Maine (H)         
     1W Denver U (W)     --+--2E Minnesota (W)          
                           |                             
     2W St Cloud (W)     --+--1E New Hampshire (H)      
3W Mich State (C)                  4E Boston Univ (H)   
6W CO College (W)                  5E Cornell (E)      

The Gory Details

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.


Last Modified: 2020 February 1

Joe Schlobotnik / joe@amurgsval.org

HTML 4.0 compliant CSS2 compliant