If the season ended today, 2000 March 13

© 2000, Joe Schlobotnik (archives)

URL for this frameset: http://www.slack.net/~whelan/tbrw/tbrw.cgi?2000/pairwise.000313.shtml

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

This analysis is based on the pairwise comparisons at the time it was written. You can also go through this process interactively with the up-to-date results using the "You Are The Committee" script, or you can fill in hypothetical results for the remaining games using the conference tournament bracket or non-table hypothetical interface, and use the resulting pairwise comparisons. You can also use the self-service hypothetical interface to add, remove, or change any results you like from this season and go from there.

College Hockey America has held its conference tournament, and the remaining five Division I conferences hold theirs this weekend, after which the NCAA tournament will be seeded, just six days from now. Wisconsin, Michigan, St. Lawrence and Boston University are all already guaranteed bids for winning the regular season titles in the four established conferences. Up to four more teams can receive auto bids by winning major conference tournaments. The remainder of the field of twelve will be filled out with at-large bids based on the NCAA's system of pairwise comparisons. Here are the current pairwise comparisons among the 26 teams with Division 1 records of .500 or better:

Pairwise Comparisons
Rk Team PWR RPI Comparisons Won
1 Wisconsin (W) 25 .6109 Me ND NH SL Cg BU Ni BC Qn Mi MS SC Mk RP Mn Ck CC Cr LS FS SH NM Pv Io Ca
2 Maine (H) 24 .6000 ND NH SL Cg BU Ni BC Qn Mi MS SC Mk RP Mn Ck CC Cr LS FS SH NM Pv Io Ca
3 North Dakota (W) 22 .5896     SL Cg BU Ni BC Qn Mi MS SC Mk RP Mn Ck CC Cr LS FS SH NM Pv Io Ca
4 New Hampshire (H) 21 .5978   ND SL     Ni BC Qn Mi MS SC Mk RP Mn Ck CC Cr LS FS SH NM Pv Io Ca
5 SLawrence (E) 21 .5812       Cg BU Ni BC Qn Mi MS SC Mk RP Mn Ck CC Cr LS FS SH NM Pv Io Ca
6 Colgate (E) 19 .5720     NH   BU   BC Qn   MS SC Mk RP Mn Ck CC Cr LS FS SH NM Pv Io Ca
7 Boston Univ (H) 19 .5877     NH       BC Qn Mi MS SC Mk RP Mn Ck CC Cr LS FS SH NM Pv Io Ca
8 Niagara (A) 18 .5731         Cg BU BC     MS SC Mk RP Mn Ck CC Cr LS FS SH NM Pv Io Ca
9 Boston Coll (H) 17 .5825               Qn Mi MS SC Mk RP Mn Ck CC Cr LS FS SH NM Pv Io Ca
10 Quinnipiac (M) 17 .5737             Ni   Mi MS SC Mk RP Mn Ck CC Cr LS FS SH NM Pv Io Ca
11 Michigan (C) 17 .5755         Cg   Ni     MS SC Mk RP Mn Ck CC Cr LS FS SH NM Pv Io Ca
12 Mich State (C) 14 .5577                     SC Mk RP Mn Ck CC Cr LS FS SH NM Pv Io Ca
13 SCloud (W) 13 .5450                       Mk RP Mn Ck CC Cr LS FS SH NM Pv Io Ca
14 MSU-Mankato (W) 12 .5319                         RP Mn Ck CC Cr LS FS SH NM Pv Io Ca
15 RPI (E) 10 .5401                             Ck CC Cr LS FS SH NM Pv Io Ca
16 Minnesota (W) 8 .5475                           RP Ck CC   LS FS   NM Pv   Ca
17 Clarkson (E) 8 .5289                               CC Cr LS   SH NM Pv Io Ca
18 CCollege (W) 7 .5270                                 Cr LS FS SH   Pv Io Ca
19 Cornell (E) 7 .5083                             Mn     LS   SH NM Pv Io Ca
20 Lake Superior (C) 5 .4967                                     FS SH NM   Io Ca
21 Ferris State (C) 5 .5197                               Ck   Cr       Pv Io Ca
22 Sacred Heart (M) 4 .4857                             Mn         FS NM Pv    
23 Northern Mich (C) 4 .5128                                 CC     FS   Pv   Ca
24 Providence (H) 3 .5092                                     LS       Io Ca
25 Iona (M) 3 .4905                             Mn           SH NM    
26 Canisius (M) 2 .5066                                         SH     Io

Due to a shortcoming in the criteria currently used by the NCAA, the strength of schedule of teams playing in weak conferences is overestimated. As was the case last season with Quinnpiac, the selection committee is expected to take overall conference strength into consideration when applying the results of the pairwise comparisons. Here is at a table of conferences' performance agaisnt one another and the average Ratings Percentage Index within each conference:

Conference Avg RPI vs HE vs WCHA vs CCHA vs CHA vs ECAC vs MAAC Leader Opp RPI
Hockey East (H) .5332 13-6 10-7 3-2-1 26-15-3 5-0 Me .5249
WCHA (W) .5147 6-13 15-12-1 3-1 10-2-1 0-0 Wi .5040
CCHA (C) .4906 7-10 12-15-1 1-2-1 11-10-2 0-0 Mi .4829
CHA (A) .4900 2-3-1 1-3 2-1-1 5-8-1 15-5-2 Ni .4485
ECAC (E) .4896 15-26-3 2-10-1 10-11-2 8-5-1 4-0 SL .4813
MAAC (M) .4531 0-5 0-0 0-0 5-15-2 0-4 Qn .4358

The MAAC is clearly still a step below the other conferences, so Quinnipiac will almost certainly be left out again this year. Niagara is a tougher case. Their schedule includes a number of respectable opponents, but is still heavily influenced by seven games against CHA opponents Air Force and Army, as well as three more against MAAC opponents (out of a total of 27 Division I games). The average Ratings Percentage Index of the CHA is comparible to that of the ECAC and WCHA, but one-third of that is Niagara's own RPI, not to mention that Air Force is somewhat weaker than their RPI would indicate. The committee will probably take Niagara's pairwise comparison wins with a small grain of salt, and it's hard to tell if that would be enough to pass them over in favor of the next 3 or 4 teams. A more robust statistical measure like KRACH shows them as a top-20 but not top-15 team (18th out of 54 Division I programs; in comparison, Quinnipiac is #42).

If we make the decision to exclude Niagara as well as the MAAC teams, filling out the remaining bids becomes easy, as Maine, New Hampshire, North Dakota, Colgate, Boston College, Michigan State, St. Cloud and Minnesota State-Mankato all win their comparisons with all the rest of the teams under consideration. That gives us the following field of 12:

West East
Wisconsin (W) 5 .611 ND Mi MS SC Mk 1 Maine (H) 5 .600 SL Cg BU NH BC
North Dakota (W) 4 .590 Mi MS SC Mk 2 SLawrence (E) 3 .581 Cg BU   BC
Michigan (C) 3 .575   MS SC Mk 3 Colgate (E) 3 .572   BU NH BC
Mich State (C) 2 .558     SC Mk 4 Boston Univ (H) 2 .588     NH BC
SCloud (W) 1 .545       Mk 5 New Hampshire (H) 2 .598 SL     BC
MSU-Mankato (W) 0 .532         6 Boston Coll (H) 0 .583        

The next step is to give byes to the top two teams in each region according to the pairwise comparisons. Wisconsin, North Dakota, and Maine clearly qualify; the final bye in the East would go to St. Lawrence given the current comparisons, since they and Colgate each win three comparisons against other Eastern teams, but SLU wins the individual comparison with Colgate.

Next, two Eastern teams are switched to the West regional, and vice-versa; based on the pairwise comparisons, that would be UNH and BC for St. Cloud and Mankato. This will also turn out to be desirable to avoid intraconference games in the regionals, so the only reason not to do it would be an attendance argument. The two WCHA teams to be swapped are from Minnesota, and the regional is in Minneapolis, but Big Ten schools Wisconsin, Michigan and Michigan State, as well as rival North Dakota, would presumably generate enough interest in Minnesota. Working with these regions, we get:

West East
Wisconsin (W) 1 .611 ND 1 Maine (H) 1 .600 SL
North Dakota (W) 0 .590 2 SLawrence (E) 0 .581
New Hampshire (H) 3 .598 BC Mi MS 3 Colgate (E) 3 .572 BU SC Mk
Boston Coll (H) 2 .583 Mi MS 4 Boston Univ (H) 2 .588 SC Mk
Michigan (C) 1 .575   MS 5 SCloud (W) 1 .545   Mk
Mich State (C) 0 .558     6 MSU-Mankato (W) 0 .532    

With the natural seedings, we would have potential second-round intraconference games between Maine and BU and SLU and Colgate. The remedy is to switch the seeds of BU and Colgate (simultaneously exchanging SCSU and Mankato to preserve the first-round pairings). This produces the following tournament brackets:

5W Michigan (C)                    6E SCloud (W)      
4W Boston Coll (H)                 3E Boston Univ (H)   
     1W Wisconsin (W)    --+--2E SLawrence (E)   
     2W North Dakota (W) --+--1E Maine (H)         
3W New Hampshire (H)               4E Colgate (E)       
6W Mich State (C)                  5E MSU-Mankato (W)  

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: 2012 March 25

Joe Schlobotnik / joe@amurgsval.org

HTML 4.0 compliant CSS2 compliant