If the season ended today, 1998 February 2

© 1998, Joe Schlobotnik (archives)

URL for this frameset: http://slack.net/~whelan/tbrw/tbrw.cgi?1998/pairwise.980202.shtml

Everybody's on the same page now; Mankato State is eligible for the tournament and included in the NCAA Tournament selection procedure. With that settled, let's calculate the selection criteria, starting with game results from the Division I Composite Schedule on US College Hockey Online. (USCHO also maintains a current RPI and current PWR, which will agree with these independently-calculated numbers until Monday's games.) This week there are 22 teams with .500 or better winning percentage, and here are the results of pairwise comparisons among them.

    Team         PWR  RPI                  Comparisons Won
 1 Mich State     21 .607   NDBUNHMiWiMmYaBCSCCgOSCrNMNELSCkCCPnRPPvDa
 2 North Dakota   20 .626 __  BUNHMiWiMmYaBCSCCgOSCrNMNELSCkCCPnRPPvDa
 3 Boston Univ    19 .618 ____  NHMiWiMmYaBCSCCgOSCrNMNELSCkCCPnRPPvDa
 4 New Hampshire  17 .619 ______  MiWiMm__BCSCCgOSCrNMNELSCkCCPnRPPvDa
 5 Michigan       17 .596 ________  WiMmYaBCSCCgOSCrNMNELSCkCCPnRPPvDa
 6 Wisconsin      15 .589 __________  Mm__BCSCCgOSCrNMNELSCkCCPnRPPvDa
 7 Miami          15 .585 ____________  YaBCSCCgOSCrNMNELSCkCCPnRPPvDa
 8 Yale           15 .559 ______NH__Wi__  __SCCgOSCrNMNELSCkCCPnRPPvDa
 9 Boston Coll    13 .573 ______________Ya  SC__OSCrNMNELSCkCCPnRPPvDa
10 St Cloud       12 .546 __________________  CgOSCrNMNELSCkCCPnRPPvDa
11 Colgate        12 .553 ________________BC__  OSCrNMNELSCkCCPnRPPvDa
12 Ohio State     10 .534 ______________________  CrNMNELSCkCCPnRPPvDa
13 Cornell         7 .517 ________________________  NMNELS____PnRPPvDa
14 Northern Mich   7 .534 __________________________  NELSCkCC__RPPvDa
15 Northeastern    6 .533 ____________________________  LSCk__PnRPPvDa
16 Lake Superior   6 .504 ______________________________  CkCCPnRPPvDa
17 Clarkson        6 .521 ________________________Cr______  CCPnRPPvDa
18 CO College      6 .534 ________________________Cr__NE____  PnRPPvDa
19 Princeton       4 .510 __________________________NM________  RPPvDa
20 RPI             2 .480 ______________________________________  PvDa
21 Providence      1 .490 ________________________________________  Da
22 Dartmouth       0 .448 __________________________________________  

To play the tournament selection game, let's first assign the automatic berths for regular season champions to the current leaders: Wisconsin, Michigan, Yale and BU. The pairwise comparisons tell us unequivocally that the at-large bids should go to Michigan State, Wisconsin, Miami, St. Cloud and Ohio State in the West, and New Hampshire, BC and Colgate in the East. Since there's an East-West imbalance in the field, we designate Ohio State an honorary Eastern team before ranking the teams within their regions:

      West                                  East
1 Mich State   5 .607 NDMiWiMmSC | 1 Boston Univ   5 .618 YaNHCgBCOS
2 North Dakota 4 .626   MiWiMmSC | 2 Yale          3 .559   NHCg__OS
3 Michigan     3 .596 __  WiMmSC | 3 New Hampshire 3 .619 __  CgBCOS
4 Wisconsin    2 .589 ____  MmSC | 4 Colgate       2 .553 ____  BCOS
5 Miami        1 .585 ______  SC | 5 Boston Coll   2 .573 Ya____  OS
6 St Cloud     0 .546 ________   | 6 Ohio State    0 .534 ________
    

It's easy to see that Michigan State and North Dakota claim the two byes in the West (unless Michigan and/or Wisconsin get the automatic byes for winning their respective tournaments). In the East, BU gets the top bye. Yale and New Hampshire each win three comparisons with "Eastern" teams (including Ohio State), but the Bulldogs win the comparison with UNH, so they would be the second seed. This is despite the fact that in the overall PWR above, UNH ranks fourth in the nation, and Yale only eighth. The two Western teams that UNH beats and Yale loses to don't help the Wildcats here. The teams to be swapped, based on the numbers, are again straightforward in the West (Miami and SCSU); in the East we see Ohio State should be swapped back to the West regional; Colgate and BC each win PWCs with two teams in the East, but Colgate wins the individual comparison with BC and thus outranks them. Making those swaps gives regionals of:

      West                                  East
1 Mich State (C)   1 .607 ND     | 1 Boston Univ (H)   1 .618 Ya
2 North Dakota (W) 0 .626        | 2 Yale (E)          0 .559   

3 Michigan (C)     3 .596 WiBCOS | 3 New Hampshire (H) 3 .619 MmSCCg
4 Wisconsin (W)    2 .589   BCOS | 4 Miami (C)         2 .585   SCCg
5 Boston Coll (H)  1 .573 __  OS | 5 St Cloud (W)      1 .546 __  Cg
6 Ohio State (C)   0 .534 ____   | 6 Colgate (E)       0 .553 ____  
    

Now let's inspect things for conference matchups. If we seeded the tournament as is, Michigan and Ohio State would meet in the first round and Colgate might play Yale in the second. That's easily remedied by swapping the five and six seeds in each region. (Ordinarily when two seeds are swapped to avoid a second-round intraconference game, the other two are also switched to maintain the first-round pairings, but in this case, that would result in a potential--and more likely--second-round game between UNH and BU. So here are the brackets:

5W Ohio State (C)                   6E St Cloud (W)
4W Wisconsin (W)                    3E New Hampshire (H)
      1W Mich State (C)    2E Yale (E)

      2W North Dakota (W)  1E Boston Univ (H)
3W Michigan (C)                     4E Miami (C)
6W Boston Coll (H)                  5E Colgate (E)
    

Are we done, on the first try for once? There's one potential second-round matchup in the West, but that was inevitable given the conference imbalance. From an attendance standpoint, you certainly can't complain about the field in Ann Arbor (although confused fans might show up expecting football games :-)); you might argue the issue of keeping BC in the East over Colgate, but the latter is rather closer to Albany, plus putting BC in the East would mean adding a potential all-Hockey East game in the second round.

The Gory Details

If you want to see the reasons why each of the pairwise comparisons turned out the way it did, here they are.


Last Modified: 2011 October 9

Joe Schlobotnik / joe@amurgsval.org

HTML 4.0 compliant CSS2 compliant