Last Modified: 2011 October 9
Joe Schlobotnik / joe@amurgsval.org
URL for this frameset: http://slack.net/~whelan/tbrw/tbrw.cgi?1999/pairwise.990308.shtml
Remember, you can go through this process interactively using the "You Are The Committee" script.
Well, the regular season has ended in the five division I conferences, and we can begin to make some more definitive statements about the NCAA Tournament selection procedure. First, we know that the regular season champions of the four established conferences, North Dakota (WCHA), Michigan State (CCHA), Clarkson (ECAC) and New Hampshire (Hockey East) will be in the field of twelve. We also know that Army, Air Force, Alaska-Fairbanks, Miami, Western Michigan, Dartmouth, Union and Northeastern are finished for the season. And while Canisius, Iona and Fairfield will all see action in the MAAC playoffs for the next weekend or two, they are mathematically eliminated from NCAA contention, since they will all finish below .500 in Division I play even if they run the table, and there is no automatic bid for the MAAC champion. Boston University, UMass-Amherst, Brown, Lake Superior State, Merrimack, Michigan Tech and Minnesota-Duluth are also doomed to losing records, but can still qualify for the NCAAs if they win their respective conference tournaments.
To get a feel for how the four to eight at-large bids might shape up, and how the tournament might be seeded, let's run through the process using the results (through the end of the regular season) from US College Hockey Online's Division I Composite Schedule. We find 26 teams with .500 or better records, or Teams Under Consideration, and perform pairwise comparisons between them, with the following results:
1 North Dakota 25 .650 NHMeMSCkCCDUQnBCSLMiOSNtNMRPPnCgCtMkPvNiMLFSYaBGHC
2 New Hampshire 24 .636 __ MeMSCkCCDUQnBCSLMiOSNtNMRPPnCgCtMkPvNiMLFSYaBGHC
3 Maine 23 .618 ____ MSCkCCDUQnBCSLMiOSNtNMRPPnCgCtMkPvNiMLFSYaBGHC
4 Mich State 22 .603 ______ CkCCDUQnBCSLMiOSNtNMRPPnCgCtMkPvNiMLFSYaBGHC
5 Clarkson 21 .583 ________ CCDUQnBCSLMiOSNtNMRPPnCgCtMkPvNiMLFSYaBGHC
6 CO College 20 .588 __________ DUQnBCSLMiOSNtNMRPPnCgCtMkPvNiMLFSYaBGHC
7 Denver U 19 .557 ____________ QnBCSLMiOSNtNMRPPnCgCtMkPvNiMLFSYaBGHC
8 Quinnipiac 18 .554 ______________ BCSLMiOSNtNMRPPnCgCtMkPvNiMLFSYaBGHC
9 Boston Coll 17 .574 ________________ SLMiOSNtNMRPPnCgCtMkPvNiMLFSYaBGHC
10 St Lawrence 16 .550 __________________ MiOSNtNMRPPnCgCtMkPvNiMLFSYaBGHC
11 Michigan 14 .550 ____________________ OSNtNMRPPnCgCtMkPv__MLFSYaBGHC
12 Ohio State 12 .527 ______________________ NtNM____CgCtMkPvNiMLFSYaBGHC
13 Notre Dame 12 .542 ________________________ NMRPPnCgCtMkPv__MLFSYaBGHC
14 Northern Mich 12 .535 __________________________ RPPnCgCtMkPvNiMLFSYaBGHC
15 RPI 12 .533 ______________________OS____ PnCgCtMkPvNiMLFSYaBGHC
16 Princeton 11 .533 ______________________OS______ CgCtMkPvNiMLFSYaBGHC
17 Colgate 8 .521 ________________________________ CtMkPv__MLFSYaBGHC
18 Connecticut 8 .521 __________________________________ MkPvNiMLFSYaBGHC
19 MSU-Mankato 6 .518 ____________________________________ PvNiMLFSYa__HC
20 Providence 6 .513 ______________________________________ NiMLFSYaBGHC
21 Niagara 6 .479 ____________________Mi__Nt______Cg______ __FS__BGHC
22 Mass-Lowell 4 .505 ________________________________________Ni __YaBGHC
23 Ferris State 3 .508 __________________________________________ML YaBG__
24 Yale 3 .503 ________________________________________Ni____ BGHC
25 Bowling Green 2 .502 ____________________________________Mk__________ HC
26 Holy Cross 1 .493 ____________________________________________FS____
The first thing we need to do, thanks to an NCAA News report which stated that the selection committee "reserves the right to evaluate each team based on the relative strength of their respective conference", and the fact that the new MAAC conference plays no games against any teams from the four established ones, is determine whether the MAAC has reached "competetive equity". Looking at the average ratings percentage index of each of the five conferences, as well as their performances against the four eligible Division I independents:
vs Indies vs Army vs Niagara vs AFA vs Mankato
Avg RPI PF-PA Pct PF-PA Pct PF-PA Pct PF-PA Pct PF-PA Pct
HE .526 14- 2 .875 12-0 1.000 0- 2 .000 2-0 1.000 0-0 .---
WCHA .506 28- 8 .778 0-0 .--- 0- 0 .--- 10-0 1.000 18-8 .692
CCHA .505 5- 5 .500 0-0 .--- 2- 4 .333 0-0 .--- 3-1 .750
ECAC .497 28-10 .737 10-0 1.000 10-10 .500 2-0 1.000 6-0 1.000
MAAC .452 8-22 .267 5-5 .500 0- 4 .000 3-5 .350 0-8 .000
it looks like that standard has not been met. Since the records of MAAC teams in the last 16 games and vs other teams under consideration, and to a lesser extent their RPIs, will be dominated by games against other MAAC teams, this means that those criteria will not be an accurate reflection of their strength relative to the rest of Division I. So it seems likely that the committee will simply exclude Quinnipiac, UConn and Holy Cross from consideration for at-large bids.
Maine, Colorado College, Denver, Boston College and St. Lawrence all win comparisons with all of the remaining teams, so it's a no-brainer to give them at-large bids. Dropping the teams that have no shot from the bottom, we take a closer look at the comparisons among the following eight "bubble" teams:
Team lPWR RPI Comps Won 1 Michigan 8 .550 OSNtNMRPPn__CgMkPv 2 Ohio State 6 .527 __ NtNM____NiCgMkPv 3 Notre Dame 6 .542 ____ NMRPPn__CgMkPv 4 Northern Mich 6 .535 ______ RPPnNiCgMkPv 5 RPI 6 .533 __OS____ PnNiCgMkPv 6 Princeton 5 .533 __OS______ NiCgMkPv 7 Niagara 3 .479 Mi__Nt______ Cg____ 8 Colgate 2 .521 ______________ MkPv 9 MSU-Mankato 2 .518 ____________Ni__ Pv 10 Providence 1 .513 ____________Ni____
Michigan is in, Princeton, Niagara, Colgate, Mankato and Providence are out, and among the remaining four OSU and Notre Dame win two comparisons and lose one each, while NMU and Rensselaer each win one and lose two, so the former two get the last two at-large bids. The resulting tournament field has seven Western teams and only five from the East, so we pretend that Notre Dame, the lowest-rated Western team, is from the East:
West East
1 North Dakota 5 .650 MSCCDUMiOS | 1 New Hampshire 5 .636 MeCkBCSLNt
2 Mich State 4 .603 CCDUMiOS | 2 Maine 4 .618 CkBCSLNt
3 CO College 3 .588 __ DUMiOS | 3 Clarkson 3 .583 __ BCSLNt
4 Denver U 2 .557 ____ MiOS | 4 Boston Coll 2 .574 ____ SLNt
5 Michigan 1 .550 ______ OS | 5 St Lawrence 1 .550 ______ Nt
6 Ohio State 0 .527 ________ | 6 Notre Dame 0 .542 ________
On the basis of pairwise comparisons, the byes go to North Dakota and Michigan State in the West and UNH and Maine in the East, although an ECAC tournament championship by regular season winners Clarkson would entitle them to an automatic bye at Maine's expense. Next we want to swap two teams from each region. On the basis of pairwise comparisons, that would be SLU and Notre Dame for Michigan and OSU, which would give:
West East
1 North Dakota (W) 1 .650 MS | 1 New Hampshire (H) 1 .636 Me
2 Mich State (C) 0 .603 | 2 Maine (H) 0 .618
3 CO College (W) 3 .588 DUSLNt | 3 Clarkson (E) 3 .583 BCMiOS
4 Denver U (W) 2 .557 SLNt | 4 Boston Coll (H) 2 .574 MiOS
5 St Lawrence (E) 1 .550 __ Nt | 5 Michigan (C) 1 .550 __ OS
6 Notre Dame (C) 0 .542 ____ | 6 Ohio State (C) 0 .527 ____
This can be fine-tuned to avoid potential second-round intra-conference matchups or to improve attendance at the regionals. In the East, sending Boston College West instead of St. Lawrence would avert a possible Hockey East matchup in the second round, but at the expense of a large draw at the regionals in Worcester, Mass, so I seriously doubt the committee would do that. A second-round intra-conference game is inevitable in the West, given the presence of seven Western teams in the tournament, however some attendance-driven tinkering may be done. A natural choice would seem to be to switch Michigan with Notre Dame to increase the gate in Madison, which would also have the effect of nullifying the artificial reward Notre Dame got for being the seventh-best team in a six-team region. OSU for Denver is also possible, but seems a little unfair based on the pairwise comparisons. So let's assume Michigan and Notre Dame switch places as our "best guess" scenario:
West East
1 North Dakota (W) 1 .650 MS | 1 New Hampshire (H) 1 .636 Me
2 Mich State (C) 0 .603 | 2 Maine (H) 0 .618
3 CO College (W) 3 .588 DUSLMi | 3 Clarkson (E) 3 .583 BCOSNt
4 Denver U (W) 2 .557 SLMi | 4 Boston Coll (H) 2 .574 OSNt
5 St Lawrence (E) 1 .550 __ Mi | 5 Ohio State (C) 1 .527 __ Nt
6 Michigan (C) 0 .550 ____ | 6 Notre Dame (C) 0 .542 ____
Seeding the teams based on their pairwise comparisons leads to two potential second-round conference matchups in the West, so we switch Michigan and SLU, and also CC and DU to preserve first-round pairings, and get the following brackets:
5W Michigan (C) 6E Notre Dame (C)
4W CO College (W) 3E Clarkson (E)
1W North Dakota (W) --+--2E Maine (H)
|
2W Mich State (C) --+--1E New Hampshire (H)
3W Denver U (W) 4E Boston Coll (H)
6W St Lawrence (E) 5E Ohio State (C)
Wanna see the reasons for each of the pairwise comparisons? Here they are.