CSI: NCAA, the 2005 Edition

The last weekend of the 2004-05 men’s D-III hockey season was filled with minor and major upsets. Only two No. 1 seeds won conference championships in the six leagues with automatic bids: St. Norbert and Curry; three second seeds won: Geneseo, Middlebury, and St. Thomas; and No. 4 New England College had two upset wins to take the ECAC East.

It doesn’t seem like those threw much of a monkey wrench into the selection process. Only a couple hours after the end of the last game, the NCAA men’s ice hockey committee had chosen the field, the fastest selection in recent memory. And while there’s not much doubt over the teams selected, the seeding does need some exploration.

So, for our third consecutive year, let’s do a little forensic investigation — what we’ve come to call CSI: NCAA. Since none of us were in on the committee’s conference call, let’s see if we can piece together from forensic evidence what could have happened to choose the nine teams and decide the seeding.

As always, the first step in the selection of teams was easy for the committee. Six teams in conferences with automatic qualifiers won their tournaments and received automatic bids: the six teams mentioned above. That group constitutes Pool A.

With those six teams in place, three at-large bids were left for the committee to select: the single Pool B bid, for a team from the two leagues without autobids — the MCHA and ECAC West — and two Pool C bids, for teams that had not yet received a bid. This year, Pool B teams were included in Pool C.

The selection committee has five criteria on which to rank at-large teams: in-region winning percentage, in-region head-to-head results, in-region results against common opponents, the Strength of Schedule Index (SOS), and results against ranked teams. The NCAA has been publishing a regional ranking of teams since February 8, using those criteria.

As was the case in 2004, it appears that the criteria were followed explicitly by the committee in the selection of teams.

First, let’s look at the Pool B bid. Because Milwaukee School of Engineering of the MCHA was not ranked, we can surmise that the Raiders were not really considered for the Pool B bid, and that the comparison was made among the four ranked teams in the ECAC West: Manhattanville, Utica, RIT, and Hobart. RIT and Hobart were ranked way below Manhattanville and Utica, so we’ll just look at those two teams.

Utica won both regular season meetings, but lost the ECAC West championship game to Manhattanville, 3-1. In the five primary criteria, Utica wins only the head-to-head comparison; the Valiants prevail in all four other criteria. Manhattanville gets Pool B. (Had Utica won, the comparison would have been tougher, but the committee dodged a bullet, and got to end their call a lot earlier than probably would have been the case.)

The top two teams in the east not receiving automatic qualifiers were Trinity and Oswego; the Lakers got a boost because of losses by Babson and Bowdoin. The top two teams in the west were St. John’s and Wisconsin-Superior.

The committee then had to compare those teams to decide which would get the first Pool C bid.

First, let’s look at the top teams left in each region, St. John’s from the MIAC, and Trinity from the NESCAC.

A comparison of St. John’s and Trinity shows the Johnnies winning three of five criteria, with the two teams having no common opponents or head-to-head play, since they are in separate regions. St. John’s winning percentage was .880 to .826; the Johnnies had the edge in SOS Index, 10.2400 to 10.0000; and St. John’s record against ranked teams was 5-2-2 vs. 3-2-2 for the Bantams.

St. John’s gets the first Pool C bid.

Next, let’s compare Trinity with Wis.-Superior. Trinity wins all three of the above comparisons against the Yellowjackets: Superior’s winning percentage was 0.776, the SOS Index was 9.5517, and its record against ranked teams, 3-2-5.

The second Pool C bid goes to Trinity.

(You can see this information by viewing USCHO’s PairWise Rankings; a link to that information is in a sidebar near the top of this article.)

So without much handwringing, we have our field of nine teams.

Next, the committee was charged with seeding the teams. A page of comparisons won, like USCHO’s PairWise Rankings, generates this list:

1. St. Norbert
2. St. John’s
3. Manhattanville
4. Trinity
5. Middlebury
10. St. Thomas
14. Geneseo
18. Curry
21. New England College

(The numbers are those of the USCHO Men’s D-III National PairWise Rankings, not the NCAA’s.)

Using the rankings above, the committee first split the field. With a 6-3 split between the east and west regions, the first-round game is held in the west region, to avoid flying teams in the first round.

While this is the most controversial part of the seeding process, the committee really doesn’t have any leeway on this issue. They are told they are to rank the teams within each geographical region and to keep flights to a minimum. Here’s a section from the 2005 NCAA Men’s D-III Ice Hockey Championship Manual:

The Division III Men’s Ice Hockey Committee will rank the teams selected within each geographical region using the primary and secondary criteria. The highest-ranked teams in each region will be given consideration as first- and second-round sites. … The committee will pair the teams regionally, based on geographical location of all participants and final seeding. Flights will be kept to a minimum.

In the west, despite being the second seed in the west, and, in USCHO’s PWR, the second-ranked team overall, St. John’s will play in the first round against the team that beat it for the MIAC championship, St. Thomas. The winner of that game will face St. Norbert in the second round.

In the east, rankings would suggest this order of seeding:

3. Manhattanville
4. Trinity
5. Middlebury
14. Geneseo
18. Curry
21. New England College

(Again, the numbers are those of the USCHO National PWR, not the NCAA’s.)

The first pairing, New England College at Manhattanville in a second-round matchup, is what was announced.

But the other two suggested by this ranking, Curry at Trinity and Geneseo at Middlebury, were not the bracket picked by the committee, which has USCHO’s No. 2 East Trinity at our No. 4 East Geneseo, and our No. 5 East Curry at our No. 3 East Middlebury.

The committee actually seeded the teams this way:

1E. Manhattanville
2E. Middlebury
3E. Trinity
4E. Geneseo
5E. Curry
6E. New England College

Why did they do this, and why did they send Trinity to Geneseo?

Sometimes you need more information than just physical clues. There is conflicting information from sources close to the NCAA and both schools on whether Trinity or Curry had submitted a bid to host, but it is clear that neither had their bid accepted if they did. So the Trinity-Curry match that USCHO’s PWR suggests should have happened would have had no rink on which to be played.

Since the committee is allowed to group and seed teams taking geography into account, they switched Middlebury and Trinity in the rankings to avoid a match between two teams that couldn’t host. That sent Curry to Middlebury, and paired Trinity and Geneseo. Because the higher seed, Trinity, is not able to host, the Bantams travel to Geneseo for their second-round game.

The other pairings among those four, Trinity at Middlebury and Curry at Geneseo, would have paired seeds 2E and 3E and seeds 4E and 5E, so that option was tossed.

While this also may be a controversial decision, it is by the letter of the championship manual:

Once automatic qualifiers are identified and the Pools B and C teams are selected, the following guidelines should be followed:

  • Once selected, teams will be grouped in clusters according to natural geographic proximity. Teams will then be paired according to geographic proximity. A team may be moved to numerically balance the bracket, if geographic proximity is maintained. Teams should be paired and eligible sites should be selected according to geographic proximity (within 500 miles).
  • Teams may be seeded on a regional basis using the regional selection criteria. However, geographic proximity takes precedence over seeding.
  • Teams from the same conference do not have to play one another in the first round, as long as geographic proximity is maintained.
  • The highest-seeded team that meets all site selection criteria will be selected as the host institution, provided geographic proximity is maintained.

Our third annual investigation of this year’s NCAA selections comes to the same conclusion as the previous two years: the committee followed its process to the letter. Once again, case closed.