I can’t even begin to count the number of times I was asked by D-III hockey fans over the weekend, “Wouldn’t you like to be a fly on the wall during the conference calls?”
Since none of us outside the committees was privy to the conversations that took place during the selection of the three at-large teams or the seeding, let’s do a little forensic investigation and see if we can piece together what happened. A little CSI: NCAA, or Dragnet, with names changed to protect the innocent.
The first step in the selection of teams was easy. Six teams in conferences with automatic qualifiers won their tournaments: St. Norbert in the NCHA, St. John’s in the MIAC, SUNYAC champ Oswego, Norwich in the ECAC East, ECAC Northeast winner Wentworth, and the lone upset, Trinity in the NESCAC.
With six teams in place, that left three for the committee to choose: the single Pool B bid, for a team from the two leagues without autobids — the MCHA and ECAC West — and the two Pool C bids, for teams in the six autobid conferences that didn’t win their league titles.
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, strength of schedule as determined by in-region opponents’ winning percentage, and results against teams already in the tournament.
It appears that the criteria were followed explicitly by the committee.
(In my predictions before the bids were announced, I said, “… if the selection committee follows the NCAA championship manual to the letter, the decisions should be straightforward.” I was correct in eight of nine picks; I got one wrong because I didn’t follow them to the letter.)
Pool B was a no-brainer; it came down to Elmira and RIT in the ECAC West. While Marian was the top eligible team from the West, it probably wasn’t offered up.
Both RIT and Elmira had 17-4-2 records within region, and the teams went 1-1-1 head-to-head during the season. RIT won the comparison in results against common opponents, 14-3-0 to 13-3-1, while Elmira had a higher strength of schedule. Elmira’s better record against teams already in the tournament, 1-3-0 vs. 0-2-1, gave it the nod.
That leaves us with the most controversial picks, the Pool C at-large bids, which were expanded to two last season.
Sometime around 6 p.m. Eastern time on Sunday, each regional committee, comprised of coaches and athletic directors, got on the horn to discuss what teams should be considered from that region for the at-large bids.
In the East, it was probably a pretty easy discussion. Middlebury, as the lone top seed to be upset, and Plattsburgh, the next-highest eligible team, were probably the top two considered, although a case could be made for New England College, Colby, or Bowdoin. However, ranking the teams based on the number of comparisons won, in a method similar to USCHO’s PairWise Rankings (see sidebar), Middlebury and Plattsburgh are the top two.
Next the Eastern committee had to rank those two teams. Middlebury and Plattsburgh split the five criteria, with the Panthers taking winning percentage and record against teams in the tournament. Plattsburgh had a better record against common opponents and a higher strength of schedule, and the two teams split head to head.
While all five are supposed to be equal, it appears from our review of this pairing — and another one we’ll look at during the seeding discussion — that record against teams already in the tournament, is “more equal than others,” and Middlebury got the nod over Plattsburgh. The championship manual does suggest that the committee can do this: “A team may have an excellent winning percentage; however, the strength of schedule and results against teams in the tournament must also be examined.”
In the west, the top two teams to be considered were defending national champion Wisconsin-Superior and a team that lost both the NCHA semifinal and consolation, Wisconsin-River Falls. Superior went 2-0-1 against River Falls this season, including beating the Falcons in the NCHA semi, so it’s likely most fans, and perhaps even the Western committee, expected Superior to be the top team from the West looking for that Pool C bid.
As Chris Lerch noted in his Q & A about the tournament selections, River Falls won three of the five criteria against Superior: winning percentage (by a half-game), record against common opponents, and record against teams already in the tournament. The Yellowjackets were stung by a dismal 0-4-1 season against St. Norbert; their only win against a team in the tournament was over St. John’s.
The Western committee, possibly after some head-scratching or even head-shaking, submitted Wisconsin-River Falls and Wisconsin-Superior to the selection committee.
The selection committee then reviewed the two Pool C bids from each region. They probably first compared the top two; since Middlebury wins the comparison with River Falls in three criteria: winning percentage, record against teams in the tournament, and strength of schedule, the Panthers got the first Pool C bid.
Next, they likely compared River Falls and Plattsburgh, the top remaining candidates. River Falls topped Plattsburgh in winning percentage and record against teams already in the tournament, while Plattsburgh had a higher strength of schedule. River Falls gets in.
And we have our field of nine teams.
Next, the committee was charged with seeding the teams. A page of comparisons won, like the updated Pairwise Rankings linked in the sidebar at the right, generates this list:
1 St. Norbert
11 St. John’s
13 River Falls
If that were the case, the seeds would have been: 1W) St. Norbert, 2W) St. John’s, 3W) River Falls; and 1E) Norwich, 2E) Oswego, 3E) Elmira, 4E) Middlebury, 5E) Trinity, and 6E) Wentworth.
Since they weren’t, it’s clear that again the committee used head-to-head comparisons among the five criteria to decide.
In the West, St. Norbert was clearly the top team. But how did River Falls get a higher seed than St. John’s even though the Johnnies had beaten the Falcons during the regular season? Again, it came down to the criteria. River Falls had a higher winning percentage, won two more games against common opponents, and had a better record against teams already in the tournament.
In the East, the seedings are similar to those rankings, but the positions of the top two and second two are swapped. Let’s see if we can figure out why.
Comparing Norwich and Oswego, each wins two comparisons: Norwich had a higher winning percentage, and led in common opponents 6-1-0 (.857) to 10-2-1 (.808), while Oswego had a higher strength of schedule and a better record against teams in the tournament, 3-0-0 vs. 1-1-0. Apparently, record against teams in the tournament was again given a little more weight. Oswego gets the top seed over Norwich.
Looking at Middlebury and Elmira, the Panthers win three of five criteria over the Soaring Eagles: winning percentage, and records against common opponents and teams in the tournament. Middlebury gets the higher seed.
By 9 p.m. or so, the committee had finished its work, and no marathon was necessary. The selection committee followed the championship manual to the letter.