Time for a little bracketology postmortem.
So the good news after the 2021 NCAA Women's Hockey Tournament bracket was announced Sunday night? The bones of the bracket made perfect sense. Northeastern earned the No. 1 overall seed and play Robert Morris. Wisconsin, as the WCHA champion, was the No. 2 overall seed. The top-six spots were correct with the ECAC champion, the second and third WCHA teams, and the second Hockey East team, filling the ranks.
The biggest question - giving the final at-large bid to either the third Hockey East team, the fourth WCHA team, or a second CHA team - ended up being a third Hockey East team in Providence. Doing so took away any need to avoid inter-conference matchups, pushing second-ranked Hockey East team Boston College into a matchup with second-ranked WCHA team Ohio State.
The bad? I, along with everyone else outside of Amsoil Arena and/or the selection committee (including the UMD beat writer), did not have Minnesota Duluth ranked as the third WCHA team.
That bit of change flipped the bracket upside down and took out a team everyone besides the committee saw as a lock. Consensus aside, the only way Minnesota would miss was a combination of two separate, questionable subjective decisions. To see a Gopher bubble burst, it would need both Minnesota Duluth being considered the third-ranked WCHA team and the WCHA only getting three teams in the NCAA Tournament in favor of Hockey East or the CHA. Surprisingly this happened, creating a bracket without Brad Frost's team for the first time since 2007.
Take away one of the choices and the bracket holds up better. I had the third-ranked WCHA school, no matter who, as the top unseeded team. Providence's spot was going to be a political question in a year where non-conference games did not exist. I thought one decision could happen but, like everyone else, never expected the committee to make both.
Honestly, I expected Penn State to be in the NCAA Tournament as an at-large bid before Minnesota's bubble was burst.
Why did the consensus have Minnesota as the third-ranked WCHA team?
Minnesota was the consensus pick because looking at the selection criteria data, the Gophers (11-8-1) were ahead of Minnesota Duluth (11-6-0) outside of the percentage in a win/loss record. Pairwise does not work across conferences this season without inter-conference play. It does work inside where schedules are unbalanced yet similar. When comparing only WCHA teams against one another, Minnesota had a higher Pairwise, RPI, and KRACH compared to UMD. Both teams lost in the WCHA Final Faceoff semifinals.
More importantly, the Gophers swept Minnesota Duluth in Duluth in the only two games the teams played. Head-to-head results were very applicable there. Both teams also had the same number of wins against Wisconsin and Ohio State. Minnesota went undefeated (9-0-0) against everyone outside of those two teams while Minnesota Duluth (9-1-0) lost to a two-win Bemidji State squad that was one of the worst teams in college hockey.
The Bulldogs are also two weeks removed from needing a third-period comeback one day and a last-second goal the next to sweep Minnesota State, however, that is not part of any selection criteria. A 1-0 win counts the same as a 10-0 win.
How did the selection committee justify Minnesota Duluth?
Note that without getting an answer from the committee, who have not given one publicly, this, like the selection criteria, is completely subjective.
The criteria used are vague enough to justify anything for a year where each team faces unique challenges.
In this case, it seems comparisons were puts aside, and win/loss record and Minnesota Duluth finishing higher in the WCHA thanks to playing fewer games give the Bulldogs the edge. UMD played three fewer games than Minnesota due to Covid canceling two separate series - including a January 4-5 one against Minnesota at Ridder Arena and six fewer games against Wisconsin and Ohio State, going 2-3-0. Using a straight win/loss and points percentage helps Minnesota Duluth and punishes Minnesota for playing more games against a tougher strength of schedule (also a criterium).
While the human element could come into play when deciding teams within conferences, there are no "last 10 games" or "how did Team X play down the stretch?" criterium. Minnesota finishing 3-7-1 after 9 of 11 games against Ohio State and Wisconsin, and Minnesota Duluth finishing 6-2-0 should not make a difference. All games are supposed to count the same, according to the criteria.
How did the selection committee justify Providence?
Again, this is my own belief. The selection committee justified Providence reaching the Hockey East championship game, a similar resume with a higher finish playing in a different conference, and creating an original matchup.
Many of the same things that could be said about Providence can be said about Minnesota, however, this separate decision makes more sense on the surface in a year without inter-conference matchups.
Providence, like Minnesota, struggled against the top team in its conference, going 0-4-0. The Friars did take 2 of 3 games against second-place Boston College and beat/tied fourth-place UConn. (PC also lost to fifth and sixth-place Maine and New Hampshire, respectively.) The 2-5 record against NCAA teams is similar to Minnesota's 4-7-1 record - if UMD is considered the fifth seed, as they are in this bracket - then the Gophers' two wins against the fifth seed should count as such.
The Friars, through no fault of their own, did not get another chance to face Boston College since Maine upset the Eagles in the Hockey East quarterfinals. Needing OT, PC advanced to the championship, unlike Minnesota, who lost to Wisconsin in the semifinal.
Lastly, this might be the most subjective decision: If Minnesota is the fourth WCHA team and the choice is between the Gophers, who went 0-4-1 against Wisconsin and just lost Saturday, and Providence, who have not faced the Badgers, would the committee be more swayed into choosing the fresh matchup for Wisconsin?
What about Penn State, who went 16-3-2?
Sadly, it seems like CHA regular-season records and points matter like those on "Whose Line Is It Anyway?" Like the 2021 NCAA Tournament, one weekend in Erie, PA takes precedence over months of work.
What can we learn from this for future years?
This year's selection is unique based on Pairwise being unusable across the sport. So it should be a unique year without bracket precedent for the future. But if there is a true winner this weekend, it's the use of Pairwise, as flawed as it can be at times. A flawed math system tops flawed people.
In fairness to the committee, I don't think a single bracket exists which would garner 100% agreement. Someone would be left out. Someone's feelings would be hurt. No result exists that would not bring up fears of mistrust and backroom deals. This is college hockey. Look at how long it has taken to get a "consensus" on settling overtime.
Still, there was a break when it came to Minnesota where the consensus and selection committee do not line up with one another. Whether there was less chance of getting four WCHA teams than believed (very possible) or a lack of understanding of what criteria were more important than to the committee (extremely possible), the two decisions that left the Gophers out should not be a surprise.
Two things to take away for this year that could be implemented for the future men's hockey selection are, first, that the polls still don't matter. Minnesota was ahead of UMD and Providence all season. They still are on the Monday after Selection Sunday. Second, there will be more of an effort to balance the field nationally and get fresh, non-conference matchups. That extra team in each conference should be seeing what it can do to add to its resume.
Unfortunately, there is a negative takeaway for women's hockey teams to take from these decisions. Using the Pairwise has come at a cost where limited non-conference games make it better to win against bad teams than lose against good teams. This cost the third WCHA team behind Wisconsin and Minnesota for several years in the 2010s and multiple Hockey East teams.
One example that comes to mind is Boston University needing to win the 2015 Hockey East tournament because the Terriers, with Marie-Philip Poulin, played and lost to Minnesota in a non-conference game that flipped an entire conference's comparison.
Using subjective reasoning did not change that. Minnesota ended up being punished for being the third-best team in a season where the Gophers, Wisconsin and Ohio State played the majority of the season against one another. Bad losses mattered less than losing in general or head-to-head matchups. Minnesota Duluth and Providence, with slightly better records and slightly weaker strength of schedules, were deemed higher-ranking than Minnesota.
If the lesson is to not play good teams - something that likely also cost the ECAC a second bid this season with all four teams beating up on one another - that is a worrying trend. The top teams facing one another can create a self-fulfilling prophecy at times, but it has also been difficult in non-pandemic years for several teams to schedule non-conference games. The Minnesota Cup exists for this reason.
I will be curious to see how this year's selection criteria affect future years, whether it is placing more strengths in certain aspects, rewarding teams and conferences more for non-conference games, or adding criteria like the final 10 games.
The Gophers being left out of the 2021 NCAA Women's Hockey Tournament comes down to two separate decisions. The committee went against the consensus to take UMD as the third WCHA team and did not take a fourth WCHA team. It's easier to understand the latter decision than realizing it turns out the former was a contest.
At the end of the day, in a season defined by unrest and the pandemic, it is fitting the most important games might have been two not played between Minnesota and Minnesota Duluth.
