Ready for some Madness?
March is nearly upon us and, after a two-year absence (and an eight year absence in non-COVID years), Boston College men’s hockey is a lock to make the NCAA tournament.
This means for some of our new readers, they will be learning about the process of setting the NCAA hockey bracket for the first time. Let’s dive on in!
College hockey bracketology is fun, because unlike in basketball and other sports where everything is based on complete speculation and subjective arguments over whether X team locked themselves in and whether Y team deserves it over Z, there is a fair, objective ranking system that determines which teams make the NCAA tournament. It judges all teams the same whether they are a blue blood or a smaller program. (Click here for a mathematical explainer.)
While the ranking system isn’t perfect and can always be tweaked, it has two huge benefits:
- Fairness, of course – everyone knows exactly which factors matter in the rankings – winning games, especially against top teams; on the road; and in regulation – and can set their schedules and their tactics accordingly.
- When we get down to the last week(s) of the season, we can have true “Game 162” style moments where teams know for sure whether they have played themselves in to or out of the tournament. In 2019, Minnesota-Duluth needed a series of results to go their way to eke in to the NCAA tournament by .001 point. They all happened, and Duluth got in, going on to win the title. In 2022, Northeastern seemed like they might be safely in even after getting knocked out of the Hockey East tournament, but local rival Harvard springing an upset to win the ECAC title knocked them out. The stakes of these moments were clear, making it much more exciting.
Let’s take a dive in to what the tournament would look like if the season ended today, and in so doing, give an explainer on how the field is set.
First, we set the 16 team field – 6 autobids from conference champions, and 10 at-large teams based on the Pairwise rankings.
The 6 autobids will be set by the conference tournaments, but for sake of this exercise let’s assume the #1 seed in each league wins the conference tournament.
Below are the 16 teams that would make the tournament today, along with their odds of making the tournament according to CHN’s Pairwise Matrix.
#1 BC – Hockey East autobid [100%]
#2 BU – Hockey East [100%]
#3 North Dakota – NCHC autobid [100%]
#4 Wisconsin – Big Ten [100%]
#5 Michigan State – Big Ten autobid [100%]
#6 Denver – NCHC [100%]
#7 Quinnipiac- ECAC autobid [99.9%]
#8 Minnesota – Big Ten [99%]
#9 Maine – Hockey East [99%]
#10 Providence – Hockey East [70%]
#11 St Cloud State – NCHC [69%]
#12 Western Michigan – NCHC [73%]
#13 UMass – Hockey East [78%]
#14 Michigan – Big Ten [65%]
#22 RIT – Atlantic Hockey autobid [40%]
#35 Bemidji State – CCHA autobid [24%]
Looking at the race to get in to the NCAA tournament, here are the teams still semi- realistically in the mix:
#15 Colorado College [51% chance of qualifying]
#16 Cornell [49%]
#17 UNH [16%]
#18 Omaha [10%]
#19 Northeastern [2%]
#21 Notre Dame [3%]
And of course, every team in a conference still has a shot to win its conference tournament.
There aren’t many teams in the race for an at-large bid, but the absolute scramble from #10 on back is huge, as the cutoff for the NCAA tournament will fluctuate based on how many conference champions come from outside the top 16. It is a mathematical certainty that the Atlantic Hockey and CCHA winners will come from outside the top 16, meaning a top 14 spot will be the bare minimum needed to qualify.
From there, upsets in other conference tournaments can move the cutline up by as much as 4 places. In particular, the ECAC is vulnerable to this as Quinnipiac is the only team above the line right now. If Cornell or Colgate were to win the ECAC tournament (which is very possible!), it would move the cutline up to at least 13.
Bottom line – for the teams in spots 10-14, they can’t feel like they’re comfortably locked in right now – they need to climb up the rankings.
None of this impacts BC, as the Eagles have essentially clinched a top-4 spot in the tournament. The question is if they finish #1 overall, or if BU and/or North Dakota can pass them.
Now we get to the fun part, setting the bracket.
First, we assign each 1-seed to its closest regional
#1 Boston College: Providence regional
#2 Boston University: Springfield, MA regional
#3 North Dakota: Sioux Falls, SD regional
#4 Michigan State: Maryland Heights, MO regional
Set up the 16 team bracket based on bracket integrity (1 vs 16, 2 vs 15) etc, before making needed adjustments
This would give us:
Providence
#1 BC
#35 Bemidji St
#8 Minnesota
#9 Maine
Maryland Heights
#4 Wisconsin
#13 UMass
#5 Michigan State
#12 Western Michigan
Springfield
#2 BU
#22 RIT
#7 Quinnipiac
#10 Providence
Sioux Falls
#3 North Dakota
#14 Michigan
#6 Denver
#11 St Cloud State
Next – any team hosting an NCAA regional that qualifies for the tournament must be assigned to the regional it is hosting.
The hosts are Brown [Providence], Omaha [Sioux Falls], UMass & AIC [Springfield], and Lindenwood [Maryland Heights]. As of now, the only host in the field would be UMass, so they’d be automatically assigned as the 4-seed in the Springfield regional.
That requires us to make some changes.
Providence
#1 BC
#35 Bemidji St
#8 Minnesota
#9 Maine
Maryland Heights
#4 Wisconsin
#14 Michigan
#5 Michigan State
#12 Western Michigan
Springfield
#2 BU
#13 UMass
#7 Quinnipiac
#10 Providence
Sioux Falls
#3 North Dakota
#22 RIT
#6 Denver
#11 St Cloud State
Next: Unless it is completely unavoidable, avoid in-conference matchups in the first round.
This is where whoever finishes lower between BC and BU is in position to potentially get screwed. With UMass in the tournament as a 4-seed, BU as a 1-seed cannot face them in the first round.
As a result, BU would get shipped out west – in this instance to Maryland Heights (in order to preserve North Dakota playing in Sioux Falls).
Our initial bracket also has Big Ten foes Wisco and Michigan facing off in the first round, so we’ll have to modify that. Sioux Falls’ bracket also has Denver taking on St. Cloud State in the first round.
Making needed swaps for in-conference matchups while maintaining bracket integrity as best as possible is an inexact science and involves some committee discretion. It would likely look like this:
Providence
#1 BC
#35 Bemidji St
#8 Minnesota
#9 Maine
Maryland Heights
#2 BU
#22 RIT
#7 Quinnipiac
#11 St Cloud State
Springfield
#4 Wisconsin
#13 UMass
#5 Michigan State
#12 Western Michigan
Sioux Falls
#3 North Dakota
#14 Michigan
#6 Denver
#10 Providence
Next: Adjustments can be made within seeding bands [i.e., swapping a 2 seed for a 2 seed, such as #5 for #6] to improve attendance. However, you cannot move teams into a different seeding band [i.e., swapping #10 for #8, a 3 seed for a 2 seed].
The committee has a lot of work to do here on this one and this is not one of the instances where it’s fairly straightforward.
Sioux Falls will sell out as long as it has NoDak. All good here. Maryland Heights is, unfortunately, likely to be an attendance disaster no matter who is in it, so you would hope that they have some teams with big fanbases; Michigan fits the bill here.
Then you get to the Eastern regionals.
In an ideal world, they’d want to put Providence in Providence. But it’s not mandatory, as PC isn’t the host, and the existing Providence bracket has two big drawing Eastern teams (BC + Maine) and also has perfect bracket integrity. I think the likely result would be for them to leave it alone.
But…
Springfield has UMass but could use another Eastern team or two for attendance. The likeliest swap here would be to move Providence in to this regional, giving you #5 Michigan State vs #10 Providence. But that would send Western Michigan to Sioux FAlls against Denver – an in-conference matchup.
So the committee could either leave it alone and count on UMass to carry Springfield attendance, and BC and Maine to carry Providence.. Or make a convoluted series of swaps to get Providence into the Providence regional.
Based on recent precedent, I would expect the committee to come as close to full bracket integrity as possible.
As such, I think they leave this alone:
Providence
#1 BC
#35 Bemidji St
#8 Minnesota
#9 Maine
Maryland Heights
#2 BU
#22 RIT
#7 Quinnipiac
#11 St Cloud State
Springfield
#4 Wisconsin
#13 UMass
#5 Michigan State
#12 Western Michigan
Sioux Falls
#3 North Dakota
#14 Michigan
#6 Denver
#10 Providence
But it wouldn’t shock me to see them swap Maine for Providence here, and potentially Quinnipiac and Michigan State to get QU into the Springfield bracket.
I’m guessing the committee is hoping UMass either moves in to a 3 seed, as that would solve a lot of problems for them.
We’ll see what it looks like next week!