NCAA D3: We Have a Bracket!
The college hockey season is winding down. The conference tournaments have all crowned their champions. River Falls, WI, awaits at the end of the month. And the D3 bracket has been revealed! Here's a look at how it got constructed.

The college hockey season is winding down. The conference tournaments have all crowned their champions. River Falls, WI, awaits at the end of the month. And the bracket has been revealed!
As a reminder, the D3 bracket has 12 berths this year, as mandated by the D3 access ratio of one berth for every six teams playing. Of those, eight are automatic qualifiers (AQs); the other four are the top teams in the NPI that did not get an AQ, either because they lost in their conference tournament or their conference is ineligible for one. With a 12-team bracket, four teams will get first-round byes.
Constructing the Bracket
Here’s a table showing the top 19 teams in the final NPI, published by the NCAA on 9 March. (Why 19 teams? There are 44 D1 and D2 teams playing for the women’s NC championship. Their bracket has 11 berths, one for every four teams. If D3 used that same ratio, we’d have a 19-team bracket and all the teams listed below would be in. Wouldn’t that be nice?) The AQ teams are indicated in boldface. The at-large selections are indicated in italics. Schools left out have a shaded background. And the rightmost column shows the school’s location, which will become important in bracket construction.
Based on this, we can seed the 12 teams based on their NPI:
On page 15 of the prechampionship manual, the bracketing principles are laid out. One welcome change for this year is the explicit statement that the top priority is “protection of the top four teams in the bracket,” amplified further in the directive that “once selected, teams 1–4 will receive the bye and host the quarterfinal round.” In the past, geographic considerations sometimes meant that a team with a high seed would be forced to play a first-round game instead of getting a bye, and it’s nice to know that won’t happen. Augsburg, Amherst, Middlebury, and UW-River Falls get the byes this year.
The other three criteria prioritized for site selection are facility quality, geography, and attendance history, in that order. Geography, for D3, means that the bracket should be set up to minimize flights. A flight is permitted if teams are more than 500 miles apart according to the official NCAA mileage calculator. Bracket integrity is not included in this list of criteria, although there is a note that teams in the same conference will not be paired in a first round game.
If we make a bracket based only on seedings, we get this (home teams in boldface):
There are problems with this bracket: two of the first round matchups would require flights. Colby and St. Norbert are 1364 miles apart, and Gustavus and Wilkes are separated by 1129 miles. And as many as three more flights might be required for the quarterfinal round, depending on the first round results! However, there’s an easy and minor fix: switch the 7 and 8 seeds to send Wilkes to Colby (484 miles, phew!) and St. Norbert to Gustavus (346 miles). Both Gustavus and Colby would maintain the right to host a first-round game. Colby would arguably have a slightly easier matchup and Gustavus a slightly tougher one than the seeds would dictate, with potential repercussions into the quarterfinal round, but them’s the breaks in D3 sports. That gives a modified bracket:
There’s still a minor issue with the quarterfinal matchups. Given the top four seeds, there’s going to be no way to avoid having one flight in this round, as two of the quarterfinalists are in the west but only one first round winner will be a western team. Since all the other quarterfinal matchups as set by the seeds are well within the 500-mile limit, there’s no reason not to give the bracket as much integrity as possible and fly Plattsburgh or Elmira to River Falls for their quarterfinal. (And hey, should either pull the upset win, they might not even need to fly home between the quarterfinals and the semifinals, as River Falls will host the final weekend!) This is the bracket the NCAA went with, one that is about as straightforward as D3 gets and one that most should be happy with.
Any surprises? Any snubs?
I did not expect to see Middlebury slide down the NPI ratings, from the initial release (NPI of 64.867, #1) to the final release (NPI of 61.830, a loss of 3.037 points, #3). Middlebury ended the regular season by going 1-3, winning at home vs. Williams and suffering the three losses on the road, one at Williams and the last two at Colby. In the NESCAC tournament, they avenged their end-of-season losses by beating Williams in OT in their quarterfinal game at home and winning over Colby in their neutral site semifinal, before losing to host Amherst in the conference final. In the end, their slide didn’t even cost them a bye in the NCAA tournament. However, it’ll be interesting to see if they can solidify their play for a deep tournament run.
Connected to Middlebury’s slide is Colby’s ascent. When the second rankings were released, before the last weekend of the regular season, Colby was at #11 with an NPI of 58.895 and below three SUNYAC teams, not even on the bubble. But then Colby ended the regular season at home with two shutout wins over Middlebury. This weekend caused their NPI to shoot up to 60.358 in the release on 24 February, a rise of nearly 1.5 points. More importantly, it sent them up the rankings to #7, above two out of the three SUNYAC teams they'd trailed. And in the final NPI release, Colby was still ahead of two SUNYAC teams. The NPI system seems to love the NESCAC, possibly due to the conference's overall high strength of schedule. But Colby, the last team in, also went 4-1 against teams with an NPI higher than theirs, showing the ability to beat good teams.
And then, the sad story of Oswego. A month ago, Oswego was in great position for a berth, just behind Plattsburgh in the initial NPI but ahead of them in the SUNYAC standings. They ended their regular season by sweeping Morrisville at home and then tying Cortland at home before beating them on the road. They won their semifinal matchup against Canton before falling at home in the championship game to Plattsburgh. In the end, that loss doomed the team to staying home, the first team out. On 7 March, the day before the SUNYAC championship game, Oswego had an NPI of 60.430, good enough for #6 and higher than Colby's eventual final NPI. After the game (and the two NESCAC semifinals and all the other conference championship games), Oswego’s NPI had fallen by 0.752 to 59.678, and down to #9 in the rankings. Between Oswego’s weaker schedule and only going 1-3-1 against teams with higher rankings, they ended the season with an NPI 0.096 lower than Colby’s, the difference between being in and being out.
The take-home lessons I see from all of this for teams with postseason aspirations:
- Construct your nonconference schedule to include strong teams. Strength of schedule is of utmost importance to NPI, which will keep you in the running for an at-large bid should the need arise. You can’t control the strength of your conference opponents, but by seeking strong opponents when you can, you’ll strengthen your entire conference.
- Win the big games. Big games are those against teams good enough to give you a quality win bonus, and especially the conference final that will clinch an AQ.
- Win at home. Home losses hurt more than road losses under the NPI formula, especially when you lose to a strong team.
- One game can raise or lower your NPI by about 0.75. That can mean the difference between playing on or going home.
Final Notes
Congratulations to Plymouth State, the 2025 MASCAC champion! Juliet Rutigliano scored the game-winning goal 6:55 into the fourth overtime period, giving Plymouth State the 2-1 win over Rivier. This was the second-longest D3 women’s game ever, just shy of the record set a week earlier when Augsburg beat St. Mary’s at 8:06 of the fourth overtime. The MASCAC does not have an AQ this year, for which Colby should be grateful. Plymouth State, the highest ranked MASCAC school, was only at #32 in the final NPI rankings, so there will be no MASCAC representation in the women’s tournament this year. But Plymouth State is the first MASCAC champion and should raise a banner in the fall.
And now, a sad note: Northland College, which sponsors both men’s and women’s D3 hockey, is closing at the end of the spring semester. The women’s team was at the very bottom of the final NPI rankings. Their last win came on 6 January 2024, in OT against Aurora. The men’s team wasn’t much better, at #86 out of 87 teams. It’s always sad to see a school close, especially one with hockey teams. But if there’s a bright side to this news, it’s that there won’t be any overall loss of opportunities for students who want to play D3 hockey. Beloit College announced shortly afterward that they will add both men’s and women’s hockey and begin play in the fall. Beloit has already hired both coaching staffs from Northland and is working with students from Northland who wish to transfer to complete their undergraduate degrees. Expect Beloit to compete as an associate member of the WIAC, as Northland has.
There are still three more weekends of D3 hockey! All first round and quarterfinal games are scheduled for 3 PM local time on Saturday. These games should all have streams available from the host schools. For the semis and final, the NCAA takes over the broadcast streams. Semifinal games are Friday 28 March at 4 PM and 8 PM EDT, and the final is Sunday 30 March at 4 PM.
Comments ()