Around the Nation

NCAA tournament selections explained

By Matthew Webb
Senior Writer,

Well, the men's and women's tournament brackets have been released which means the annual stream of questions and complaints have been pouring in for about eleven hours now.

Trinity was again a topic of discussion as a result of Monday's selections as the Bantams surprisingly vaulted up the East Region rankings. It turns out that rise was justified.

In fact, it's been awhile since we've seen this much howling about the tournament selections -- though in this instance it has little to do with the actual teams selected for either tournament, but rather it's the arrangement of the men's bracket that has been dropped directly into the crosshairs.

Which is foolish. Because the committee got it right. Again.

We'll get around to explaining why a bit later, and we'll also touch on some other things people may be wondering about, namely the women's bracket, the final men's regional rankings, the men's Pool B and C selections, and of course, after being able to take a year off we must once again explain why our Men's Bracketology was wrong. After all, we know exactly why it was wrong...and we're not happy about it.

So sit back as we commence our annual exercise aimed at answering any and all questions that we've seen arise since this morning's NCAA tournament brackets were revealed. As always, it's going to be a fun ride.


Once again we lead off with the women as the bracket offered little to nothing to complain about. In fact, if anyone actually is complaining...we aren't aware of it as we haven't noted a single complaint all day.

Our Women's Bracketology was once again spot-on as it nailed the bracket perfectly, so kudos to Ray Biggs for that, though despite this year's selection being so straightforward we'll touch on a few things in case there are some newcomers to this process.

  • Why is Adrian playing Elmira instead of a West Region team? Easy, despite Adrian being in the West Region, Elmira is the only other team in the tournament that is within the NCAA's allowable 500-mile travel limit for the first round. That means the Bulldogs and Soaring Eagles were destined to again meet in a quarterfinal, and it will again be played in Adrian. Why is this? Because the Bulldogs are ranked higher in the NCAA's final regional rankings.
  • Why is Middlebury on Plattsburgh State's side of the bracket? The way the numbers fell this year makes it look pretty clear that the top four seeds, in order, are Plattsburgh State, Adrian, UW-River Falls and Middlebury. As such, Plattsburgh and Middlebury are lined up on one side of bracket as the No. 1 and No. 4 seeded teams while Adrian and River Falls fall on the opposite side as the No. 2 and No. 3 seeded teams.
  • Why does St. Thomas have to fly to Plattsburgh? As a flight is required no matter what it makes sense to send the lowest seeded team, St. Thomas, to play the highest-seeded team, Plattsburgh. This is actually the exact same thing that happened in 2015. The result that year? A 4-1 Plattsburgh win.
  • Why did Gustavus Adolphus get in as the last team instead of perhaps Endicott? The Gusties simply had the better numbers. A tremendous season for the Gulls, no question, but once Gustavus dropped into the Pool C mix last week it was clear Endicott just wasn't going to win that comparison. A relatively poor SOS and practically non-existent record-against-ranked-teams were the two metrics that sealed Endicott's fate.

Does that about do it? I think so. Yet again the women's committee went straight by the book and spit out a field that makes perfect sense. For that, it and its members should all once again be commended. 



And now onto the men, which unfortunately present a variety of complexities that the women do not. Where shall we start?

Oh, Trinity

The Bantams have had a habit of popping up in this column over the past few years and here they are again. This time it's because the NESCAC champions made a surprising charge up the East Rankings over the past few weeks and surged all the way to No. 2 in the final rankings, which surprised a lot of people. This is even higher than we foresaw, which is saying something as we had good reason to believe Trinity would be on the upward move. Let's compare our projected final East rankings with the real one:

1. Norwich 1. Norwich
2. Hobart
2. Trinity
3. Oswego State
3. Hobart
4. Trinity 4. Oswego State
5. Endicott 5. Endicott
6. Hamilton 6. Hamilton
7. Plattsburgh State
7. Plattsburgh State
8. Utica
8. Utica
9. Williams
9. Williams
10. Amherst
10. Amherst

We won't spend too much time on this as it ends up not being too consequential to the bracket but we will note that the numbers actually back up Trinity's rise to No. 2. You can peruse the final East and West data sheets if you'd like, and if you'll do you'll see that the Bantams boast a very strong RNK (thanks to Williams and Amherst remaining ranked) and also have the best SOS of any team anywhere near it in the rankings.

Surprising at the time? Yes, but a quick look at the data (which was not available yesterday) shows the move made quite a bit of sense. The irony is that Trinity's ascent won it a date with perhaps the hottest team in the nation in Plattsburgh, while Hobart holding at three gets it a home game against a tournament newcomer in Endicott. While time will tell how things unfold many would argue Trinity actually has the tougher draw here.


It's Elementary

For all the carping that has gone on today and will likely continue to go on, very little has to do with the five teams selected for Pool B and C, which is borderline amazing, especially as Utica and UW-Eau Claire were likely the first two teams out and both have come out on the short end of the selection in recent years so fans of both might have reason to be a little on edge. It took us literally 30 seconds to pick the five teams, and we suspect it might not have taken the committee much longer. From our final Bracketology, here's the numbers, and note that RNKs have been adjusted to account for the fact Babson did indeed fall out of the East Rankings:

Pool B

WIN  .7860  .8040
SOS  .5280  .5240
RNK  .3750 (1-2-1)  .6875 (5-2-1)
H2H  -  -
COP  -  -


Pool C

Hobart  .7860  .5280  .3750 (1-2-1)
Oswego State
 .7960  .5200  .5000 (3-3-0)
Hamilton  .7780  .5060  .5000 (2-2-3)
Utica  .6670  .5280  .4000 (2-3-0)
 .7410  .5410  .2143 (1-5-1)
UW-Eau Claire
 .6960  .5340  .0000 (0-6-0)

In both comparisons the winners are clear. Stevens Point's edge in Pool B only got larger when Babson became unranked and in Pool C the top four are obvious. We can't recall an instance where it was this definitive, but it really was which is probably why no one's complaining about it.

But have no fear, as while those were elementary we're slowly working our way to what it is that people are complaining about. I can't wait!


Semifinal Switcheroo?

This is a minor one, but a few people have pointed it out and it's a good observation so we'll give it a mention. It seems quite clear now that if the entire field was seeded the top four are Norwich, St. Norbert, Stevens Point and Trinity, in that order, yet the bracket is set up so that Norwich would face Stevens Point in the semifinals and not Trinity. But why?

Regional ranks/seeds. Both semifinals are set up so a regional No. 1 would meet a regional No. 2, so the committee likely had some flexibility here and opted for the East-West match-ups. As 2W equals 2E in the eyes of the NCAA, this falls completely within the tenets of the process.

It also creates the potential for some extremely interesting match-ups.



And here we go...

"Bracket integrity" is a phrase I've been using for years in Bracketology and it seems to finally have become en vogue among the self-appointed bourgeoisie and its digital cocktail circuit. So you're all welcome for that but if being used it needs to be used properly, and to that end the only rational definition of "bracket integrity" is: "the bracket that maximizes seed integrity while adhering to NCAA travel restrictions."

It is this concept that today's avalanche of complaints focus on. Well, those complaining might not like it but I'm here to tell you: the bracket released this morning has more integrity than any other option that includes an 8-4 split and these 12 specific teams.

Here's why: Adrian being in the fold always creates some unique options due to its complete geography, and this year just two of those were plausible. They were:

  • Putting Adrian/Oswego/Hobart together such as we did in our final projection
  • Going with two West Region quarterfinals such as the national committee did

The second option is far superior and there really isn't even much discussion to be had about it. Or at least rational discussion.

Consider the following graphics. On the left we have the bracket we proposed on Sunday night and on the right we have the real bracket. However, all team names and records have been removed and only regional ranks are shown.


So what's there to complain about? And I'm not asking that flippantly, I'm asking because I'd really like to know what there is to complain about as anyone who can look at the one on the left and conclude it has more "bracket integrity" than the one on the right is, quite simply, wrong. The one on the left almost looks laughable by comparison.

The real bracket is literally perfect by regional rank and we know the NCAA places high emphasis on regional designations and ranks -- it's a key tenet of the entire Division III concept which is why it's a point we've repeatedly tried to hammer home in our Bracketologies.

If anyone wants to try to put a case together, based on something other than a tiresome emotional investment in a specific school, team or region, that the bracket on the left has more "bracket integrity," in whatever haphazard manner you feel like defining it, I'm all ears. Try me, because you don't have a case to make and I'd probably find it rather amusing. But feel free to try me anyway as I'm always looking to be entertained.

Now, while the bracket is seeded perfectly by regional rank it of course isn't completely perfect as Norwich, a regional No. 1, is indeed stuck in the first round. Sadly there's no way around that one due to travel restrictions, but hey, it's not like we haven't seen this before as it happened to St. Norbert in 2010 and 2012 and it's quite likely we'll end up seeing it again at some point. Such is the nature of the beast.

But regarding the bracket itself, it gets as close to perfect as one of these can be for a 12-team field that's handcuffed by the current travel restrictions. It's seeded perfectly by regional rank, the top six ranked teams all get home games, the top four ranked teams in each region are in or can play their way into the quarterfinals, the top two in each region into the semifinals, and we're set up for potental semifinals of 1Ev2W and 1Wv2E.

But somehow, despite all that, people have spent all day and thousands of words braying that they think the one on the left is a superior option? Just look at them. You're wrong. It's not even close. It's so far from close that it shouldn't even be a topic of discussion.

The committee, yet again, did it by the book and did it right and should be commended for adhering exactly to the process. This seems to be becoming a trend, which is a great thing.


Why Oh Why?

Last and probably least, we must sadly return to having to explain why our Bracketology was wrong. But first, an anecdote:

My parents live in Nevada so every Christmas I fly into Vegas and make the 76-mile drive to the city of Mesquite for a 10-day extended stay. Nevada means casinos, and as I may or may not be inclined to place a bet or six from time to time, the opportunity is there. Sometimes the opportunity is so grand that I don't even wait until December to hammer my money in.

Such was the case with the recent UFC fight between Ronda Rousey and Amanda Nunes. I was so certain that Nunes was going to win and the opening line was so good I called my mom in October and asked her go make the bet for me out of fear the odds on Nunes would come down as it was such an obvious bet to make. Surely everyone else must be able to see something that I felt was so obvious.

I then spent over two months telling everyone that I would almost bet my life on the fact Nunes would win, and would probably do so in less than 90 seconds. Easiest bet I ever made. Come the night the December 30 the fight was on and all of 48 seconds later Nunes had caved Rousey's face in and I was a winner.

While happy I won, I sat on the couch staring at my dad and asked, "why on Earth did I not bet five grand on this? What was I doing?" After all, I had been the one carrying on for months that it was the surest thing I'd ever seen.

Atypical for me, I didn't fully execute on my judgement. But so how does this tie into Bracketology and why we had it wrong?

Because I made the same mistake again. I should have known better and I blew it. Ray didn't blow it, other D3hockey staff that doesn't even exist didn't blow it, nope -- I did.

A long-running goal of Bracketology is to illustrate and implement the process. To this end, I've talked a lot over the years about whether the goal of Bracketology is to be predictive or educational, and ultimately settled on the latter for a variety of reasons. I won't bother with another winding discourse through what all those reasons are, but suffice to say our only stated goal is to put the screws to the process and come up with the absolute best bracket we can.

And we didn't, because I just couldn't fully sell myself on the fact the committee was really going to go the route it did and I made a decision based upon that premise, which actually runs counter to the entire purpose of the exercise.

Thus I settled on the option that bundled Adrian/Oswego/Hobart as, after all, it had become a very popular option. The only problems were that it was the inferior option, I knew it was the inferior option, and I went with it anyway. I let the biases created by people who used to ravage this process in the name of self-gain creep back in instead of doing what I knew was right and trusting the committee to do the same -- which is exactly what it has been doing for some time now.

We warned in the column itself, on Twitter, and elsewhere that two West Region quarterfinals was a bonafide possibility, would be a certainty in an 11-team field, and at the last hour went so far as to say "we don't think its appearance is completely out of the question. In fact, we give it a better chance that probably most anyone else does..."

But I got cute and didn't pull the trigger on it.

The lesson? Trust this process and stick with it and you'll likely follow a winding path to the correct conclusion. Let extraneous influences creep in and cloud the process and you will be lead astray. A lot of today's discussions are plenty enough proof of that.

But alas, while our final bracket didn't match the one revealed this morning we're actually glad we were wrong because it means the committee was right. Additionally, we remain wholly confident that our Bracketology editions remain enlightening, entertaining, and educational, and help bring clarity to the process for a lot of people.

Nonetheless, it sure is nice to nail the sucker like we did last year. We should have had this one and damnit, it's really going to grind me for the next 364 days that we didn't. But ah well, live and learn, right? To that, we assure you all that we will continue to do just that as this is an exercise that can be perfected. And to that, I can assure you we're getting close. Real close.



[Click here to read full column]


Matthew Webb


Matthew Webb is a 2002 graduate of UW-Stevens Point and has been following Division III hockey since the 1990s. He is the former managing editor of, has covered seven of the past eight Division III Frozen Fours and was the publisher of

Previous Columns