Explaining the Final Bracket: Committee Q&A and Analysis
All of the talk about whether or not Denver, as the tournament's top overall seed, would be "protected" from having to play in the Fargo Regional, seemed to be accurate when the Committee announced Denver would play in the Midwest Regional in Cincinnati.
However, that was not really the reason at all. In fact, this year's bracket was as straightforward as can be compared to recent years.
The Committee starts out with the list of 16 teams. It then tries to group them in fours like so: 1-16, 8-9 in one region ... 2-15, 7-10 in another ... and so on. Teams then get flipped around for a variety of factors. Some, such as avoiding intra-conference first-round matchups and host teams needing to be in that Region, are absolute and cannot be ignored. Other factors are more subjective or vague, such as "protecting" certain teams, or moving teams around for attendance reasons, or other logical factors.
My general philsophy has been: The Committee should absolutely feel free to move teams around logically, in order to improve the tournament, for whatever reason. But sometimes, it veers too much into things like "maximizing attendance" to the overall detriment of the integrity of the bracket. Other times, it sticks too rigidly to those seeds for no good reason.
There's also a rule, or guideline really, that you first place the No. 1 seeds in order — starting with No. 1 overall, then No. 2 overall, etc... — in the regions "closest to home." They deviate from this however, and this year, it wasn't considered at all, as you'll see.
Ultimately, the only actual switch this year was taking Cornell from Providence and moving it to Manchester. This created a 5 (Lowell) vs. 11 (Cornell) matchup, and a 6 (WMU) vs. 12 (Air Force) matchup, instead the "natural" 5-12, 6-11.
Here is a conversation with NCAA Ice Hockey Committee Chair Tom McGinnis, the associate athletic director at Minnesota.
CHN: It looks like once you placed the top seeds, it was straightforward — you just bracketed it in.
McGinnis: Yeah, we stayed very close to seeding this year. So we filled it all in. No first-round matchups against conference opponents. Then we looked for any opportunities to provide better atmosphere in venues with potentially moving a team or two. The Cornell-Air Force switch was just to get two Eastern teams in Manchester just because we already had two in Providence. But then beyond that, any other change that was considered would have some ripple effects, so as a Committee it was decided this is what we'll go with.
CHN: What about, say, switching Western Michigan and Union, getting Western Michigan to Cincinnati (instead of Providence)? The school is somewhat close to Cincinnati.
McGinnis: I don't know if that was felt to be a significant switch. When you look at it the first time through, you had Cornell (in Providence) so you couldn't move Union there. After (Cornell was moved to Manchester) you could've made that (Union-WMU) switch as well, but I don't think they felt that was a dramatic enough change.
CHN: It's interesting. Each year the people on the Committee change a little bit. Some years there's an extreme emphasis on attendance. Maybe five years ago was more like this year, whereas I feel over the last five years, there's been more and more emphasis on attendance where even some of these marginal moves were made. And this year you said maybe it's not worth it.
McGinnis: Maybe "not worth it" is not the right word, just maybe that we already had some teams there that would cause some draws. How many do you really need to move into one region?
CHN: So did you guys believe that Penn State was a school that could draw there in Cincinnati?
McGinnis: I think (Penn State) is almost equidistant between Providence and Cincinnati. So if you're truly just looking from their campus in State College — and obviously they have fans from different places — it was pretty (similar) distances. So the idea was, if we're holding to the seeds as much as we can, hopefully their fans will want to go out for the first time in the tournament, and it's really exciting for them, and it's a spot people can get to.
CHN: So obviously everything trickles from the first major decision, and I put it out there as a possibility, knowing you could do this, which is putting Denver to Cincinnati. So can you just talk about that thought process.
McGinnis: Obviously you have each No. 1 seed, but we didn't necessarily place the No. 1 seeds in locations. It was more, put the bracket together and look at where they need to go. And then true bracketing, obviously North Dakota being a host would need to be in Fargo, so that four-team group went to that location. As opposed to saying 'We're going to put Denver in a spot,' we put the brackets together (first) and then said where is the best spots for them to go.
CHN: [Note, North Dakota finished 10 overall, so naturally lines up with the 7 overall, and the 2 overall in one bracket foursome.] Right, but was there a conversation about keeping the top seed in the tournament (Denver) away from Fargo?
McGinnis: No. The first principle was seeding the tournament 1-16, seeing where teams needed to be. It wasn't so much keeping Denver away from Fargo because good, bad or indifferent, it's a neutral site. I understand there will be a fair amount of green jerseys in the building, but that wasn't a factor because it worked out. Now, had the numbers been different and North Dakota been an 8 or 9 seed, maybe we'd go down that path, but we didn't need to.
CHN: I will say, in some years past, there have been philosophical — Committee members have said, they're going to try to protect the 1 seed if possible.
McGinnis: I think protection for the top seed, in most instances, they need to play the 16 seed. So whoever that automatic qualifier is from somewhere that isn't one of the top 16 teams, that's when we say "protect the top seed." That's what we need to protect, not so much the location of where they're playing.
CHN: So it all flowed from there. ... The other big issue is Providence in Providence. Now, obviously the bracket just worked out that way, so you can say that's the case, but it is an elephant in the room, people weren't happy about it two years ago (when Providence was also a four seed in Providence, and won the Regional against Miami and Denver). So was there any thought, even for a second, of switching them with Notre Dame or something like that?
McGinnis: No. The idea is, we try to avoid those first-round matchups, but then, we are trying to provide atmosphere. And whether you're a 1 seed, 2 seed, 3 seed or 4 seed, if you've got an opportunity when we can to get a couple teams geographically close to the site, we're going to do that. And that's just how it worked out. I understand the conversation that happened a couple of years ago, but in the format we have now, we pick these sites, it's not their home site — now granted they are close and have an opportunity to have a lot of fans there, but the biggest concern in picking these sites at this point is making sure nobody plays in their home rink. Some of them are going to be geographically close to certain schools.
CHN: The argument I've made is that, because they are a low seed, they don't deserve that benefit. It's a nice benefit they're getting, despite not really earning it.
McGinnis: I totally understand that perspective, that's the flip side argument — but that's not the process we follow at this point.
Some people like to criticize our projections when they're not right. They miss the point of the projections. You can't read the minds of Committee members, many of which change from year to year. I try to lay out all of the possible scenarios, and then explain the one I think is most likely to happen. The possibility of things being different is baked right into the projection article, if anyone bothers to read it all.
That said, we missed one scenario the Committee went with this year, which was to ignore the idea of placing No. 1 seeds first, and simply group the foursomes together and logically let them fall into whatever Region they may.
I have no problem with this, necessarily, so long as the end result is sound. It was just a bit of a different process than we're used to. But that's the way it always is — the members change, and every year it seems some process is slightly different than the year before.
In the end, the only major issue I have with the bracket is Providence being in Providence. Just like in 2015, I don't think this is right. And this is not personal to anyone at Providence, it just so happens to be the example here.
As I said above, I get and generally agree with the idea of maximizing attendance where possible. But not at the detriment of bracket integrity. Providence, as a low seed, should not get that benefit, to Harvard's detriment. The idea McGinnis said about these just being "neutral sites" after all, is a bit willfully blind to the reality. Harvard doesn't want to play Providence there. It has a right to be annoyed.
What should happen is, the Committee should allow itself to move teams around the bands to maximize attendance and avoid that matchup. Providence and Notre Dame could've flipped. Or flip Providence and Ohio State. Then, to help attendance in Providence, you could've put Boston University there, or Union, and put Western Michigan in Cincinnati. Yes, Providence will be a bigger draw in its home town than BU or Union, but BU or Union will get enough people to make it OK, while Providence doesn't get that benefit it doesn't deserve.
And putting Western Michigan in Cincinnati has the added benefit of helping that bracket too, even if just a bit. That might be a marginal boost, as McGinnis alludes to above, but it's something, and it can be done without any disadvantage.
Ultimately, as I told McGinnis, we'll complain about this stuff for a couple days, and then forget about it and just enjoy the tournament. I've never doubted the integrity of the Committee to do what those guys believe is right. But these are topics at least worth discussing when determining ways to improve things in the future.