NCAA Tournament Selection Redux

Posted by: Mike Machnik

The field of 16 was announced this afternoon, and there weren’t any surprises as far as we were concerned. But let’s take a quick look at a couple of the issues that did come up and how they were resolved. (Adam will likely have his own analysis soon as well.)

The big issue was how the committee was going to break the tie for 11th place in the pairwise. Adam and I agreed that the recent precedent was to break those ties with RPI, not the comparisons among the individual teams, and so we ended up ordering those three teams as follows: #11 St. Lawrence, #12 Maine, #13 UMass. Note that this drops UMass into the band of No. 4 seeds in the tournament — the #13-16 teams.

Indeed, that’s what the committee did. If they had used the comparisons, UMass would have been #11, St. Lawrence #12, and Maine #13. The impact is on which team is a No. 3 seed and which is a No. 4 seed. SLU is a No. 3 seed either way.

As a result, UMass, as a No. 4 seed, ends up facing No. 1 seed Clarkson in the East Regional. And Maine faces St. Cloud. (There’s more detail — UMass should have faced UNH, but was switched with Miami to avoid the intra-conference matchup, and the same for Maine, which was switched with SLU to avoid BC. But you can read about that in the bracket prediction article mentioned above.)

Note that the tie for 8th between Michigan and Michigan State — also big, because one would be a No. 2 seed and one a No. 3 seed — would go to Michigan whichever method was used, RPI or the head-to-head comparison.

Another issue was where the four No. 1 seeds would go. Northeast host UNH was going to Manchester, which left the other three. Normally, the No. 1 seeds are to be placed in the closest regional to them in order of seeding. That would have meant Minnesota to Grand Rapids, then Notre Dame to Rochester, and Clarkson to Denver.

However, that just doesn’t make sense for a number of reasons which should be clear. Fortunately, the committee was able to take a small degree of latitude and shift the teams to what we got — Minnesota in Denver, Notre Dame in Grand Rapids, and Clarkson in Rochester. That was made possible by a few factors, including Minnesota needing to fly either way, and West host Denver not making the tournament.

We did think the committee would make that move too, and they did.

The third major issue we discussed was whether or not Air Force would be kept in Denver. On this one Adam and I differed. One of us felt the committee would go right by the numbers — since they were going to do so in breaking the #11 tie — and match up #1 overall Minnesota with #16 overall Alabama-Huntsville, along with sending #15 Air Force to Grand Rapids to face #2 Notre Dame. The other one felt the committee would keep Air Force close to home since they were SO close and would not need to fly, there wasn’t much difference “by the numbers” (both teams were not teams under consideration and were well down the list in RPI), and because like the No. 1 seeds situation, it just “made sense”. This was a toss-up and while we ultimately went by the numbers, I will say I personally like the way it worked out better with Air Force making its first ever tourney appearance in its backyard.

Overall, we were as close as you could get without being 100% on the money. But as we’ve been saying, that’s really a reflection of how refined and predictable the process has become, even down to recognizing the cases where the committee could and would make slight deviations for the better of the tournament as a whole. To be honest, I don’t look at the whole thing until the night before the selection, because things can change so much even in a day. But the process is so refined that in about 15 minutes we have the brackets drawn up, the issues identified and the predictions made.

Best of luck to all the teams and here’s to an exciting tournament!

Leave a Reply