ncaa tournament part 2

About ten years ago I was laid up for a month. With nothing else to do, I spent the time watching every game of March Madness that was televised. I fell in love. This was basketball at its purest and finest. Not like the NBA where the rules only seemed to be whatever the officials decided they were on a given night. College players were expected to play by the rules, and the officiating was quite consistent. If a player got away with something, it was only because an official had missed it, and not because of who the player was, or how it might affect ratings.

Ever since then, I have always made it a point to watch as many games as I could each year, using vacation time on at least two occasions in order to be able to watch the finals.

Having said all that, I have always been amazed at the oddities the selection committee seems to come up with in regards to the scheduling of games. This year, both Utah and Arizona had to travel all the way to Miami to play against each other, and BYU and Texas A&M played each other in Philadelphia, among others.

So this year I decided to see for myself if it could be done better. First I searched the internet to see if I could find out exactly how the committee does its work. I found a rather detailed article in Wikipedia, which described the process pretty much as I thought it would be. Once the teams are selected, they are all placed in ranked order from 1-65, although exactly how they determine the ranking must be a state secret. The authors of the article seem stumped by it, as they point out how the committee so often snubs the Missouri Valley Conference when its teams are ranked in the top 40, and how Utah State was snubbed in 2004, despite being 25-2 and being ranked in the top 25.

The article goes on to describe how the top 4 teams get the number one seeds, the next four get number two seeds and so on. Then the highest number one seed is placed in the same region as the lowest number two seed (number eight ranked overall). By extrapolation, the teams with the following rankings would all be in the same region:
1,8,9,16,17,24,25,32,33,40,41,48,49,56,57,64. This is all exactly as I would have expected. The article then goes on the say that “adjustments” are made for a variety of reasons. Ah, Ha! A loophole! First, the top three teams in each conference must be placed in separate regions. This means that if the 1-65 ranking system would place the top two teams in a conference in the same region, one of the teams is reseeded and moved to another region. If a conference has more than three teams in the field, and two teams are in the same region, they will change the seeding so that they would not play each other until the regional final. This is not an absolute rule, if it contradicts another rule, but they will do it whenever possible. They will also change seeding to allow teams to play closer to home, if it does create too large a disparity in the expected performance of the regions. There are also other unnamed reasons the committee will change seeding.

In order to test this out, I needed to know what the 1-65 ranking was. I could not find this anywhere. I don’t know if it’s another state secret of the committee, or I just didn’t know where to look. So I tried to recreate it as best I could. It was not too difficult, actually. I knew that Louisville was #1 overall, so I could easily reconstruct that one. And I decided that Connecticut must be #4 overall, since they would play Louisville in the national semi-finals should both teams make it that far. The problem was not knowing who was #2 and who was #3 between Pittsburgh and North Carolina. I picked North Carolina as #2, but that was entirely arbitrary. If I was wrong, then half of the teams would have to be flip-flopped in my list, but all the flip-flops would only be one place change each, so it didn’t make a whole lot of difference. The biggest problem is that my list is actually a list of teams AFTER all the adjustments were made, not the original list.

To make a long story short, after I studied it from several different angles, and listed different possibilities, I discovered that, although I would have done things differently, for the most part, the brackets that the NCAA selection committee came up with cannot be said to be wrong. (Note how carefully I just worded that.) It’s just inevitable that some teams are going to have to travel a long way to play. I think the problem stems from having TWO first round sites way up in the northwest corner of the country when there are only a handful of teams within a thousand miles of there in the tournament. If they feel there must be two sites in the west, the it would have been better if one of them had been in, say, Utah, Colorado, Texas, or even Arizona. I realize the sites are chosen a long time ahead of time, but surely the NCAA has a general idea of the locations of most of the schools that are likely to be in the tournament. Only five of the top 16 teams in the field are located west of the Mississippi, and only two of those are west of the Rocky Mountains. In fact, in the entire 64 team field, only about a dozen teams are in the western third of the country.

I think the most important thing in bracketing is that the top 16 ranked teams be seeded properly so that if all 16 make it to the regionals, #1 would play #16 and so on. And just like in football, I think this seeding should be done using a set formula, not left to the “judgments” of people who may or may not have a vested interest in the outcome. Once that’s set, don’t make any adjustments. As far as the rest of the teams in the field, I think it makes little difference who plays who. If, in general, the lower seeded teams are playing the higher seeded teams, and the teams in the middle are playing each other, it shouldn’t make a whole lot of difference what their exact seeding is.

At first I was a bit upset at the fact that Cleveland State had to travel all the way to Miami to play, although I doubt the players considered that bad, and that Akron had to go all the way to Portland, Oregon to play. I figured it was just the usual “dissing” of anything Cleveland. But once I started making up my own brackets, I discovered that, because of their seeding, those were really the only two places they could have been assigned.

As noted above, two first round sites were up in the Pacific Northwest. Of the top 16 seeds, Washington and Gonzaga were assigned there, as being the closest top seeds. But that left two openings. As #2 seeds, Memphis and Oklahoma were allowed to play in Kansas City, so Kansas was sent to Minneapolis as the next closest place, and Missouri was shipped out west, probably because Kansas was ranked higher overall. It may seem a tad unfair that #3 Missouri had to travel while #4 Gonzaga got to play closer to home, but there just wasn’t anywhere else to put them. As far as the other western opening, it seems it would have to have been either Xavier or Wake Forest, and since Wake Forest was “needed” in Miami, Xavier got the call. Every other one of the top 16 did get to play closest to home. So it would seem that when all is said and done, I don’t have anything to really beef about, and there is not as much that I would change as I thought.

As for the two games I mentioned at the beginning? I would have sent Utah to Portland, and Illinois to Miami. I would have sent Arizona to Portland and Western Kentucky to Miami. I would have sent Texas A&M to Kansas City, and Maryland to Philadelphia. And I would have sent BYU to Kansas City and Clemson to Philadelphia.