NCAA Tournament Selection and Seeding
Back in December, I wrote an article giving a basic overview of how the BCS selection process happens, why certain conferences are better represented than others, and what goes into choosing the at-large bids. Now, it’s that time of year for another major postseason production in college sports. Only this time, we have a tournament of epic proportions, featuring the 68 best teams in a showdown that can only have one winner. Yes, ladies and gentlemen, it’s time for March Madness (only on CBS!). I always get excited this time of year. I know there are a lot of people who think that college basketball is only entertaining during the tournament, and I do agree with them to a certain extent. While it’s fun to root for your favorite team all season long, there seems to be something about do-or-die situations in a tournament bracket that give even the fair-weather fans a good rush of adrenaline. It’s a magical time when no work gets done, and no one cares that it doesn’t.
Like the BCS selection process, there are rules in place to help the NCAA tournament committee pick exactly who should be playing in March. Let’s start with the first major difference between college football and basketball: There is no limit to the number of teams that any one conference can have in the tournament. It was a point of controversy during bowl season that Arkansas (ranked well ahead of teams like Michigan and Virginia Tech) was not allowed to play in a BCS bowl game. The only reason for that was because the SEC already had two teams playing in a game (LSU and Alabama). But with the enormous field of potential teams for basketball, that rule is nonexistent. So let’s move onto the selection process.
NCAA Team Selection
It can’t be easy deciding who the best 68 teams in the country are, especially when there are literally hundreds to choose from. Luckily, the process is made a little easier by the individual conferences. The first 31 teams in the tournament come from automatic bids. There are 31 conferences that participate in the tournament each year, and the champion of each conference (30 conference tournaments, and the Ivy League regular-season champ) is automatically invited to dance.
That leaves 37 at-large teams that still need to be invited. Like I said earlier, there is no limit to the number of teams that can be invited from any one conference. That means, conceivably, there could be a season where all 12 teams from the Big Ten are in the tournament. (But, since Northwestern is still struggling to make the tournament for the first time, that’s unlikely to happen anytime soon.) The tournament committee, composed of 10 members, begins by making two columns on a piece of paper. Yep, you read that right. The first column is used for every committee member to pick 37 teams that he or she believes should be an at-large team (even if the conference tournament hasn’t happened yet, and a team could later be an automatic qualifier). Once this column has been made, any team that receives at least eight of the ten possible votes will be put into the at-large pool. So, there could in fact be a year where the second column isn’t used at all—if 10 grown adults can all pick the same 37 teams the first time around.
The second column is for teams that should receive some consideration. Each member can add as many teams as is humanly possible to this column, but it should be kept to teams that would have a legitimate chance to be in the tournament, and not, for instance, a team at the very bottom of a low-tier conference. This column is then used to create an “under consideration board.” Yep, that’s what they call it. (I’d like to take this time to apply to the NCAA Tournament selection committee as an advisor on naming things and otherwise picking better ways to make important decisions.) To get on the “under consideration board,” a team must meet one of three criteria: 1) Have more than one vote in either column, but not enough to be an at-large team; 2) Be recommended by more than one member; or 3) Win or share the regular-season championship for a conference.
Once the “under consideration board” has been established, it will be closed and the final selection process will begin. It’s important to note, however, that “closing” the board doesn’t actually mean anything, because a team can be removed and added to the at-large pool if it receives at least eight votes, and other teams can be added to the board if they receive at least three votes. When all this voting takes place, I don’t know. Because if everyone can vote for teams at any time they choose, then it really seems like there is no reason to have the initial ballot and go through the process of the “under consideration board.” But hey, I don’t make the rules. I just try to comprehend them. The next phase takes those teams on the “under consideration board,” and begins ranking them to determine who gets added to the at-large pool. Depending on how many teams are still on the board, the committee members begin ranking the teams. (20+ teams: Rank the top eight; 14-19: top six; 0-13: top four.) No committee member can participate in ranking with 24 or less teams in the pool if his or her team is one of those. The eight (or six, or four) teams that showed up on the most ballots move onto the next round of consideration.
With the next eight teams in hand, they are ranked from best to worst (one being the best, and eight being the worst) by each committee member, and the ranking of each team is totaled across the ballots. The four teams with the fewest number of points (i.e., if one particular team is ranked “1” on all eight lists, it would total eight points) are added to the at-large pool, and the other four are held over to the next round. The committee members then return to the remaining teams on the “under construction board” and rank their top eight. The four with the most votes are then added to the list with the four teams that were held over, and the second round follows. This process is continued until all 68 teams have been selected. It sounds like a lot of work, and it sounds like a lot of unnecessary work. But that’s just the way it goes, and it inevitably involves some teams that could make very good cases for their inclusion getting the cold shoulder instead.
A few final notes on the selection process:
- If a team goes through two rounds of the selection process without making the at-large pool, it gets thrown back to the “under construction board,” where it can be then moved back to the consideration rounds by receiving the proper number of votes.
- Making it into the at-large pool guarantees a team nothing. If the committee members have the requisite number of votes to do so (all but two of the eligible votes), a team can be removed from the at-large pool and returned to the board. Why would this happen? Either a late season slump, poor seeding in a conference tournament, or a down-right embarrassing performance in a conference tournament.
- At any time during the process, the committee can start seeding teams and removing teams from the “under construction board.”
The first step in the process is for the committee to rank the teams from 1 to 68. These “true seeds” stay the same through the whole process. But, like everything else, there are always problems that may keep some teams from being ranked in their “true seed” order. What kind of problems? How about the same bias issues that show up in the selection process. The same procedure of selecting eight teams, ranking them in order from best to worst, and the four teams with the lowest point totals are ranked from one to four in ascending order (lowest point total gets ranked first). The process continues until all 68 teams are ranked. After the “true seed” list has been assembled, teams can only be moved to different spots on the list by all but two of the eligible votes from committee members.
When it comes to finally putting everyone in the proper regional bracket, there are a lot of considerations that go into the task. Here’s an abridged list of some of the rules that the committee must keep in mind:
- More than two teams from a conference can’t be in the same regional bracket, unless that conference has nine or more teams in the tournament.
- Conference teams can’t meet each other before the regional finals, unless there are nine or more teams from that conference in the tournament.
- Teams 65-68 (and the last four teams that were placed in the at-large pool) will play in first round elimination matches to round out the field of 64 that will continue to the first major weekend of play.
- Teams can’t be placed in a region where they will play in an arena they’ve already played in more than three times during the season (i.e., Dayton couldn’t play in Dayton during the first round play-in games).
- Protected seeds will be placed in regions that keep them as close to their home crowds as possible.
- Rematches of regular-season games will be avoided—if possible—until the round of 16.
- Rematches of previous years’ games (in the tournament) will be avoided until the round of 16, if possible.
Now, once the “true seed” list has been compiled, and all the rules have been written on what I can only assume is the world’s largest white board that is constantly reviewed with every single damn selection, it’s time to actually seed the teams. Teams 1-4 will become the #1 seeds in the four regions; teams 5-8 will be the #2 seeds; 9-12, #3; and 13-16, #4. Seeds 1-4 are placed in regions in whatever way best balances their strength. (Using the points assigned to obtain the “true seed” order, the balance will be achieved by making sure that there is no more than five points difference between the lowest and highest seeds. Therefore, one bracket may have “true seed” #4 and #13, if that’s the way the numbers work. It also explains why you often have those regions where #4 seed teams will blow away #1 seeds in the Sweet 16 or Elite 8, because according to the tournament selection committee, those two teams may only have a 3-point difference, and really be more evenly matched than the bias of the committee suggests.)
Once the top four are in place, seeds five through 16 are chosen. Instead of going by “true seed” number, the committee will look at the point totals achieved for each team, and make sure that each tournament seed number is filled by four teams with the same point total (if possible). So all of the #6 seeds could have the same point total of “4,” but because of how the selection process worked to create the “true seed” list, one could conceivably have a #6 seed whose “true seed” was 53.
And there you have it. It’s a fairly long and arduous process. I hope I explained it well enough, because even I don’t know if I could process all the rules that seem to be in place for no particular reason while attempting to be fair and unbiased toward these teams.