Let Teams Choose Their NCAA Bracket Position

Nate Silver has (another) truly insightful post demonstrating the possible perverse advantage of receiving an 11th seed instead of an 8th seed in the NCAA tournament.

He explains:

[An average] team like Arizona would have a considerably better chance — about two-and-a-half times better, in fact — of winning its second round game and advancing to the Round of 16 as a No. 12 seed than as a No. 8 or No. 9 seed. This, of course, is because it has not yet had to face the No. 1 seed.

He also understands why this is a problem:

I think you need to be concerned any time you create an incentive for losing — which is what the current tournament structure does. For example, a team playing in a conference championship that figured to be a No. 9 seed if it won the tournament but a No. 11 seed if it lost might have some incentive to tank.

But when it comes to solutions, he misses an obvious one (which is somewhat inspired by Scorecasting’s suggestion that NFL coaches bid for possession in overtime).  The NCAA could rank the 16 teams in a bracket  from best to worst as it does now — but then allow the teams (starting with the number one seed) to choose which bracket position they would want to occupy.   The first and second ranked teams would almost certainly choose different halves of the bracket.  The 15th and 16th ranked teams would almost certainly still have to play the top ranked teams (because no other bracket positions would be left).  But we would quickly see whether the 8th ranked team would follow the Silver reasoning and opt for what is now the 12th ranked bracket position to increase its chance of surviving to the Sweet Sixteen.

Silver prefers a solution where the NCAA ranks only the top 6 teams, and the rest of the bracket position are determined by lot.  He envisions a tournament selection show “expanded from one hour to two, with the last half involving Vanna White and Dean Smith drawing ping-pong balls out of a tumbler.”  But I’d prefer for the second hour of the show to be having coaches on the clock choosing in succession their bracket position.  Creating an additional strategic decision might be even more fun than drawing ping-pong balls.



You neglect other factors, in particularly geography. Most teams will choose a seeding that will keep them close to home, rather than one that gives them easier opponents. it's not obvious how you could simultaneously choose location and seed and do so in an equitable manner.


What about a system - and this would been going back to the old way of determining locations and doing away with pods - of having seeding done like we do now, but go back to where half of one bracket is all played in one location, and the other half played in the other location, and then basically re-seed (or re-arrange) the bracket in each location.

So you have a 1/16, 8/9, 5/12, and 4/13 in a given location, and the 1, 9, 12, and 4 wins. Then re-seed the teams so that the 1 plays the 12 and the 4 plays the 9, instead of doing 1 vs. 9 and 4 vs 12 as it is today.

Then when you get to the regional semifinals, you can re-seed again, based on who is left.


I have thought about this before and it brings up a very interesting point. Though, I would argue that it has a very large flaw in that lower ranked seeds get a choice of the opponent while the higher ranked seeds do not. This would allow the lower ranked teams to pick opponents that may be better against an "average" team, but that is a favorable match-up for the lower rank teams. This advantage seems as though it may give a team even more incentive to lose so that they can choose who to play.

Perhaps the way around this is to rank the teams 1-16 and the higher seed chooses the opponent and the region?

Chris Bennett

Or they could reseed, a la the NHL playoffs.

Chris Bennett

If the 8 beats the 1, it gets the 1-seed's draw for the rest of the tourney. So in effect, 8s are one big upset away from an easier draw.

Not so for 11 seeds, for example, who replace a 6 if they win, but get a 6-seed's tougher draw the
as long as they survive.


I like the idea however I think you over value the incentive of playing a weaker team. If a coach has hig confidence in his preparation skills he might select a strong team to play against the first weekend because he'll have the opportuntiy to prepare for that team. Also there is the matchup factor. NCAA bball has so much to do with matchups and playing a "worse" team might be less appealing than playing a team that you simply match up with well. And finally your dealing with competetive college athletics. There is a macho factor that would result in mid level teams selecting the best team to play the first weekend because of course to be the best you have to beat the best. That sais none of these things necessarily precludes what is an interesting (albeit unrealistic) idea.

Steve Nations

This is a great idea.

Another great idea would be to realize that a playoff with 68 teams is just an unholy mess. I mean, really. Am I the only sports fan in this country who thinks the NCAA basketball tournament is a monstrosity that should be done away with? And why did the tournament expand in the first place? Because some team that was in theory the 68th best in the country was upset that it didn't make the playoffs? Sixty-eighth best teams do not get to whine about being left out of the playoffs. And when football goes to a different playoff format with 8 teams, how loudly do you think the 9th ranked team will complain? Especially if it has the same record as the 8th ranked team, which would be the case probably half the time, at least. I could go on and on. . . .


or let the top 8 teams choose their slot, and opponent.


This does not factor in that the system is designed to reward the 4 and 5 seeds and help them win their first game by giving them an inferior opponent. The 8 and 9 seeds may be upset that they have to face a 1 seed in the second round, but to avoid that, they should have played better and worked up from an 8 to a 6. If you let the better half of the teams choose their position in order and then let them choose their opponent in order, you would get the exact same bracket. The 1 seed will choose the worst teams, the 6 seeds will choose an 11 seed and an 8 seed would be stuck with a team of comparable ability.


Why would you want to create a system that gives the bottom half of teams the advantage of choosing their opponents?


On the topic of coaches being involved in choosing the bracket position: Europe's main rugby league competition (Super League) has a feature in its playoffs called Club Call, in which the highest ranked team from the quarter-finals gets to pick the team it wants to face in the semi-final. This was "a first for any world sport" when it was introduced in 2009.

(More info here on the playoff structure here: http://en.wikipedia.org/wiki/Super_League_play-offs)

The NCAA should go the whole hog and rank the teams 1 to 68, and then let the seeds have a choice in turn on who they want to play, down the list until no-one's left. Live on TV of course. It'd give a new meaning to the phrase The Big Dance.


As others have suggested, the advantage of choosing later is considerable. Perhaps the following reconciliation:

Each team, in seed order, can choose a bracket position, or pass. The lowest remaining seed must choose if all the other teams have passed. After a team chooses, restart the process with the remaining teams.

Presumably, all teams would pass, and the 16th seed's "choice" wouldn't matter. Then, the top seed could choose to play the 16. It gets interesting after that - would the 2-seed choose to place themselves in the opposite side of the bracket, or pass, intending to play the #15 team? If both the 2 and 3-seeds pass, would the #4 team choose to place itself in the opposite side from the #1, deviating from the traditional configuration?

Of course, there is the risk that the school with the best math department would have a relative advantage, and it's supposed to be about, um, basketball.



This is a crazy idea. Here's what we should do.

1. Rank the teams based off of regular season performance.
2. Stop playing for a few weeks.
3. Have the number 1 team play the number 2 team.
4. The winner of that game is the national champion.

It simplifies everything. You eliminate all the whining about whether or not someone should have made the tournament despite 68 teams getting in. Only a few teams would have the ability to whine. The regular season is more important and every game is like a playoff.

Also, I'm kidding. College football is the worst system.

But most college coaches wouldn't like this. Anytime that someone can shake things up by making an unusual choice in the first round, incumbents complain. What if a 4 seed level coach decides he matches up better against the 1 seed and takes him in the first round? It would shake up the tournament a ton. Lots of fun though.



A variant system:
Starting with seed #1, each team can pick their slot or pass. When a team picks rather than passes, start back at the top of the list. The lowest seeded team that has not yet picked must pick if all higher teams pass.


I suggested Tox's variant, apparently while Tox's comment was awaiting moderation - so if the moderators don't kill it as unoriginal, I promise I wasn't copying.

However, I disagree on Tox's analysis of what will happen. First choice-round I agree on: team 16 will have to pick their spot. However, team 1's 'pick a spot whenever we want to' advantage is too valuable to blow on choice-round 2 for no greater return than one single easy first tournament-round match. I think it will be a team in the range 6 to 10 that will pick the spot next to 16. If it is 9 or 10 which takes that spot, then in choice-round 3, team 2, 3 or 4 may well take a spot to play the winner of that match in tournament-round 2.

Team 1 will probably watch which half or perhaps quarter of the draw seems easiest to them, and then pick the last available slot in their preferred section. Teams 1 through 4 can feel entitled to being in a quarter of the draw which doesn't contain another top team. The longer they delay, the more information they have to make the best choice. As soon as one quarter is all-but-one full, team 4 is under great pressure to make their pick - otherwise one of the lower teams will fill out that quarter, and team 4 will be forced into a quarter containing another top team. Team 4 doesn't have to pick the all-but-one-full quarter - if they pick a different quarter, teams 1 through 3 will probably avoid picking to go in with them.

Extending this logic, when an eighth is all-but-one full (i.e. in a tournament of 16, has one of its two spots picked) then team 8 is under pressure to pick a spot, and this happens in choice-round 2. So 8 vs 16 is a likely result of the first two choice-rounds. (But not inevitable - team 8 might feel that team 16 is underrated and don't want to play them, or maybe that team 7 is overrated so they're looking to avoid a first round match against 1-6 or 9 and so are happy to let team 9 pick before them.)


Daniel H.

Is anyone concerned a potential 11 seed will actually tank on purpose? This seems highly unlikely considering 1) the teams are presumably highly motivated by competitive spirit to win their conference tournament, and 2) teams at or around that seed position are in a precarious position. If they lose, there's no guarantee the selection committee won't leave them out of the tournament altogether. Even in situations where a loss would clearly not hurt a team's chances of making the tournament, and could have long-term benefits (e.g. losing to a bubble team from its own conference, potentially increasing the number of tournament teams from that conference and thus the conference's perceived strength), I have never once heard of a team losing on purpose. These are 18-21 year-old kids we're talking about, and I think they genuinely want to win each game regardless of its long-term consequences.


Nathan Robertson

They should take note from the Eurpoean Champions League soccer tournament, who re-draw after every round to see who they are facing.


doesn't this give later teams the advantage of selecting their opponent?

Freakonomics- An Interesting Spin on Traditional Economics

[...] Freakonomics, one of my favorite books and blogs, is a collaborate project between an economist, Steven D. Levitt, and a writer, Stephen J. Dubner. Initially, the pair thought it would be interesting to join forces in writing a book that analyzed an array of topics under an economic microscope.  What they initially thought would be a small project escalated at a rapid pace.  A book became two books became a documentary became a blog became a podcast!  Freakonomics is now a widely recognized term.  The blog is so interesting because the authors of the book gather economic articles concerning a wide variety of topics and post them in one place.  One of my recent favorites suggests that the NCAA let teams choose their own bracket position, as it would make March Madness that much more interesting.  You could read that article (and many others!) here. [...]