Don't Mourn the NCAA Tournament
Expansion only makes the tournament better, and if we're being serious, we don't do a very good job of seeding the tournament right now.
Much hand-wringing has been made about the NCAA Tournament expanding. Critics say this move dilutes the tournament. It’s just a money grab. It sets a dangerous precedent and threatens to ruin the one thing college basketball has going for it.
I say bring it on. Let everyone in. Expanding the NCAA tournament only makes it better and here’s why:
The tournament was never really a 64-team competition anyways.
Tournament admission does not go to the best teams.
The tournament is ultimately about volume.
1. The Tournament was never a 64-team competition
Looking back at tournament history from 2025 to 2000, the only team to win the NCAA tournament with a seeding worse than 3 is UConn. UConn did this twice, as a 4 and 7 seed. They have won six national titles since 1999, making them the most successful college basketball program in the nation. They are the exception. Not the rule.
Since 1985, the rule are this:
65% of NCAA tournament winners are one seeds;
75% are one or two seeds;
85% are one to or three seeds;
90% are one, two, three or four seeds.
College basketball has a bit more randomness than college football, but after a full season, we know who the best teams are. We can narrow the top teams down to about a dozen or so elite teams and be +90% confident that the best team is among them.
So what are we doing with the other 48 teams?
2. Tournament admission is not for the best teams
The NCAA tournament itself is part of the prize. Just getting in is a reward for dozens of teams that have no hope of making it. No team seeded 9 or above has ever won. But an 8 seed has. So maybe it’s possible.
Conservatively, let’s say that seeds 10-16 have a less than .005 chance of winning the tournament. There are 24 of these teams that get admitted. Each of which averages a 0.02% of winning the tournament. These are not top teams. And at the bottom of the pecking order, we don’t try to pick top teams; we reward teams.
Seeds 10-16 go to conference championship winners from bad conferences. They go to 20-win teams that are “deserving”, but come from bad conferences, play poor competition, and don’t have real recruits.
It’s not obvious that we could differentiate teams at this level even if we wanted to. As we move towards middling teams, the margin of error on our predictions about their performance increases.
And often, these middling teams are swingier. They have good nights; they have bad nights. Who is to say which version of the team will show up on any given night? That makes predicting their success over a single game even less accurate.
Consider the following underdog win rates (rounded to nearest 5%):
4 seed v. 13 seed: 20%
5 seed v. 12 seed: 35%
6 seed v. 11 seed: 40%
7 seed v. 10 seed: 40%
8 seed v. 9 seed: 55%
Between seeds 6 and 11, it’s not obvious that we can tell which team is better than the other with any real confidence – let along rank teams in order with any meaningful confidence. For 8s and 9s, it is actually more likely than not that the bracket seeders get the ratings of the teams backwards: the rank the worse team better and the better team worse.
Why do we think we can tell 10s and 11s apart? Or 12s and 13s? Or 15s and 16s?
We probably can’t. Not on any one night.
3. The tournament was always about volume
The. NCAA tournament is about volume. The best part of the tournament is opening weekend. 20 million people watch. That’s the same number of people that watch the NCAA final.
Why do they tune into these games? Because it’s an event. Everyone is playing all at once. If you watch the games at a sports bar, every TV will have a different game. There’s constant action. With the rise of sports gambling, it’s not uncommon for someone to have 10 or more bets resolving over the course of the day. It feels like a holiday.
This effect is created entirely by volume. No one would tune in for any of the individual matchups. But collectively, they become an event.
This scales. More is better here. The NCAA has it right. The limiting factor is how many weekends they can sustain engagement. Within the context of a single weekend, more is better.
A modest proposal for a 112 team NCAA tournament.
Double the opening weekend volume. Let 112 teams in.
64 teams–seeds 13 to 28–play on day one. Then the next day, they face off against seeds 5 to 12. Then, Saturday, teams seeded 1 and 2 play the weary victors from the first two days. And Sunday, teams seeded 3 and 4 play the Thursday/Friday victors, who have now had a chance to rest a day.
Then you’re on to your Sweet 16 for the second weekend after 80 games of basketball. +30 games on Thursday and Friday. 8 games each on Saturday and Sunday.
Open the flood gates. Roll out the ball. Let them hoop.