March Madness (i.e., the NCAA basketball tournaments) will soon be upon us. It is the one time during the year that many people who usually think about brackets not at all will briefly become very interested in them. Alas that that interest will focus on brackets that are so grievously flawed.
I’ve written before about what’s wrong with the NCAA tourney, but I’ll do it again, presently, pulling together some ideas that are otherwise scattered among several posts. But first, lets give credit to a college basketball tourney that does things pretty darned well: the B1G (i.e., Big Ten) tournament recently concluded at Madison Square Garden in New York.
For reasons I won’t go into here, the Big 10 now has 14 teams. The obvious format for an elimination tourney for 14 teams is a 16 bracket, with two byes. Here is such a bracket, with conventional seeds according to the regular season result for each team: B1Galt.
But that’s not what the Big 10 does–they do something much more sensible: B1Gact. As this was the bracket actually played, I’ve filled in the result. The main difference between the two is that the basic structure of the real bracket comes from a 32 bracket, with multiple byes rather than just a couple. Four teams get no byes at all, and play in the first round. Six teams get a single bye, playing first in the second round. And the top four teams get a double bye, appearing first in the third round.
Why is this better? Let me count the ways.
- The first round won’t be mostly blowouts. The teams at the bottom of the table get a chance against another poor team that they might actually beat. Now, perhaps it would be better to excuse the worst teams from the tourney all together, but if they are to be included, give them one game they might win.
- The regular season means something. Getting into the top four, and earning the double bye makes an important difference. The besetting embarrassment of such conference tourneys is that it tends to render all of the other games meaningless.
- It evens out the fatigue factors. Since the tourney is played on a single court, if you ran the 16 bracket over the five days, you’d have to split the first round between two days, which means that some teams would get a day’s rest between two successive games, while others wouldn’t. Of course, any team that expects to win through from the first round will have to play through considerable fatigue, but that’s appropriate, I think, as an additional reward for the regular-season success of the teams that get byes.
We’ve seen this technique before for competitions like the UEFA Champtions League in this post: The Milan Miracle.
Pretty much everything (and more) that’s right about the B1G championship is wrong with “March madness”. More on that to come.