Restructuring the March Madness bracket

Realistic rating: 5.5

With a No. 7 and No. 8 seed qualifying for the Final Four, this year’s tourney again highlights some of the same issues that have always plagued college basketball—the extreme unpredictability of the postseason. While most fans probably love this aspect of the sport, it brings up a number of problems with the college basketball season as a whole, namely: the meaningfulness of the regular season and the frequency with which the best team actually wins the championship.

Perhaps the most fundamental concept every sport must confront is how to balance fairness vs. excitement—or in other words, how often should the best team win? There are some sports that probably have too high a percentage (e.g. sprinting, NBA) at the expense of excitement/unpredictability, while others have far too low a frequency of the best winning but enjoy a very high level of excitement/unpredictability (e.g. MLB).

College basketball falls about as far on the excitement end of the spectrum as any sport. This makes the postseason extremely fun and unpredictable, but the regular season, well, is just about meaningless. Plus, the “champion” is rarely the best team.

This is due to the postseason’s structure—a 68-team single elimination tournament in which every team, with the exception of the eight involved in play-in games (no, I’m not calling those “round one”), has ostensibly the exact same starting position, requiring the six victories to win the championship (i.e. there’s not much reward for regular season performance other than a higher seed, which in many cases is actually a negative.) Thus, as much as ESPN likes to hype a regular season game such as No. 1 Syracuse vs. No. 5 Duke last month, it really doesn’t matter who wins, as there’s not much reward for the victor.

A couple of weeks ago on Deadspin, I wrote a piece discussing some of these issues with this format and proposed a few alternatives, such as a World Cup style, a bye tiered bracket, and a standard bracket with randomized lower seeds. (The heavy math for these proposals is there.) While these were all certainly improvements on the current format, they were, as Mike Ehrmantraut would say, a “half measure.”

Expanding (Yes, Expanding) the Bracket

Considering there are 351 Division I basketball teams, allowing 68 (19.4%) into the postseason a good figure if you must have a single-elimination, limited-tier (by “tier” I mean giving certain teams byes or other hard advantages other than playing a weaker first round opponent). However, contrary to what most hardcore fans may believe, making the postseason more exclusive will only make the regular season more meaningful for roughly the top 25% of teams. With fewer teams in the tourney, the regular season would be far less meaningful for the bottom 75%, because for the vast majority of teams in smaller conferences, their season is entirely meaningless only a dozen games into the season. (Note: Yes, there are incredibly dumb conference tournaments that render single-bid conference regular seasons meaningless, but we’ll just ignore that for now.)

But how can you make the regular season more meaningful for all 351 teams without making the postseason even more of a crapshoot? Here are a couple of straightforward solutions, which could really be applied to all sports:

1. Simply have best-of-three series at certain rounds of the tourney. This would be highly impractical for a number of obvious reasons, though, so let’s assume we have to stick to single elimination. Also, while decreasing the randomness, best-of-three series still wouldn’t make regular season success more meaningful.

2. Have a dramatically bye-tiered bracket. The example for this kind of setup in the Deadspin piece only had three bye tiers: #1-4 seeds on bye to the round of 32, #5-8 at the normal starting position, and #9-16 with a play-in game to the first round. This still didn’t result in quite as much significance for regular season success as I would have liked, though, but how about some of these?




In each of these brackets, there’s now a much greater reward for landing a higher seed, which would make the regular season far much more meaningful while also giving much higher likelihood to the chances of the best teams being there the final weekend. I’m a huge proponent of tiered bracket systems in which higher seeds get greater rewards (in the form of byes or otherwise) and wish more leagues other than the NFL would adopt such measures. (More on this in future posts to be sure.)

I’m not sure if I’ll be rooting for UConn or Kentucky this weekend. Should either win, it will be a glaring mark on the legitimacy of the NCAA champion title, but at the same time, it might garner further exposure for the sham the current format really is.

Bottom line: These proposals really don’t shake up things all that much (i.e. still single-elimination bracket, still four regions, etc.), but with the NCAA making a boatload of money as is, the probably won’t want to change much. Still, with even the highest regular season game (a 5000+ game sample size) rating falling well below almost all of the 68 tournament games, there’s certainly a missed revenue stream opportunity.


One thought on “Restructuring the March Madness bracket

  1. Pingback: How well do Vegas futures predict results in each sport? | Lower The Mound

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s