Crazy Idea: Allocate Set Number of Tournament Bids By Conference Ranking

Before I get to the idea, though, some quick thoughts on Michigan State. This will, I promise you, have some bearing on the crazy idea. Apparently, some people who will remain anonymous think that Michigan State shouldn’t be in the tournament and putting them in is a sign of a drinking problem. It doesn’t matter what you or I think, and I think they have played pretty poorly of late, because they are in right now. If you want to know why, it’s because they play in either the #1 or #2 ranked conference by any measure, and conferences that finish that high get at least half their teams in almost without fail (only 1 conference that ranked in top 2 in last 8 years failed to get at least half its league members in the tournament). Michigan State is likely going to finish at least 9-9 in one of the top two conferences in the country. In fact, I would say based on historical allocations, the Big Ten will get six in–Ohio State, Purdue, Wisconsin, Illinois, Michigan, and Michigan State.

Which leads in to my crazy idea. Get away with all this 37 at-large stuff, where we debate a bunch of teams that you can’t keep track of across conferences, trying to make comparisons between a bunch of teams with tenuous connections who have various pros and cons in favor of being selected for a spot that likely won’t have an impact on the national championship. Allocate bids to conferences based on measured conference strength during the season.

If I can draw an analogy to Europe and the Champions League, the equivalent of the “at-large” spots granted to each country is based on past performance. We know that the top leagues generally rotate between Spain, England, Germany and France, but the order changes over time. You do relatively better, you get more bids. But they don’t debate how the fourth place team in one league compares to the fourth place team in the other based on recent performance, road record, or strength of schedule. They are pre-assigned.

So I say pre-allocate the vast majority of the current “at-large” bids to conferences. In practicality, this is what ends up happening anyway. I checked the last eight seasons, to see what percentage of conference members got NCAA bids based on where the conference ranked.

The top conference averaged 54% of it’s teams in the tournament;
The 2nd conference averaged 52% of it’s teams in the tournament;
3rd and 4th conferences averaged 45% of teams in;
5th averaged 40% in;
6th averaged 35% in;
7th and 8th averaged 27% in;
9th, 10th and 11th best conferences averaged 18% in; and
12th to 14th best conferences averaged 14% in.

In 8 years, there were only 9 at-larges that came from outside the top 11 ranked conferences.

Using those historic percentages as guide, and then reducing them slightly, conferences are allocated tournament slots based on their relative ranking by whatever system the committee wants to use: RPI, Pomeroy, Sagarin, or a composite of several. At some cutoff date (let’s say right after Bracketbusters), the numbers are crunched and the bids by conference are allocated. The percentage allocations would be:

#1 conference: 55% of teams get allocated a spot for the NCAA tournament;
#2 conference: 50% of teams;
#3 conference: 45% of teams;
#4 conference: 40% of teams;
#5 conference: 35% of teams;
#6 conference: 30% of teams;
#7 conference: 25% of teams;
#8 and #9 conference: 20% of teams;
#10 and #11 conference: 15% of teams.
all others: 1 automatic spot allocated

Applying those to this year, just using the RPI conference rankings for simplicity, and rounding to the nearest whole bid number, you would get the following allocations: Big East (9); Big Ten (6); Big XII (5); Mountain West (4); Atlantic Coast (4); Southeast (4); Pac-10 (3); Conference USA (2); Atlantic 10 (3); Colonial Athletic (2); Horizon (2); everyone else (1).

If you add those up, you will see that 33 of the current at-larges would be used, leaving only 4 true “at-larges” where the committee has discretion to add more teams to the field beyond the conference allocations. Most of those allocations account for the teams I already had as an at-large in my latest projections anyway. I didn’t quite have Richmond in (some other projections do) but would I complain if an automatic allocation put them in because their conference earned three spots? Not at all. Nobody that really clearly deserves in is getting jobbed here, and there is a lot more certainty for teams that know what they need to do. Those four spots are open for top teams from weaker conferences (let’s say a Utah State if they don’t win the tournament) or teams from bigger conferences that exceed the allocation.

The SEC knows they are getting 4 teams for sure. Tennessee’s recent bad play puts them in jeopardy because the conference wasn’t relatively strong enough among the Big Boys, and that Alabama-Georgia game to end the season becomes huge. Winner is in, loser is in the mix for one of those true select “at-larges”. Villanova’s recent collapse brings Cincinnati and Marquette into the mix for catching them for the 9th Big East spot, and the other two are also in the mix for the at-larges. UAB, Southern Miss, and Memphis, they are in a three-way battle for two spots from Conference USA, and they know that whoever finishes third is out. But at least they know for sure that two are in based on the conference ranking.

Like I said earlier, in practicality, this is what generally happens anyway. The benefits of such a crazy idea include (1) it would promote scheduling and importance of non-con games because conferences would have a concrete reason to try to perform better non-conference; (2) it would give coaches and fans greater certainty in knowing what they need to accomplish toward the end of the season to get in the field (“we need to be in the top 3 in our conference”); (3) even though it sounds complicated, it’s less complicated than what actually happens, and would simplify the number of discretionary decisions for the committee. This in turn, should decrease, relatively, any criticism by limiting the number of “bubble teams” truly up for debate; (4) when we know what conferences will have each number of bids, we can focus on talking about the teams that matter, rather than this damn bubble.

[photo via Getty]