Death to the At-Large Bid

MSGIn three weeks an era will end for many college sports fans. There is plenty of NCAA competition left after the Division I Men’s Basketball tournament is over, but for a large chunk of casual fans, the end of the tournament marks the end of the college sports season until football starts again in August.

When those fans come back, whether they know it or not, the NCAA will supposedly be revolutionized. Athletic directors will have more authority. Student-athletes may have a more prominent voice as well as a real vote. Most notably, the largest and richest conferences may have autonomy to pass their own rules on some issues, most notably cost-of-attendance scholarships and miscellaneous benefits to athletes.

But that change is doomed to failure, just like the original round of reform was coming out of the Presidential Retreat in 2011. And one thing is holding the NCAA back from making this change. If the coming governance changes are truly going to be a revolution, then like all revolutions it needs a battle cry. And that cry should be “Death to the At-Large Bid”.

Why the At-Large Bid Needs to Go

Last year, after already suffering the defeat of the miscellaneous expense allowance the year prior, NCAA President Mark Emmert used his State of the Association address to once again push for more reform:

Using his annual State of the Association address to both applaud the NCAA’s century of progress and yet challenge administrators to aspire for even higher standards, Emmert posed a set of commitments necessary to sustain the collegiate model, not the least of which is a values-based approach to legislation rather than pages of competitive-equity rules that in the end are unenforceable and do not support student-athlete success.

One of the principles Emmert promoted, and has since the Presidential Retreat, is fair competition. Unlike competitive equity, fair competition recognizes the advantages some institutions have over others, especially in their ability to generate revenue. Focusing on fair competition rather than trying to enforce competitive equity across the entire Division I landscape would allow the NCAA to adopt rules that some conferences and institutions could afford but others could not.

This was pitched as being acceptable because intra-conference competition should be the focus. All 340-plus Division I members are not supposed to be trying to compete with all the others all the time. And the smallest are not supposed to be competing with the largest day-in and day-out. Instead, fair competition pushed the idea that conference should compete amongst themselves to get into the NCAA tournament where they get their shot at each other, often in an inherently unpredictable single-elimination bracket format.

But one feature of NCAA tournaments binds all 32 Division I conferences together: the at-large bid. In all sports, including football, some portion of the tournament field is comprised of at-large bids. At-large bids are baked into the NCAA rules as well. Once a team sport tournament reaches a certain time, it has to be at least 50% at-large bids. In men’s basketball, an NCAA rule guarantees 34 at-large bids.

If a conference decides to stop competing against other conferences, especially those with more prominent members and bigger budgets, it starts to forfeit a chance at those at-large bids. Taken to the logical conclusion in men’s basketball, eventually each power conference programs would be fighting for one of 37 spots (conference automatic qualification plus all 36 at-large bids) while every other team has only one chance to make the field.

This applies to most team sports, but men’s basketball throws it into the sharpest relief. Beyond the benefits of exposure in the NCAA’s most popular event, at-large bids are worth a substantial amount of money. Each at-large bid guarantees a conference another unit for the NCAA’s basketball revenue distribution, plus the chance to earn more. Starting in 2013–14, each unit will be worth a minimum of $1.5 million over its six-year lifetime.

For the Division I’s middle class, this is important income. The power conferences get big distributions (over $20 million to the Big Ten and Big 12 in 2012–13) but have even bigger television contracts. The low-majors generally get their one unit each year and budget accordingly. It’s conferences like the Mountain West, WCC, American, A–10, and new Big East that can make significant gains in revenue by amassing basketball units.

As long as conferences are fighting it out in a zero-sum game for at-large bids, there is only so far they can go playing by different rules. To start to disconnect conferences from each other and allow a wider range of rules that fit the vastly different institutions that make up Division I, the NCAA needs something different, especially for its signature competition.

Enter Coefficient Rankings

Instead of coming up with selection criteria and meeting in a room to pick the best 36 teams, the NCAA should award spots based on a conference’s performance in previous NCAA tournaments. This is similar to how UEFA decides how many spots each country gets in the UEFA Champions League and Europa League. The NCAA already calculates tournament performance over time, in order to distribute revenue from the basketball fund.

From that basic idea, three details would have to be sorted:

  1. Would advancing to later rounds be worth more? If so, how much more?
  2. Would there be a cap on the number of teams from any one conference?
  3. How far back would tournament performance be tracked?

Giving a bonus for later rounds, setting a conference maximum, and shorter period (say the previous two or three years vs. the previous four to six) would lead to more volatility, but might track current quality more closely. Conversely no bonus for later rounds, no conference maximum and a longer period both means more stability (i.e. conferences will generally have the same numbers of bids) and a bigger advantage to conferences which get more teams into the tournament.

The obvious downside is that such a system does not take into account the current year. If say the WCC has two bids and is exceptionally strong, with four teams deserving of an at-large bid, two of those teams will be out of luck. To some degree, a conference’s access in tournament will also always be based on student-athletes, coaches, even whole institutions that are no longer there.

On the flip side though, it does make the process of selection much easier. Every team goes into the season knowing exactly what it needs to do to get into the tournament. It also increases the value of the conference regular season, at least in multi-bid conferences.

Such a system would not require any other change to the college basketball calendar. The nonconference season would still be important for seeding and preparation purposes. A team could punt the nonconference, but would risk a much lower seed. Conference tournaments could also be used to award bids and likely still would, especially in single-bid conferences.

How It Looks in 2014

So how would this play out this year?

To calculate the number of bids each conference has earned, I used the last three years of NCAA tournament performance. That would allow an individual player or recruiting class to effect the coefficient and benefit from it. No bonuses were awarded for later rounds, each game was worth one unit except for the championship game, just like revenue distribution. There was also no cap on the maximum number of bids for any one conference, although this created an obvious problem this year.

This is a particularly complicated time for such a system like this given conference realignment and the split of the Big East and American Athletic Conferences. For standard realignment, the units stayed with the conference where they were earned. For the Big East/American, each conference was given the units of its current members. The units earned by departed members of the old Big East (Notre Dame, Syracuse, West Virginia, and Pittsburgh) were forfeited and were not used in the calculations.

Each conference’s units from the last three years were added up and divided by the total to get a percentage. That percentage was multiplied by 68 to get a number of bids for each conference. Any conference which had earned less than one bid was rounded up to one. Any conference which had earned more than one bid was rounded down to the previous whole number. That resulted in 65 bids awarded. Three conferences (Missouri Valley, Pac–12, WCC) finished within X.99 bids, so they were rounded up.

That leaves us with this breakdown of bids by conference:

  • Big Ten: 9 (!)
  • ACC: 5
  • Big 12: 5
  • SEC: 5
  • American: 4
  • Atlantic 10: 4
  • Pac–12: 4
  • Big East: 3
  • Mountain West: 3
  • Colonial: 2
  • Missouri Valley: 2
  • WCC: 2
  • All other conferences: 1

A couple things stand out, starting with the Big Ten’s nine guaranteed bids. Part of that comes from being second in units over the last three years behind the old, combined Big East, which split some of its units and forfeited a lot that Syracuse had amassed, plus four second (third according to the NCAA) appearances in 2011. The A–10 and Colonial are also trading on the performance of past members, but given the quality of the A–10 this year, they actually get penalized some.

In this sort of system, the bids are the conference’s to award, subject to whatever limitations the NCAA may impose. The assumption was that the first bid in each conference would go to the conference tournament winner if the conference held one. Any other bids a conference had were awarded in order of regular season finish.

The combination of awarding bids based on past tournament performance and using just the conference regular season resulted in nine changes from the 2014 field:

In:
– SMU
– Minnesota
– Illinois
– Indiana
– Towson
– Indiana State
– UNLV
– Georgia
– Arkansas

Out:
– Memphis
– Dayton
– UMass
– NC State
– Baylor
– Oklahoma State
– Xavier
– Arizona State
– Stanford

In this case, some teams which were not on the bubble (on the right side) were replaced with some teams which also were not on the bubble (on the wrong side). One way to fix this would be to limit the Big Ten’s bids, for example to six so a team had to finish in the top half of the conference to go to the tournament. The next three conferences in line to be rounded up are the American, ACC, and the Big 12. That would put Memphis, NC State, and Baylor back in the field at the expense of the three Big Ten teams.

The other school of thought would be to let the Big Ten send its 7th–9th place teams, all of which would receive double digit seeds, and see what happens. If those teams get hot and make a run, it would consolidate the Big Ten’s nine bids. On the other hand, if they really do not belong in the field, they will lose early and the Big Ten will get minimal benefit from having them in the tournament.

Conclusion

There are a lot of unanswered questions here. Whether this makes for a better or worse basketball tournament is one. Whether it makes for a more interesting basketball season is another. And whether the details could be sorted out and something like this could ever happen is a third.

But it accomplish a more important goal which is to reduce the need for conferences to compete against each other all the time. There is still competition for seeding. And revenue still provides an advantage in a single-elimination tournament. But as long as the NCAA does not find a replacement for the at-large bid, it can never truly escape the shackles of competitive equity.

 

Posted on by John Infante
This entry was posted in Bylaw Blog, Bylaws. Bookmark the permalink.
Join the #1 RECRUITING NETWORK

3 Responses to Death to the At-Large Bid

  1. Sam says:

    First off, I love ideas like this. I doubt it will ever happen, and I’m not sure if it’s a net benefit, but I like seeing them kicked around.

    A few questions:
    First, what about the ways in which basketball and soccer are pretty different? A champions league type system would move conference tiebreakers from something nobody cares about into the deciding factor in determining berth in the NCAA tournament. So, that’s head-to-head records, which while people may know of them, they aren’t as prominent as goal differential in soccer standings. More importantly though, most soccer leagues play double(or quadruple, in smaller leagues) round-robin schedules, where everybody plays everybody home and away. In college basketball, a twelve or fourteen team conference playing an eighteen game schedule will have significant scheduling asymmetries. In theory, the selection committee can take imbalanced conference schedules into account, while the champions league format presumably wouldn’t.

    Second, I’m wondering why you used raw wins in determining the coefficients, rather than something like what UEFA does, which is basically wins divided by teams in tournament. I don’t know if this would have produced different results for this year, but it would presumably be “self-correcting” going forwards. If the Big Ten gets 9 bids this year, their coefficient would probably go down. Let’s say, for the sake of an example, that the six Big Ten teams will win twelve games in the tournament. And let’s also say that adding Illinois, Indiana, and Minnesota will result in one more win. Under the “raw wins” system, this would move the Big Ten coefficient for this year from 12 to 13, while under a UEFA-like “average wins” system it would go from 2 to 1.44. In fact, this is the danger of the raw wins system, that it creates a snowball effect, where more teams get in, resulting in more wins, necessitating a “top half of conference only” condition.

    I hope this doesn’t come across as hostile to the idea, I think it’s an interesting idea(though probably one that will never be implemented) and would like to see it clarified. At the very least, we can come up with terms for creating hypothetical brackets in future years.

    • John Infante says:

      1) That’s up to the conference really. The incentive is to send your best teams to the tournament so they earn more spots or keep the ones you have. If an unbalanced schedule means you routinely send bad teams to the tournament, that will hurt you in the long run.

      2) I did not do anything more complicated that straight NCAA units because that was enough math for me already.

  2. Larry says:

    I feel like this puts the Cinderella or underdog teams behind the eight ball even more than they already are. You say use conference tournaments to determine single bid conferences but that totally devalues the regular season. It seems like under this system the conference tournaments would have to be eliminated… Which or course maybe wouldn’t be such a bad thing.

Leave a Comment