A New Plan – How to fix the NCAA Tournament

I recently had an epiphany where I thought that I could take all of the data from the regional ranked data sheets, put them in Excel and it would magically spit out the strongest region. In the end, after a lot of effort, I had a lot of data with no real purpose. I’m going to use some of this data below but not in the way I initially envisioned.

When I set this data aside I had time to think and eventually my thoughts turned, as it often does, to the field of 64. There are 19 or 20 at-large bids each year that the NCAA selection committee messes up. They certainly don’t mess up all of the at-large bids but there is usually anywhere between 2 to 5 teams that don’t make the tournament that probably should have based on the selection criteria and just plain common sense. The current method of selection is really not constructed to fix this problem and this got me thinking of a different solution.

The main issue with the NCAA Tournament, and this may be true for all divisions across all sports, is that we have the Automatic Qualifier (AQ) conferences. These conferences get to determine one of the teams that will participate in the NCAA Tournament when a number of conferences really don’t produce a team that is NCAA Tournament worthy. I started thinking what would happen if every region was responsible for sending their best eight teams (sort of) to the tournament while maintaining the automatic qualifier conference dynamic.

Here’s the quick overview of my plan. The conference tournaments get played out and each region determines their top 8 teams as they do today with the regional rankings. If the top 8 teams fall out correctly then the region is done and they have their representatives. Most likely this will not be the case so at this point regional play-in matches will need to be scheduled. The New England Region has 8 AQ conferences so worst case is that none of the top 8 teams in the region win their conference. Regardless of the number of teams, the regional play-in matches will pit the regionally ranked teams that did not win a conference against the teams that were not regionally ranked but did win an AQ conference bid. The matches would be played at the AQ conference winning school. Remember that in every region except the West, travel can be accomplished through driving. (That last statement isn’t entirely true but don’t you fret about it.) My plan means an extra week in the schedule as the end-of-the-year order would now go conference tournaments, regional play-in and then NCAA Tournament. The extra week to the season can be picked up through an earlier start or more likely the elimination of a week during the season. I would argue that this isn’t a deal breaker because now we are truly a regional sport and the need to play outside of the region is eliminated during the season.

I want to put this plan into action and I will but let’s discuss some of the problems first. This plan is certainly not fair to the AQ conference winners that are not regionally ranked. True, but the current system is not fair to the teams that are better. No additional travel is placed upon the AQ teams and they still have the advantage of a home match. If you can’t beat a team coming into your home then how were you going to beat a better team (in theory) at a neutral site? Some of the matches could involve an additional flight and I would put this cost on the school. My response to this is if the school doesn’t want to bare this cost then they don’t have to play the match and the AQ school retains their bid. With the current system the regional ranked team that has to fly to the play-in match is most likely already done and they can certainly retain the status quo if they choose to do so. The proposed plan benefits the smaller regions with fewer AQ conferences than the larger regions. It also hurts the larger regions because the regional bids are capped at 8 and in the current system the larger regions almost always exceed this number. I can’t really argue the numbers on this point but I would remind the Northeast that they have a number of AQ conferences that really don’t deserve an NCAA Tournament bid. Because they have more of these conferences, they now have more opportunities to get teams in that would have never sniffed the NCAA Tournament under the current system. Where I really see a problem with this proposed plan is if you had a large region with the top 7 or 8 conferences in the nation. We don’t have anything like that today and I would argue if this developed over time that I would lead the charge to change to a different system. I mentioned the lost week of play in-season as being a potential problem. My thinking is that a number of schools use a weekend to play out-of-region and these types of matches really don’t improve their regionally rank that much in this proposed plan. It would be much better to beat the number 6 regionally ranked team in-region then taking down Emory, WUSTL and Calvin one weekend. (Okay, maybe not but you see my point that in-region wins against regional ranked teams are now vital to any team wanted to get an NCAA Tournament bid.) An offshoot of this is that it does make our sport more regionally based but I would argue that is what the NCAA is forcing us to do now.

Some of the fringe benefits from the new plan would be a reduction in travel costs for teams that normally go out-of-region during the season. I still believe the elite teams with aspirations of a National Championship will still travel because of how it prepares their teams but for the most part other teams will not. Another benefit is that this will really ramp up the regional rivalries and make in-regional matches even more important than they are in the current system. I can also see academics receiving a benefit because most of the teams will have a full week and weekend after the conference tournaments with no matches giving them time for practice and school work. Maybe the best benefit to the fan is that the best 8 teams from a region will play in the NCAA Tournament.

Based on the results from last season, let’s see how this proposed plan would have played out. For each region I have listed the AQ conference teams that would be forced to host a regional play-in match. I have also listed the ranked teams that would play in those matches. Finally, I have listed teams that would have received a bid without the need for a play-in and teams unaffected by this new plan.

Central’s AQ conference winners that would host – Dubuque, Greenville and Minnesota-Morris.
Central’s Regional Ranked teams to play – Bethel, Wartburg and Saint Olaf*.
Central’s NEW NCAA teams – None
Central’s unaffected NCAA teams – Gust. Adolphus, WUSTL, Saint Benedict, Saint Thomas and Augsburg.

*The new plan would definitely have to take into account distance as well as rankings when making the pairings.

Great Lakes’ AQ conference winners that would host – Rose-Hulman and Bethany
Great Lakes’ Regional Ranked teams to play – Otterbein and Baldwin Wallace*.
Great Lakes’ NEW NCAA teams – Muskingum
Great Lakes’ unaffected NCAA teams – Calvin, Wittenberg, Hope, Ohio Northern and DePauw.

*Baldwin Wallace was most likely the 9th ranked team as Thomas More received a ranking but also a Pool B bid so would not be considered here.

Mid-Atlantic’s AQ conference winners that would host – Pittsburg-Bradford, Notre Dame and Stevenson.
Mid-Atlantic’s Regional Ranked teams to play – Franklin & Marshall, Mary Washington and Susquehanna.
Mid-Atlantic’s NEW NCAA teams – None*
Mid-Atlantic’s unaffected NCAA teams – Juniata, Chris Newport, Carnegie Mellon, Eastern and Johns Hopkins**.

*The Mid-Atlantic Region would drop from 10 bids (not including Pool B Marymount) down to 8 bids under the new plan. (As a counter point, I would say that 11 teams are now considered for those 8 bids.)

**Johns Hopkins was the 9th ranked team but was considered since Marymount received the Pool B bid moving JHU into the top 8 at that point.

Midwest’s AQ conference winners that would host – Cornell.
Midwest’s Regional Ranked teams to play – UW Stevens Point.
Midwest’s NEW NCAA teams – Concordia Wisconsin.
Midwest’s unaffected NCAA teams – Chicago, UW Eau Claire, IWU, UW Whitewater, Aurora and Carthage.

New England’s AQ conference winners that would host – Wentworth, E. Connecticut State, Worcester State, Southern Vermont and Maine Maritime.
New England’s Regional Ranked teams to play – Wesleyan, Wellesley, Springfield, MIT and Amherst.
New England’s NEW NCAA teams – None*
New England’s unaffected NCAA teams – JWU (RI), Bowdoin and Babson.

*The New England Region would drop from 11 bids down to 8 bids. On the flip side, 13 teams would be in consideration for those 8 bids.

New York’s AQ conference winners that would host – Nazareth, Kean, Morrisville State and Mount Saint Vincent.
New York’s Regional Ranked teams to play – Stevens, Vassar, Stockton and Clarkson.
New York’s NEW NCAA teams – None
New York’s unaffected NCAA teams – Ithaca, RIT, SUNY New Paltz and Hunter.

South’s AQ conference winners that would host – Meredith.
South’s Regional Ranked teams to play – Bridgewater.
South’s NEW NCAA teams – Hendrix, Sewanee and Randolph-Macon.
South’s unaffected NCAA teams – Berry, Emory, Washington & Lee and Birmingham Southern.

West’s AQ conference winners that would host – Pacific Lutheran.
West’s Regional Ranked teams to play – Whittier*.
West’s NEW NCAA teams – La Verne, UT-Dallas.
West’s unaffected NCAA teams – CMS, Colorado College, UMHB, Trinity and Cal Lutheran.

*We have a flight! Whittier would have to fly to play Pacific Lutheran. (Although, knowing how close that 8th spot was I can almost guarantee in the new plan that Pacific Lutheran would have been the 8th seed.)

That’s a lot to digest, I know. There is still a problem with my little plan because I didn’t account for the two Pool B teams in my final count of 64. I took away 2 bids from the Mid-Atlantic Region and 3 bids from the New England Region for 5 total but added 7 at-large bids that didn’t require a play-in match. To compensate for the Pool B bids, I would eliminate the weakest 8th ranked teams, which pretty much gives the Selection Committee the only thing to worry about at this point with the new plan. For the sake of argument, let’s say it’s Baldwin Wallace (Great Lakes) and Bridgewater (South). (I didn’t feel like picking on the New York Region here.) That means Muskingum would have a play-in match as would Randolph-Macon as opposed to a free at-large pass. There’s my field of 64…err…84!

So with 84 teams that means we have 20 regional play-in matches so we can eliminate 20 more teams to get us to the proper 64 count. We’ve added 15 teams that will get a chance to prove they belong in the NCAA Tournament that were previously eliminated and we’ve given 5 teams at-large bids without the need of a play-in. We’ve eliminated all of the cross-region arguments that takes the Selection Committee all night and most of the morning to resolve and left them with who gets removed for the Pool B teams. We’ve given them the best 8 teams per region which makes the regionals easier to organize and with all of their extra time they could even mix some things up a bit to try to even up the brackets. With that said, another benefit of doing this new plan is that the brackets start becoming fairer and there really isn’t a need to ship teams around. Would I still try to swap some teams around to strength the New York Region? Yup, but I’m not going to cry (much) if they don’t.

The bottom line with this plan is that the bigger regions are hurt more than the smaller regions that actually benefit a great deal in some cases. I would argue that the bigger regions are already hurting their own regional ranked teams with so many AQ conferences and this new plan actually gives more teams a final chance to play into the tournament. With that said, there is a push to move to 10 regions which would basically mean taking the Mid-Atlantic, New England and New York Regions and expanding them to 5 total regions. You could still use this plan but the math will differ slightly. It would mean more bids would go to these three (now five region) as they would fill up half of the tournament. I would probably give 7 bids to the existing 5 regions and let the new 5 regions split the remaining 29 bids, which would mimic the current system (although this still doesn’t account for the Pool B bids). Over time, adjust the available bids per region based on the competition.

Way back at the beginning of this article I promised you data so here you go. Some of this data allowed me to rationalize taking bids away from the larger regions.

I only have the last two years of regional ranking datasheets but I tried to determine some averages and where the regions fell around that number. From this data I determined that the average SOS across the nation in DIII in 2017 was 0.510 and in 2018 it was 0.509. Why aren’t these 0.500? I have no clue but I think it’s just a function on how SOS is calculated. The only regions that had a higher than average SOS for the last two years were the Great Lakes, the Mid-Atlantic and the West. That means that only these three regions played a harder than average schedule as a whole. The only regions to have a better than average winning percentage over the last two years were the Central, the Great Lakes and the West. This is really a show of depth in the region but also indicates that these regions did go out-of-region and pick up some wins. When you couple the Great Lakes and the West being better than the average SOS, it means they were playing tougher teams in other regions. I then looked at the average ranked win percentages over the last two years and the only regions to exceed these numbers were the Central, the Mid-Atlantic, the Midwest and the West. I’m not entirely sure what this number actually proves because the larger regions can rank more teams. Still, it’s probably another measure of depth in the region and means some quality wins outside the region had to occur.

Some of the weaker numbers I found were produced by the New York Region, which isn’t too surprising if you follow DIII volleyball. As a region they were nowhere close in SOS when compared to the other 7 regions. Their ranked wins numbers were also disastrous. These are clear indications of a poorly performing region and a region that benefits the other regions around them that come in and pick up wins. They also finished last in winning percentage in 2017 and just behind the South Region in 2018. As a group, the South Region had a pretty bad 2018 while at the same time improving their ranked wins percentage. (No idea what that means statistically.) Here is a rank of the regions from 2018 and 2017 based on winning percentage and SOS (summed):

Region Pct SOS Total
Great Lakes 2018 0.5351 0.5321 1.0672
West 2018 0.5114 0.5359 1.0472
New England 2018 0.5064 0.5087 1.0152
Central 2018 0.5100 0.5044 1.0144
Mid Atlantic 2018 0.4974 0.5095 1.0069
Midwest 2018 0.4923 0.5066 0.9989
South 2018 0.4713 0.5005 0.9717
New York 2018 0.4774 0.4873 0.9647


Region Pct SOS Total
Great Lakes 2017 0.5498 0.5442 1.0940
West 2017 0.5088 0.5358 1.0445
Mid Atlantic 2017 0.5156 0.5160 1.0316
Midwest 2017 0.4996 0.5196 1.0192
Central 2017 0.5008 0.5059 1.0067
South 2017 0.4864 0.5087 0.9951
New England 2017 0.4927 0.5003 0.9929
New York 2017 0.4622 0.4800 0.9422

As I said at the beginning, I’m not sure what all this means but I’ll keep tracking. I think you can probably make a case from the numbers that the Great Lakes and the West Regions are the strongest and that the New York and the South are the weakest. Keep in mind that this considers the entire region and not just the NCAA Tournament teams.

Back to the plan and does it have any chance to be considered and possibly implemented? Hells no. Do I think it gets us closer to a tournament that includes the best teams? Hells yes. In the end, putting the best teams in the tournament really isn’t a goal for the NCAA. It’s what the fans want and it’s what is fair but in the end that has meant nothing to the NCAA. I think the larger goal of this article is that we need to start thinking outside of the selection criteria box as a way to move us to a better tournament. We also need to start figuring out ways to quantify the quality of our regions and allowing those results to filter into the at-large selections. I’m in no way confident a suitable solution is out there but on the flip side it just means I’ll continue to have more things to write about in the future.



One thought on “A New Plan – How to fix the NCAA Tournament

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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