Virginia Commonwealth and Butler are not Equal to Northern Illinois and Air Force
By Jason Lisk
I’m all for fanciful ideas and interesting thoughts. I just don’t think we should be extrapolating to conclusions that the author, and the title, want to make. First, no one, and I mean literally no one has advocated for a 64 team tournament that I am aware of. Sure, at least that many teams play in bowl games traditionally*, but there is no one out there pushing for over half of the division I teams to be part of a playoff.
Which brings us to my next issue with this exercise. It assumes the likelihood of an upset for the 44th best football team versus one of the best teams in the country is the same in football as basketball. It’s not even close. There are almost 3 times as many Division I basketball teams, and the tournament field still represents the top 15-20% of basketball teams. Doing a 1 for 1 comparison doesn’t give you teams at the same level of achievement. Teams like Butler and VCU still represent the extreme upper tier of programs from non-BCS conferences. The difference between Kansas and Virginia Commonwealth was large, no doubt, but not as large as it would be in football. Ken Pomeroy had that Southwest Regional Final game at 88% in favor of Kansas before the tip, which is pretty close to being in line to historical averages of a #1 seed versus a team seeded between 8 and 12.
Let’s go with that for now, a 12% chance of losing. Who would Oregon or Auburn have a 12% chance of losing to? It wouldn’t be Northern Illinois or Air Force, teams that ranked in the 30’s and 40’s respectively. Just over the last 3 seasons, the top 5 teams by simple rating system finished 17-0 against teams ranked 31-40, and 17-0 against teams ranked 41-50. Only one top 5 team over that span had one of their losses to a team eventually ranked below 21. It’s late and I don’t have time to look further, but my intuition and observation tells me that the 40th best college football team doesn’t win anywhere close to 12% of the time against a team that is top 5 at the end of the regular season, whereas it is far rarer to see the 40th best college basketball team beat a #1 seed in the regular season.
No, the similar comparable to Kansas losing to VCU, in terms of likelihood of upset, would be somewhere closer to the 16th-18th best team in college football beating the 3rd best in a bowl game, still unlikely but in the realm of real possibility. If you want a better comparable for VCU or Butler, in terms of true probability of beating one of the big teams, it’s more likely 13-1 WAC Champion Nevada. If we scaled the basketball tournament to the football field, we would get about a 20 team tournament in football, and a team like Nevada surviving to the semifinals would be the equivalent of what we have seen in this wacky basketball tournament.
Earlier this week, I looked at how frequently the best teams in football and basketball won the national championship, and hypothesized that, if we wanted to really maximize how frequently the true “best” team won each sport, we would reduce the playoff in basketball (to something like 16-24 teams) but increase football (to 4-6 teams).
Some people, such as Dan Wetzel, do advocate for a 16 team playoff with every conference getting a bid, like basketball. That would make it similar to basketball in terms of the percentage of time a true underdog gets through (about once every 10 years or so). I wouldn’t go crowning Northern Illinois or Air Force as national champions any time soon though. And I certainly wouldn’t extrapolate the results of a basketball tournament to a 64 team football field and blindly copy in the results.
*a tradition since 2008
[photo via Getty]