Which teams should have been most upset at the NCAA?
Compared the rest of the NCAA, big-time college football decides who wins the national championship in a truly bizarre fashion. Rather than a full-blown playoff (like the lower-level “Football Championship” division), they heavily rely on polls and a small panel of experts to decide whom the top four teams are, who then have a small playoff, while the remaining teams with winning records play in bowl games. By small, I mean there are over 100 teams in the Football Bowl Division, but only four teams have the opportunity to play for a championship.
I believe most fans would say this is an improvement over the previous arrangement, where a combination of computers and polls would decide who the top two teams were, who’d then play for the championship. Prior to 1998, there was no such agreement in place, and there were several “split championships” where two teams could make a case that they were the best team in college football.
Regardless, the current playoff format helps adds a lot of clarity, but even with a four-team playoff, there are teams (and fanbases) who find themselves on the outside looking in.
The biggest contention in these arguments is usually the strength of their respective team’s schedule compared to others with a similar record. For example, this year, UCF had a perfect record, including a bowl win against Auburn, who beat two of the four teams (Georgia and Alabama) that made into the four-team playoff. The knock on UCF? Their overall schedule was relatively weak.
With this in mind, I created a scatterplot from the Sagarin Ratings and the win/loss record for Football Bowl Division teams (Figure 1).1 I was mostly interested in teams that were at the periphery of the point cloud. The teams that were closer to the top right were teams that performed well against hard schedules, whereas the teams at the bottom left were teams that performed poorly despite easy schedules. Perhaps more germane to this debate, the teams that were closer to the top left corner were teams who won most of their games but against an easy schedule. Finally, on the bottom right were teams with bad or mediocre records, but that record came against stiff competition.
Out of curiosity, I posted Figure 1 to the college football subreddit and fully expected to see the UCF fans crying foul over being left out of the playoff. Then, I expected counter arguments from Alabama fans defending their inclusion as the fourth team in the playoff based on strength of schedule. Despite losing a game, Alabama had a harder schedule.
What I didn’t expect were the Wisconsin fans. Their argument? They had comparable strength of schedule to Alabama, they lost only one game like Alabama, and that one game they lost was a conference championship. As a result, they had a better win percentage than Alabama. Or, as one commenter stated, “Wisconsin: Better record. Better schedule. Better loss. Alabama: Better brand.”
Alabama mauled last year’s champion Clemson and clawed their way to victory over the best Georgia team since Hershel Walker played.
In retrospect, I have a hard time saying that Alabama should have been left out of the playoff, but I also sympathize with Wisconsin fans. However, that discrepancy between the Massey Composite Ranking and the Final AP Poll illustrate that Wisconsin was heavily penalized for their loss to Ohio State, perhaps fairly or not.
If anything, this year’s “Wisconsin Problem” further illustrates the need for an eight-team playoff, and perhaps one where a panel doesn’t decide the seeding that weights losses in championship games more heavily than those in the regular season.
1 I used the “rating” instead of the “ranking” because the rating is a continuous variable, and is more appropriate to use in a basic scatterplot. The Sagarin Ratings were downloaded on Dec. 10th, before the start of the bowl season.