The Initial College Football Playoff Ranking Is Out, Does The Selection Committee Suffer From Confirmation Bias?
This weeks post from guest blogger Renwei Chung focuses the spotlight on college football and the role that confirmation bias may, or may not, have played in the recently announced College Football Playoff Ranking. Renwei attends SMU Dedman School of Law. He has an undergraduate degree from Michigan State University and a MBA from the University of Chicago. He is passionate about writing, psychology, and economics. You can contact Renwei by email at projectrenwei@gmail.com, follow him on Twitter (@renweichung), or connect with him on LinkedIn.
The first College Football Playoff Ranking was announced this week and of course it’s a little controversial. This is what makes sports so fun, we can talk incessantly about it and there is no bullet proof logic or absolute truth.
The college football ranking is just as much an art as a science. It was ultimately compiled by the Selection Committee using statistics, polls, and like anything else, some bias. The committee has stated that it doesn’t look at what might happen, rather, it analyzes what has happened. In other words, the initial college football ranking is meant to measure a team’s performance over its potential.
Let’s analyze the Selection Committee’s logic in choosing Alabama as a top four team, even though it has already suffered a loss this season. Most analysts believe Alabama will be in the college football playoff at the end of the season. After all, Alabama’s football program is the gold standard in which all other programs measure themselves against. Most fans also believe Alabama’s football team will be in the final four at the end of the season.
But has the Committee let confirmation bias cloud their judgment? In other words, did the Committee’s choice of Alabama in the initial ranking because it presumptively believed it would be in the final ranking ultimately create an unfair playing field?
Recently, I’ve had the opportunity to work with GapJumpers, a platform that allows companies to conduct blind auditions and overcome bias to hire their employees. This blind audition is meant to measure both performance and potential.
I wonder how a blind process would affect the Committee’s rankings. In other words, if Alabama’s brand and name plate was removed from its regular season resume would it still be ranked so high?
If every team was judged purely by its performance on the field and strength of schedule, would some undefeated teams like TCU or Baylor be ranked higher?
Many people like to cite the one-loss Ohio State team that climbed to the final four ranking last year as proof that the committee ultimately gets it right based on the merits. However, even after Alabama suffered one early loss last season, it was still ranked sixth in the initial poll. In other words, it only needed to climb two spots to reach the 2014 college football playoffs while Ohio State had to climb twelve spots. Of course, last year’s season was much different than this year’s season.
So what if we viewed this season through a “blind” lens like GapJumpers implements, would teams like Alabama be ranked so high? USA Today’s Mike Foss highlights five statements made by Jeff Long, the Committee’s Chair:
1. “Alabama, from our point of view, had a stronger schedule in the games that they’ve won. They’ve got three wins against teams with better than .500 records.”
2. “How [Alabama] won those games that they won against quality opponents, I think that sends a strong message of why we ranked Alabama higher than Florida.”
3. “Our committee’s charge is to go deeper. We always look at strength of schedule, and this was discussed in great depth over the last two days.”
4. “You have to look deeper than those three wins against teams with records better than .500.”
5. “Losses against a quality opponent were also taken into account.”
Using the Selection Committee’s logic, Foss provides several counterpoints:
1. If the committee picked Alabama because of their three wins against teams with .500 or better records, why didn’t Michigan State make it in instead? The Spartans have five wins over .500 teams, and also haven’t lost.
2. Alabama beat Georgia 38–10 in Athens. A quality win. Florida beat Georgia as well, 27–3, at a neutral site. If that was the one comparable result, then fine, Alabama over Florida is a tough but understandable decision. But Florida crushed No. 18 Ole Miss — the team responsible for Alabama’s loss — 38–10. They also have comparable strength of schedules. Alabama is No. 9 on Jeff Sagarin’s college football ratings list, while Florida is 15th.
3. If strength of schedule is really a factor in the committee’s decision-making, then Baylor has absolutely no business at No. 6. Baylor’s strength of schedule is 104th in the country. 104! 104! 104! Get outta here, Jeff Long.
4. Michigan State and Utah both have five wins over team’s with winning records. Both teams have more wins over .500 teams than anyone else also have glaring strikes against them. Stanford has four wins against teams .500 or better.
5. Florida lost to No. 2 LSU on the road 35–28. Notre Dame lost to No. 1 Clemson on the road 24–22. Alabama lost to No. 18 Ole Miss at home 43–37. Which team has the worst loss?
Given the Committee’s statements and logic, do you believe the Committee got the initial college football ranking right? One would be hard-pressed to find a college football fan who doesn’t believe Alabama will ultimately be in the college football’s final four ranking. In fact, given the rigors of their regular season schedule in the SEC, it would be hard to argue that one loss would prevent them from making the playoffs. But in week nine, without the name “Alabama,” has Nick Saban’s team done enough to be ranked in the top four?
Do you believe the Selection Committee let confirmation bias cloud their judgment? Admittedly, some bias is always present in the decision-making process. But as Long stated, “the Committee’s charge is to go deeper.”
Unfortunately, I don’t believe the Committee went deep enough to avoid the inherent biases that tend to affect an organization’s selection process.
