SORTACUS and the College Football Playoffs
I've long had a fascination with the college football postseason, something I treated in detail a few years ago during BCS Week. I believe that playoff appearances should be determined by an algorithm to calculate which team is most deserving (based solely on wins and losses). This is done in nearly all sports, but due to the nature of college football it needs to be a more complex algorithm. After years building and examining ranking systems, I think SORTACUS is optimal for this task.
SORTACUS is an entirely algorithmic ranking system based solely on wins and losses. It relies on Bayesian statistics, using a prior rating and each game as an observations to calculate a posterior ratings distribution. I haven't given a full description yet, but last year I shared a few more details of SORTACUS.
It's important to note that SORTACUS is designed to rank teams in order of performance so far. In effect, this calculates which teams are most deserving of being in the playoffs. If you instead want to predict which teams will win in the future, check out my work with TeamRankings.
A weakness of this application of SORTACUS has always been it's ability to handle FCS teams. There are not enough connections between FBS and FCS teams for it to always recognize a difference between North Dakota State and Alabama. Because of this I've always ignored games against FCS teams, which can look weird when an FCS teams wins (looking at you Iowa!).
To try to get rid of this issue I have combined all FCS teams in to one giant FCS team. This team ends up going 10-103 and has a worse rating than all FBS teams. I think this is an okay approximation, although a better one would add decreased prior ratings for FCS teams and include them all individually. If we're being real though, that sounded like too much work.
As usual the experienced reader has skipped my chit chat and just jumped straight to these ratings. Below is the SORTACUS top 25 using games through Saturday. In theory, the top 4 teams are the most worthy of a playoff spot so far. Note that "Rating" corresponds roughly to how many points better than the FBS average a school has been.
And the top 4 is...
Exactly who everyone thinks it should be. Alabama has clearly run away with the regular season, and the other three one loss major conference teams all currently deserve a shot.
There are a few interesting insights above. First, take a look at Ohio State's rating. They are not just number 2, they are number 2 by a long ways. Their strength of schedule, highlighted by going 3-1 against top 10 teams, means that they deserve a shot at the playoffs, even over the eventual Big Ten champion.
While we're talking about the Big Ten, let's look at their second tier, the two loss trio. SORTACUS rates Michigan, Wisconsin, and Penn State as the top three major conference playoff replacements. The Big Ten has been a very strong conference this year, and any 2 loss team deserves playoff consideration if Clemson or Washington lose. The winner of the Big Ten championship may be the most deserving replacement team if 3 or 4 falters.
There is an outlier in the top 10: #5 Western Michigan. SORTACUS is very high on the Broncos (I had to look that up). I think there are two major reasons for this. First, like the FCS teams SORTACUS may not have enough data to appropriately judge the difference in Alabama and Western Michigan's schedules. Second though, there may be some degree of bias in the human rankings. Sure, Western Michigan is not likely the 5th best team in college football, but they may still be the 5th most deserving. It's hard to be even below average FBS squads.
How Is My Team Doing?
Schools of particular interest to the few may read this are below.
Iowa: 35th, rating of 7.21 (The Hawks are actually 13th if the loss to NDSU is excluded)
NC State: 58, 1.10
Maryland: 59, 0.66
Northwestern: 60, 0.52
Texas Tech: 83, -4.39
Purdue: 107, -10.20
Iowa St: 114, -11.87