The College Football Playoff Committee isn’t scheduled to release its first rankings of the season until a week from tonight. Our model, though, gives us the ability to estimate what those ratings will look like.
Our model uses a formula based on the previous five years of end-of-season CFP Rankings—the ones that determine which teams do and don’t make the playoff. The formula isn’t perfect. We don’t expect it to be. By gauging it each week against the real rankings, though, we can get an idea of which teams are getting the benefit of the doubt and which are being punished even more than we might expect for a soft schedule, a particular perceived flaw, or a specific loss.
Starting next week, we’ll be comparing our model’s expected rankings against the committee’s real rankings and adjusting our model accordingly, to continue to give you the best idea possible of which teams will and will not make the playoff. For now, though, we’ll simply give you a look at where our model thinks each team would land were rankings released tonight. The number to the right of each team is scaled such that the most highly-ranked team in the country is a 100.0 and the most lowly-ranked team (were these rankings to go through all FBS teams) is a 0.0. Commentary follows the rankings.
1. Ohio State (100.0)
2. Penn State (90.9)
3. Clemson (90.8)
4. Alabama (89.3)
5. LSU (89.3)
6. Baylor (83.0)
7. Auburn (81.6)
8. Florida (78.3)
9. Oklahoma (78.1)
10. Michigan (78.1)
11. Wisconsin (77.9)
12. Oregon (77.9)
13. Minnesota (77.1)
14. Utah (75.8)
15. Notre Dame (75.0)
16. Appalachian State (74.7)
17. SMU (73.5)
18. Iowa (72.5)
19. Cincinnati (72.5)
20. Georgia (72.1)
21. Kansas State (69.0)
22. Navy (67.8)
23. Indiana (66.9)
24. Texas A&M (66.8)
25. Texas (66.0)
Penn State?!
All year, our model has been bullish on the Nittany Lions relative to the media and betting markets. This gives some insight into why. Penn State has obliterated some teams, boosting how our model expects them to be perceived by objective mathematical rating systems and the subjective “eye test.” They also have road wins over Iowa (18th) and Michigan State (would be 38th), in addition to a home victory over Michigan (10th). That might not be on par with Ohio State’s résumé (blowouts of Wisconsin—10th, Cincinnati—19th, and Indiana—23rd) or LSU’s (wins over Auburn—7th, Florida—8th, and Texas—25th), but it’s good.
Penn State.
But. Look again at how close Penn State, Clemson, Alabama, and LSU are in that number on the right. It’s tight enough that even with only one of those four teams playing this weekend (Clemson, at home against Wofford), results elsewhere could shake the teams up. The bigger takeaway here should not be that Penn State is ranked second by our model’s expectations. It should be that these four teams are very much bunched in a group.
Take Baylor Seriously
The Bears are undefeated. They started the year unranked in the AP Poll, forcing them to play catch-up in that horse race, and they wouldn’t be favored against Oklahoma were the two to play today, holding them back in our model’s expectations of where the rankings will finish. Still, the Bears are undefeated and have some quality wins—at Kansas State (21st), at Oklahoma State (28th), home against Iowa State (30th). It isn’t the most impressive collection out there, but Baylor will be viewed differently than Minnesota.
For Now, Anyway
The Bears are still not favored against Oklahoma, and they’d likely have to beat the Sooners twice to make the playoff. Our model thinks the CFP Committee would like them right now. It does not think Baylor will finish the season ranked sixth.
Some Head-to-Head Notes
Our model doesn’t really evaluate head-to-head results. Yes, they’re one of the committee’s specific metrics if applicable, but history is scant on the subject, making it hard to measure the significance in a numerical sense. As with Penn State being ranked second, it’s probably better to focus on the numbers on the right than the actual rankings. With that being said, it’s possible Michigan will be ranked ahead of Wisconsin, and it’s possible a two-loss Auburn would be ranked ahead of a one-loss Florida. Thankfully for our credibility, Florida’s game with Georgia will shake things up anyway.
Georgia?
The Florida/Georgia game will also notably say something about Georgia. The Bulldogs are much lower in these estimated rankings than they are in the polls. It’s certainly odd that Notre Dame’s ahead of them, given the head-to-head result, but clearly, our model thinks the South Carolina loss is highly damaging. It’s worth nothing, though, that Georgia’s closer to Notre Dame in that number on the right than they are to Kansas State, despite being ranked five slots below the Irish and directly above the Wildcats.
Navy?
Navy’s in an interesting spot in that their win over Air Force is actually looking pretty good—Air Force is 31st in these rankings. Similarly, their loss to Memphis on the road is respectable—Memphis is 33rd. As with most anomalies in these rankings, it wouldn’t be surprising if our model is wrong. But it’s possible it’s right, and at a minimum, the truth is likely closer to the model than the AP Poll currently implies.
