Alabama Coach Nick Saban during Saturday’s loss to Auburn. (AP Photo/Brynn Anderson)

Cade Massey, a practice professor at the Wharton School, and Rufus Peabody, a Washington-based sports analyst, developed this ranking system for projecting future performance. Ratings represent a team’s predicted point differential against an average team on a neutral field. Current season statistics are adjusted for home field, opponent and game situation, blended with preseason expectations and weighted by their predictive ability.

College football’s rivalry week yielded some stunning upsets of Alabama and Miami, but these and seven other teams ended the weekend in much the same way they started: in a chaotic mix of playoff contenders.

By playing itself out of the SEC championship game, Alabama lost control of its own destiny, but our model still gives the Tide a better than decent chance of securing a spot in the final four, if the conference championship games fall as we predict. Of course, they rarely do. And an even greater uncertainty is how the selection committee will react as the contenders emerge, scathed or spotless, from the big games.

Here’s what we see at the intersection of football uncertainty and committee uncertainty as we enter the decisive weekend.

Football Uncertainty

We hope to be wrong, but outside of the SEC, we believe the conference title games will be relatively lopsided. But play enough games, and surprises will occur.

SEC: Georgia (No. 3 in our ratings) vs. Auburn (No. 5) is the clearest win-and-you’re in game, and it should also provide the weekend’s best matchup. In our ratings, the teams are separated by just .04 (with Clemson wedged between them). Will Auburn’s prolific running game rebound from this weekend’s injuries? Will a de facto home field advantage boost Georgia?
Our forecast: Pick ’em.

Big Ten: Ohio State (No. 2) vs. Wisconsin (No. 9). The Badgers feasted on weak opponents all season, while the Buckeyes endured numerous back-alley brawls. Ohio State quarterback J.T. Barrett should return after a freak collision with a cameraman during Saturday’s game. Will he be himself? We’re confident the Buckeyes will prevail, but even then, our model gives them just a 22-percent chance of making the playoff – possibly handing a golden ticket to Alabama.
Our forecast: Buckeyes by 7.

ACC: Clemson (No. 4) vs. Miami (No. 17). While Miami has been a media darling over the past month or so, we’ve consistently shorted them. We felt somewhat vindicated after the Hurricanes laid an egg Friday, and we expect little magic from them Saturday against a Clemson team that has shown well nearly all season. That said, our model suggests if Miami claims the title, they stand an 84-percent chance of making the playoff.
Our forecast: Tigers by 11.

Big 12: Oklahoma (No. 6) vs. TCU (No. 11). The Big 12 re-instituted a title game this year to boost the odds of the conference champion securing a spot in the College Football Playoff. The Sooners easily handled TCU at home earlier this month, but this time they square off at AT&T Stadium, in Arlington, Texas. If Oklahoma wins, it is a virtual lock to make the playoff. If TCU wins, it has just a 24-percent shot of a playoff berth – and will open another possible route for Alabama.
Our forecast: Sooners by 7.

Pac-12: USC (No. 12) vs. Stanford (No. 20). Earlier in the season, the Pac-12 championship game appeared likely to produce a CFP entrant. With the conference’s best teams imploding down the stretch, that possibility is vanishingly small. But USC can still theoretically get in if, for instance, they annihilate Stanford, while underdogs in other title games win. But our model suggests it won’t get to that.
Our forecast: Trojans by 5.

Committee Uncertainty

Projecting game results is a breeze compared to predicting the playoff committee’s preferences this season. But we’ve been learning as we go: “best team,” it appears, matters about as much as “best resume,” margin of defeat matters more than margin of victory, and it helps to be Notre Dame.

Our model allows us to project the committee’s “pecking order,” given any outcome, based on how the committee has acted over the past four seasons (see table). So what does the pecking order look like for the potential winners of the Power Five conferences, as well as one notable non-champ contender?

If Wisconsin pulls the upset in Indianapolis, we see them as an easy No. 1 seed. SEC champ Georgia and ACC champ Clemson are neck-and-neck behind the Badgers, though we think the committee would give the edge (55-percent) to the Bulldogs. A Big 12 champ Oklahoma slides in at No. 4, just ahead of SEC champ (but two-loss) Auburn. We see those two comfortably ahead of ACC champ Miami if the Canes manage to get past Clemson.

Close behind Miami, appearing at No. 7 in the pecking order, is SEC West runner-up Alabama. Based on how the committee historically has treated wins and losses, conference championships, “best” vs. “resume”, etc., we believe they are 84-percent likely to take the Tide over a 2-loss Big 10 champ Ohio State.

Far behind is TCU, who will need surprises both on the field and from the committee to make the final four.

Every model is imperfect, of course, so we can’t say with certainty what the committee will do, especially because it has not been – ahem – a model of consistency over the years.

How It Adds Up

Overall, in our view, Clemson is the team mostly likely to make the playoff (81 percent), followed by Alabama and Oklahoma (both at 70 percent). Next are teams who are win-and-in but have tough title matchups: Georgia (53 percent), Auburn (47 percent) and Wisconsin (39 percent). Finally we have the long shots: Miami (16 percent) with a steep hill against Clemson, and Ohio State (16 percent) and TCU (7 percent), both needing wins and a little help.

In our simulations, the most likely brackets (32 percent of the time), include Alabama, Clemson, Oklahoma and the Auburn-Georgia winner.

Fun time of year. Unless you’re in Tuscaloosa or Columbus, waiting for a committee you can’t control instead of an opponent on the field you (mostly) can.

— Bob Tedeschi contributed

Projected Pecking Order Probability of being seeded higher than…
Team Conf
Champ
Losses Likelihood Georgia Clemson Oklahoma Auburn Miami (FL) Alabama Ohio State TCU
Wisconsin Big Ten 0 27% 67% 71% 89% 94% 99% 99% n/a 100%
Georgia SEC 1 50% 55% 79% n/a 96% 97% 100% 100%
Clemson ACC 1 81% 76% 84% n/a 97% 100% 100%
Oklahoma Big 12 1 72% 61% 83% 88% 98% n/a
Auburn SEC 2 50% 75% 81% 97% 98%
Miami (FL) ACC 1 19% 59% 90% 94%
Alabama 1 100% 84% 89%
Ohio State Big Ten 2 73% 60%
TCU Big 12 2 28%