Another week of college football is done, and between USC’s victory over Utah, Wisconsin’s trouncing of Michigan, and Georgia’s triumph over Notre Dame, we got plenty of playoff-impactful results. Here’s what’s changed, as viewed by our model:
Ohio State, Welcome to Alabama Territory
Ohio State’s demolition of Miami-Ohio vaulted them upwards in the eyes of the ratings our model uses. This combined with Michigan’s demise to increase the Buckeyes’ playoff chances from an already very good 58.5% to a mighty 70.6%, within a margin of error from Alabama (73.6%). Their national championship hopes aren’t at the level of Nick Saban’s team (15.5% for OSU, 28.7% for ‘Bama), but in terms of getting into a semifinal, Ohio State has as good a shot as anyone not named Clemson.
Wisconsin, Welcome to Oklahoma Territory
While Ohio State was busy making their private-school neighbors question whether the paycheck was worth it, Wisconsin was thoroughly deconstructing Michigan at Camp Randall. Our model expected a Badger victory, but, like the rest of the world, was impressed by the thoroughness of the result. The end product brought Wisconsin a jump in Big Ten championship probability from 23.3% to 29.2%, and a more important jump in playoff probability from 21.2% to 33.0%, actually above that of Oklahoma, the clear Big 12 favorite.
The Big Ten’s divisions have sorted themselves out, at least for the time being, which is a big part of why a new feature of our model—one that tells you how often conferences send multiple teams to the playoff—has the Big Ten with nearly a one-in-four chance of sending a pair of teams to national semifinals. Wisconsin and Ohio State have plenty of difficult games remaining, including at least one with each other, but they’re at least clearly enough above the rest that the default result would be the pair playing twice, and possibly each finishing 12-1. And even if that best-case scenario doesn’t come to fruition, the Big Ten has a very good chance (92.1%) of avoiding sending no teams to the playoff, which would be a relief after the last two years.
Georgia Rises
Georgia’s victory over Notre Dame wasn’t extraordinary. The Bulldogs took care of business, but they struggled to move the ball against a Notre Dame defense that, while possibly good, had trouble stopping Louisville on the ground back on Labor Day. Still, having that caliber of win in their pocket is great for the Dawgs, who climbed from 35.1% likely to make the playoff to 40.0% likely.
Notre Dame Falls
Notre Dame still certainly has a path to the playoff, and in some ways (specifically, defense), the Irish impressed on Saturday. They covered the Vegas spread by somewhere between five and nine points, depending what day of the week you’re looking at, and gave themselves what could go down as one of the best losses in the country. History’s shown that good losses matter, and results against common opponents are one of the selection committee’s explicitly listed criteria. Other criteria can outweigh that, but if things come down to a two-loss SEC team and a one-loss Notre Dame for the final spot, and Georgia’s won the SEC, Notre Dame should have the better result against UGA, having done their losing on the road by about 40 yards.
But.
Of course.
Notre Dame would have preferred to win.
And the scenario outlined above, while realistic, is far from probable. Notre Dame now only makes the field in 7.0% of simulations. And in all seventy of those sim’s, they get significant outside help.
Bye, UCF
UCF has been a point of fascination for us not only because of their unique status as a program over the last few years, but because they help us learn how much the committee values different variables when evaluating teams. After losing to Pitt on Saturday, though, they’re toast for 2019, making the playoff in exactly zero of our simulations. The Group of Five isn’t entirely hopeless: it sends a representative to the playoff in 0.5% of sim’s, with Appalachian State, Boise State, and SMU possessing potential quality Power Five victories. But a one-in-200 chance is far from optimistic.
The Pac-12 Takes on More Water
Utah and Washington State losing pushed the Pac-12 down even further, to the point where our model right now has the conference with only a 4.3% chance of sending even just one team to the playoff. That’s comparable to where UCF was at before this week, and worse than even a one-loss Notre Dame’s chances.
This doesn’t mean the Pac-12 is the worst conference. Take away the top individual teams from each, and the league matches up well with its east coast counterpart, placing each of its second-to-fifth best teams ahead of the ACC’s second-best team in the ratings. The problem isn’t that the league’s bad—it’s that it isn’t good, like the Big Ten or SEC, and it doesn’t have one great team to mask all other shortcomings like the ACC does. And in a world in which conferences are evaluated not by the difficulty of winning them, but by their propensity to send teams to the playoff, that’s a bad place to be.
The Big 12 Needs to Consolidate Its Winning
The Big 12 does have one great team—Oklahoma—and it has few truly bad teams (even Kansas is rated better than the worst of the ACC, Big Ten, and SEC—though notably, not that of the Pac-12, where Oregon State’s shockingly holding the line). But without the coalition of strong teams featured in the Big Ten and the SEC, it leaves its top program little margin for error, a phenomenon most visible when comparing conference championship probabilities with playoff probabilities. For Alabama and Ohio State, the latter dwarfs the former. For Oklahoma, the opposite is true.
What’s a Big 12 to do? The top-heavy method of evaluating schedules means the league would do best by having its top teams (Oklahoma, gap, Texas) win out but for a Longhorn loss in the Red River Shootout, and to get two or three teams to the 9-3 territory that would boost the conference’s presence in the Top 25. Kansas State would be an ideal one to do this, with its strong non-conference road win over Mississippi State, and Oklahoma State and Baylor fit the billing well with no non-league losses themselves. Iowa State and TCU, the remaining top-35 teams in the ratings, each have a non-conference loss, but them beating up on Kansas, Texas Tech, and West Virginia would also help the league. The committee’s shown it’ll forgive a certain level of dead weight on a schedule, especially if a team’s a Power Five team. The key for Oklahoma—the Big 12’s ace—is seeing that dead weight consolidate while a few teams on the other end push against the New Year’s Six bowl threshold.
Tiers
We’ll wrap this up with some tiers. First, simply how good teams are, and these are listed in the order the ratings we use rank them, with their points above an arbitrary threshold somewhere in Division III in the parentheses. These tiers are arbitrary, and the cutoff, which this week comes after the 36th-best team, is placed where it is because of this website’s Fargo bias.
Tier 1 (the best teams): Alabama (99.4), Clemson (98.6)
Tier 2 (the challenger): Ohio State (94.4)
Tier 3 (the challengers to the challenger): Georgia (92.6), LSU (91.2), Oklahoma (90.7)
Tier 4 (very good teams): Wisconsin (88.2), Auburn (87.5)
Tier 5 (good teams): Notre Dame (86.7), Penn State (85.4), Florida (85.4), Washington (84.9), Oregon (84.8)
Tier 6 (more good teams, but not quite as good as those): Texas A&M (82.8), UCF (82.7), Texas (82.2), Mississippi State (81.8)
Tier 7 (decent but flawed teams): Michigan State (80.9), Missouri (80.8), Iowa (80.0), Michigan (79.7), Oklahoma State (79.3), Utah (79.0), Utah (79.0), Iowa State (78.3), Washington State (78.2), Kansas State (78.1), USC (77.7)
Tier 8 (decent but more flawed teams and North Dakota State): TCU (76.4), Maryland (75.5), Memphis (75.1), Boise State (75.0), Miami (74.8), Baylor (74.6), Utah State (74.3), Virginia (74.1), North Dakota State (73.7)
And now, the championship contender tiers, with their national championship likelihood in the parentheses:
Tier 1: Clemson (35.6%)
Tier 2: Alabama (28.7%)
Tier 3: Ohio State (15.5%)
Tier 4: Georgia (5.9%), LSU (4.2%), Oklahoma (4.0%), Wisconsin (3.4%)
Tier 5: Notre Dame (0.7%), Auburn (0.6%), Penn State (0.5%)
Tier 6: Florida (0.2%), Oregon (0.2%), Texas (0.2%), Washington (0.2%), Iowa (0.1%)
