College football starts in September with 128 teams and billions of different paths to the playoff. Nearly all of that multiverse has been closed off — on Saturday, it’ll be sealed for good.
There are four major conference titles at stake Saturday, but only two could really cause chaos for the playoff selection committee: those of the SEC and ACC. The Big Ten championship game matters less, from our perspective, because the winner is almost certainly getting in. And the Pac-12 championship only comes into play if Stanford wins and Alabama or Clemson loses (or both).
To break down just how those games could affect the playoff picture, my colleague Jay Boice and I ran 20,000 simulations of FiveThirtyEight’s College Football Playoff model. This gives us greater insight into both how Saturday’s games could play out and, more importantly, how the committee might make its playoff picks based on those outcomes.
Two things we know:
- Oklahoma is in. In all our scenarios, the Sooners are virtually certain to make the playoff.
- The Big Ten winner is in. Whichever team wins the championship game — Iowa or Michigan State — its probability of making the playoff also approaches 100 percent, according to the model. (Our projection favors the Spartans.) It’s very unlikely that the loser gets in under any scenario, and so throughout this analysis it’s assumed that the loser is out.
I analyzed the multiverse of playoff scenarios by first looking at the two most important games: the championship games for the ACC (Clemson vs. North Carolina) and the SEC (Alabama vs. Florida). If Alabama and Clemson win, the selection committee’s job is super easy, but if one or — Bear Bryant forbid — both lose, then it’s not so clear.
If there is chaos on Saturday, three teams are standing offstage: Stanford, Ohio State and North Carolina. But none of them controls its own destiny (even if UNC beats Clemson). The model can’t say with any great confidence which of those teams would take the place of the Tide and/or Tigers should those favorites fall — or whether Alabama or Clemson would be replaced at all.
The possible scenarios
Both Alabama and Clemson win (42 percent likelihood):
This is the simplest outcome. The Tide and Tigers are in, alongside Oklahoma and the Big Ten winner. That’s an impressive pair at the top: an undefeated ACC champion and a one-loss winner of the SEC (arguably the strongest conference in college football). This is the likeliest scenario, happening 42 percent of the time.
(Stanford fans: You may notice that the model gives the Cardinal a 3 percent chance if they win the Pac-12 in this scenario. It’s not happening; the model’s just being cautious/polite.)
Alabama wins but Clemson loses (32 percent likelihood):
This scenario is the second-most-plausible because North Carolina is more likely to upset Clemson than Florida is to upset Alabama.
What would come next is complex. (Which should come as no surprise — this is college football, after all.) If Clemson loses, four teams could be vying for the fourth playoff spot: Stanford, Ohio State, the newly crowned ACC champion Tar Heels and … Clemson. That’s right, our model thinks there’s a decent chance that the Tigers could make the playoff over North Carolina even if they lose the ACC title to the Tar Heels. At 42 percent, Clemson is better-positioned than any of the other three teams. But it’s close: Stanford isn’t far behind at 34 percent, should the Cardinal win the Pac-12.
How could Clemson make it over the team it just lost to? The Tigers have a more impressive schedule than UNC and a signature win against Notre Dame. The Tar Heels suffer both from a weaker schedule and an embarrassing early season loss to South Carolina. Would that really be enough to outweigh North Carolina’s conference championship and its head-to-head win? FiveThirtyEight suspects that the model is being slightly too kind to Clemson,1 but the committee would have a lot to think about. As my boss Nate Silver wrote last year, only a handful of No. 1-ranked teams have lost their conference championship game. On average, those teams were demoted to No. 4 after the loss. If the Tigers suffered the same fate, they’d still be in the playoff. Still, there’s no guarantee that the playoff committee would treat them similarly.
Instead of choosing an ACC team, the committee could opt for plans C or D: Stanford, if it wins the Pac-12 championship (it’s a slight underdog), or Ohio State. In some ways, this would recall the dilemma the committee faced last year: Without clarity about whether Baylor or TCU was the superior Big 12 option, it froze the Big 12 out of the playoff.
If Stanford loses, the committee still has three teams to choose from. But once again, our model thinks Clemson is as likely to make it as anyone. The Tigers are given a 46 percent shot. The Tar Heels are not far behind with a 36 percent shot. In this scenario, an emphatic win over Clemson might be enough to get the Tar Heels in, while a close one might not. Ohio State would have a 17 percent chance — but one problem for the Buckeyes is that, if the committee is OK with a one-loss team that failed to win its conference, it might decide Clemson is the superior one.
Alabama loses but Clemson wins (15 percent likelihood):
This scenario would shake things up a lot. Clemson, Oklahoma and the Big Ten winner are locks — but the fourth team is anyone’s guess.
We know, though, that it wouldn’t be Florida. Beating the Tide isn’t enough for the Gators. After an awful finish to the season — a 27-2 loss to Florida State after inexcusably close calls against Florida Atlantic and Vanderbilt — the Gators have lost more ground in the playoff chase than they can make up, and the model gives them just a 2 percent chance of making the playoff even with a win over Alabama. So an Alabama loss would probably shut the SEC out of the postseason.
But who takes Alabama’s place would be a complicated question, depending in large part on what happened in Stanford’s game against USC.
If Stanford won …
The Cardinal would then be the best-positioned team to nab the fourth slot. They’re not a lock, though — the model thinks two-loss Stanford has a 61 percent likelihood of taking Alabama’s place.
Behind the Cardinal is one-loss Ohio State, which has a 25 percent chance of making it if Alabama loses but Stanford and Clemson win. The Buckeyes are the defending national champion and had won 23 games in a row before losing to Michigan State on the last play of the game two weeks ago. We think Stanford probably has the inside track — but the committee could conclude that the Buckeyes’ overall résumé outweighs Stanford’s conference championship and two losses, especially if Stanford’s win over USC is underwhelming by its “eye test.”
Beyond Ohio State, another team could conceivably make it in this scenario: that same Alabama team that just lost the SEC championship! The model gives the Tide a 13 percent chance of snagging a playoff spot if they lose and both Clemson and Stanford win. It’s doubtful that the committee will be so forgiving, however, unless perhaps the SEC championship involves a controversial finish.
But if Stanford lost …
It would clear the way for Ohio State. The one-loss Buckeyes make it 64 percent of the time that Clemson wins but Stanford and Alabama both lose. The two-loss Tide gets a 25 percent shot in this scenario. For what it’s worth, ESPN’s Football Power Index projections say Stanford is slightly more likely to lose than to win on Saturday.
Both Alabama and Clemson lose (11 percent likelihood):
This is the committee’s nightmare scenario: unlikely to happen, but ugly if it does. All of a sudden, there would be two slots available and five teams vying to fill them.
If this happens, Stanford is the best bet to make it into the playoff. But only if it wins the Pac-12 championship. The Cardinal’s odds rise to 62 percent in this scenario — their highest probability of any of the rabbit holes we sent them down.
Assuming Stanford wins …
Then Clemson’s odds fall to 50/50. The Tigers, as noted earlier, are still considered potentially playoff-worthy by our model even if they lose to North Carolina. The ACC champion Tar Heels, however, aren’t a bad option either and are given a 43 percent chance. Ohio State is also a viable alternative at 30 percent.
Alabama is unlikely to make it; the Tide have an outside shot at 12 percent. That makes sense, as both UNC and Stanford would have conference championships over Alabama, and in the Tar Heels’ case, they’d have fewer losses, too.
However, should Stanford lose to USC …
Then this quagmire becomes slightly less difficult (four teams for two spots). Clemson, North Carolina and Ohio State stand somewhere between a 48 percent and 67 percent shot in this scenario. Two-loss Alabama would be a longer shot at 24 percent.
But to be honest, we’d be entering uncharted territory here. From an empirical point of view, these messy outcomes would be great: We’d learn a lot more about how the committee weighs conference championships, losses, strength of schedule and other factors. For football fans, though, they would be another controversial finish to a thrilling season. Which maybe should come as no surprise — this is college football, after all.