What would a 12-team playoff have looked like since 2014? Simulations, storylines and more

To see how a new development might affect the future, take a look at how it would have affected the past. That’s a go-to of mine — see: how larger playoffs would have worked in 2020 or how a much bigger, earlier playoff would have impacted things — and with the news that we could be moving toward a 12-team playoff soon, we have a reason to dip into that well once more.

For all the simulations I’ve done, I had never really paid much attention to a 12-team format. An eight-teamer — six conference champions with two at-large bids — has long seemed perfectly inclusive and interesting to me. It was the logical next step. But as it turns out, a 12-teamer works quite well in terms of political calibration.

A 12-team playoff indeed offers a playoff path to what we have long referred to as the Group of Five conferences, assuring that college football’s national title race is actually inclusive for just about the first time ever. But it also assures that the most powerful conferences and teams benefit massively from extra at-large bids. And the news that quarterfinal games would take place in bowl games (presumably the four New Year’s Six bowls that don’t host semifinals in a given year) instead of home fields conveniently allows the most influential bowls to remain relevant. We lose home-field atmospheres, but the Rose Bowl doesn’t end up with the Big Ten’s fourth-best team against the Pac-12’s third.

To best learn about what a 12-teamer changes, however, let’s hop in the simulation machine.

Below are how each of the last seven College Football Playoffs would have taken shape with 12 teams instead of four. I used the playoff committee’s rankings as they existed — there’s a distinct possibility that the committee considers teams differently with a 12-team cutoff instead of four, but we won’t know what the changes are for a while. I also worked under the aforementioned assumption that the quarterfinals will take place in four New Year’s Six bowls, with the semifinals taking place in the same bowls that hosted the four-team semis. I simulated each playoff using my SP+ rankings, and I’m including each playoff team’s odds of reaching the semifinals below, as well as the most likely champions each season. Why semifinal odds? Because I want to see what might have changed compared to the four-team playoff that we actually got.)

2014

First round

• 9 Ole Miss (9-3) at 8 Michigan State (10-2)

• 12 Boise State (11-2) at 5 Baylor (11-1)

• 11 Kansas State (9-3) at 6 TCU (11-1)

• 10 Arizona (10-3) at 7 Mississippi State (10-2)

Top-ranked snubs: Georgia Tech (10-3), Georgia (9-3)

Quarterfinals

• Orange Bowl: 1 Alabama (12-1) vs. Michigan State/Ole Miss winner

• Cotton Bowl: 4 Ohio State (12-1) vs. Baylor/BSU winner

• Peach Bowl: 3 Florida State (13-0) vs. TCU/KSU winner

• Fiesta Bowl: 2 Oregon (12-1) vs. Mississippi State/Arizona winner

Semifinal odds, per SP+

• 1 Alabama 75%

• 3 Florida State 66%

• 2 Oregon 60%

• 4 Ohio State 59%

• 7 Mississippi State 37%

• 5 Baylor 37%

• 6 TCU 27%

• 8 Michigan State 18%

• 9 Ole Miss 8%

• 11 Kansas State 7%

• 12 Boise State 4%

• 10 Arizona 3%

Odds of top 4 all advancing: 17%
Most likely champions: Alabama (37%), Florida State (19%), Ohio State (11%)

Originally, the big controversy from this season came when Ohio State usurped Baylor and TCU for the No. 4 spot. In this example, the Buckeyes merely steal a bye. Baylor draws a solid but unspectacular Boise State, while TCU hosts a three-loss Kansas State team it beat by three touchdowns earlier that season. They both likely win and advance to intriguing Ohio State-Baylor and FSU-TCU quarterfinals.

---

2015

First round

• 9 Florida State (10-2) at 8 Notre Dame (10-2)

• 12 Houston (12-1) at 5 Iowa (12-1)

• 11 TCU (10-2) at 6 Stanford (11-2)

• 10 North Carolina (11-2) at 7 Ohio State (11-1)

Top-ranked snubs: Ole Miss (9-3), Northwestern (10-2)

Quarterfinals

• Peach Bowl: 1 Clemson (13-0) vs. Notre Dame/FSU winner

• Rose Bowl: 4 Oklahoma (11-1) vs. Iowa/Houston winner

• Fiesta Bowl: 3 Michigan State (12-1) vs. Stanford/TCU winner

• Sugar Bowl: 2 Alabama (12-1) vs. Ohio State/UNC winner

Semifinal odds, per SP+

• 4 Oklahoma 80%

• 2 Alabama 54%

• 1 Clemson 53%

• 6 Stanford 42%

• 7 Ohio State 39%

• 3 Michigan State 37%

• 8 Notre Dame 26%

• 9 Florida State 22%

• 11 TCU 21%

• 5 Iowa 14%

• 10 North Carolina 8%

• 12 Houston 7%

Odds of top 4 all advancing: 9%
Most likely champions: Alabama (25%), Ohio State (18%), Oklahoma (16%)

This draw gives out two consequential mulligans. The first one goes to Ohio State, which was quite possibly deeper and more talented than 2014’s title team but slogged through a miserable 17-14 defeat to Michigan State. Instead of costing the Buckeyes a bid, it costs them a bye, and they likely beat UNC and get a shot at a flawed Bama in the quarterfinals.

The second mulligan goes to a Stanford team that began the season with the infamous “body clocks” loss to Northwestern before catching fire. In this reality, they get the six-seed and have an excellent chance of getting through TCU and Michigan State and into the semifinals.

---

2016

First round

• 9 USC (9-3) at 8 Wisconsin (10-3)

• 12 Western Michigan (13-0) at 5 Ohio State (11-1)

• 11 Florida State (9-3) at 6 Michigan (10-2)

• 10 Colorado (10-3) at 7 Oklahoma (10-2)

Top-ranked snubs: Oklahoma State (9-3), Louisville (9-3)

Quarterfinals

• Sugar Bowl: 1 Alabama (13-0) vs. Wisconsin/USC winner

• Cotton Bowl: 4 Penn State (11-2) vs. Ohio State/WMU winner

• Rose Bowl: 3 Washington (12-1) vs. Michigan/FSU winner

• Orange Bowl: 2 Clemson (12-1) vs. Oklahoma/Colorado winner

Semifinal odds, per SP+

• 1 Alabama 74%

• 2 Clemson 59%

• 3 Washington 56%

• 4 Penn State 47%

• 5 Ohio State 45%

• 7 Oklahoma 40%

• 6 Michigan 25%

• 11 Florida State 19%

• 9 USC 19%

• 12 WMU 8%

• 8 Wisconsin 7%

• 10 Colorado 1%

Odds of top 4 all advancing: 11%
Most likely champions: Alabama (34%), Clemson (19%), Washington (13%)

Ohio State, which earned the No. 3 seed in real life, only draws the No. 5 because it didn’t win the Big Ten. Instead, Penn State — which beat the Buckeyes in the regular season but lost twice — gets a bye … and probably plays Ohio State again in the Cotton Bowl quarterfinals.

Note the chaos potential of the odds above, by the way. Six teams have between a 40-59% chance of reaching the final four, and there’s only an 11% chance of the top four seeds getting there.

---

2017

First round

• 9 Penn State (10-2) at 8 USC (11-2)

• 12 UCF (12-0) at 5 Alabama (11-1)

• 11 Washington (10-2) at 6 Wisconsin (12-1)

• 10 Miami (10-2) at 7 Auburn (10-3)

Top-ranked snubs: Stanford (9-4), Notre Dame (9-3)

Quarterfinals

• Peach Bowl: 1 Clemson vs. USC/PSU winner

• Fiesta Bowl: 4 Ohio State vs. Alabama/UCF winner

• Orange Bowl: 3 Georgia vs. Wisconsin/Washington winner

• Cotton Bowl: 2 Oklahoma vs. Auburn/Miami winner

Semifinal odds, per SP+

• 1 Clemson 58%

• 2 Oklahoma 56%

• 3 Georgia 54%

• 5 Alabama 51%

• 4 Ohio State 48%

• 7 Auburn 34%

• 6 Wisconsin 33%

• 9 Penn State 30%

• 11 Washington 14%

• 8 USC 12%

• 10 Miami 9%

• 12 UCF 2%

Odds of top 4 all advancing: 9%
Most likely champions: Alabama (23%), Ohio State (18%), Georgia (12%)

Maximum chaos potential. Alabama’s late-year loss to Auburn knocked the Crimson Tide from the SEC Championship game, but when Georgia defeated Auburn in Atlanta, Bama sneaked into the four-team playoff and then won the whole thing. They obviously get in here, too, but as the No. 5 seed. And with no true standout teams, eight teams have at least a 30% chance of reaching the semis and all are under 60%.

Also: self-proclaimed national champion UCF no longer complains about getting left out of the Playoff. Instead … they have to beat Bama and Ohio State to get to the semis. Ouch.

---

2018

First round

• 9 Washington (10-3) at 8 UCF (12-0)

• 12 Penn State (9-3) at 5 Notre Dame (12-0)

• 11 LSU (9-3) at 6 Georgia (11-2)

• 10 Florida (9-3) at 7 Michigan (10-2)

Top-ranked snubs: Washington State (10-2), Kentucky (9-3)

Quarterfinals

• Peach Bowl: 1 Alabama (13-0) vs. UCF/Washington winner

• Rose Bowl: 4 Ohio State (12-1) vs. Notre Dame/PSU winner

• Fiesta Bowl: 3 Oklahoma (12-1) vs. Georgia/LSU winner

• Sugar Bowl: 2 Clemson (13-0) vs. Michigan/Florida winner

Semifinal odds, per SP+

• 1 Alabama 85%

• 2 Clemson 66%

• 4 Ohio State 59%

• 6 Georgia 50%

• 3 Oklahoma 38%

• 5 Notre Dame 25%

• 7 Michigan 17%

• 10 Florida 16%

• 12 Penn State 16%

• 11 LSU 13%

• 9 Washington 9%

• 8 UCF 6%

Odds of top 4 all advancing: 13%
Most likely champions: Alabama (45%), Georgia (17%), Clemson (16%)

Beginning in 2018, the teams at the top of the pack began to get more dominant. This was Nick Saban’s best Bama team to date, and Clemson was playing otherworldly ball at the end of the season. They stand out as favorites, and while the other two quarterfinals could have been competitive, the chaos potential is minimized this year.

Two other notes:

UCF becomes the first Group of Five team to host a first-round game, but even if the Knights win … they get Bama again in the next round. Double ouch.

Unbeaten Notre Dame draws only the No. 5 seed because the Irish aren’t conference champions. Their odds of reaching the semifinals are far lower than those of the top four seeds, but as athletic director Jack Swarbrick pointed out recently, they would have to play a conference title game to earn one of those top four seeds. That brings obvious risk as well.

---

2019

First round

• 9 Florida (10-2) at 8 Wisconsin (10-3)

• 12 Memphis (12-1) at 5 Georgia (11-2)

• 11 Utah (11-2) at 6 Oregon (11-2)

• 10 Penn State (10-2) at 7 Baylor (11-2)

Top-ranked snubs: Auburn (9-3), Alabama (10-2)

Quarterfinals

• Sugar Bowl: 1 LSU (13-0) vs. Wisconsin/Florida winner

• Cotton Bowl: 4 Oklahoma (12-1) vs. Georgia/Memphis winner

• Orange Bowl: 3 Clemson (13-0) vs. Oregon/Utah winner

• Rose Bowl: 2 Ohio State (13-0) vs. Baylor/PSU winner

Semifinal odds, per SP+

• 2 Ohio State 78%

• 1 LSU 74%

• 3 Clemson 70%

• 4 Oklahoma 46%

• 5 Georgia 44%

• 6 Oregon 16%

• 9 Florida 15%

• 10 Penn State 15%

• 11 Utah 14%

• 8 Wisconsin 11%

• 12 Memphis 10%

• 7 Baylor 7%

Odds of top 4 all advancing: 19%
Most likely champions: Ohio State (36%), LSU (28%), Clemson (14%)

There is great potential for a new crop of rivalry games in a 12-team era. With a possible Oklahoma-Georgia Cotton Bowl matchup, this might have been the third consecutive year that saw the Sooners and Dawgs face off, after the 2017 semis and 2018 quarterfinals. Meanwhile, Alabama and Ohio State might have played each other five years in a row — the 2014 quarterfinals, 2015 semis, 2016 semis, 2017 quarters and 2018 semis. Yes, we probably still get our fair share of Bama-Clemson battles, but that isn’t the only defining matchup.

Also: using the CFP rankings as they were originally drawn up, a two-loss Alabama team gets left out of these playoffs in favor of Utah and Penn State. I’m going to go out on a limb and guess the committee doesn’t rank the Tide 13th if 12 is the new cutoff.

---

2020

First round

• 9 Georgia (7-2) at 8 Cincinnati (9-0)

• 12 Coastal Carolina (11-0) at 5 Notre Dame (10-1)

• 11 Indiana (6-1) at 6 Texas A&M (8-1)

• 10 Iowa State (8-3) at 7 Florida (8-3)

Top-ranked snubs: North Carolina (8-3), Northwestern (6-2)

Quarterfinals

• Peach Bowl: 1 Alabama (11-0) vs. Cincinnati/Georgia winner

• Fiesta Bowl: 4 Oklahoma (8-2) vs. Notre Dame/Coastal winner

• Cotton Bowl: 3 Ohio State (6-0) vs. A&M/Indiana winner

• Orange Bowl: 2 Clemson (10-1) vs. Florida/ISU winner

Semifinal odds, per SP+

• 1 Alabama 78%

• 4 Oklahoma 74%

• 3 Ohio State 73%

• 2 Clemson 61%

• 7 Florida 29%

• 6 Texas A&M 23%

• 5 Notre Dame 18%

• 9 Georgia 14%

• 10 Iowa State 10%

• 8 Cincinnati 9%

• 12 Coastal Carolina 8%

• 11 Indiana 5%

Odds of top 4 all advancing: 25%
Most likely champions: Alabama (42%), Ohio State (18%), Clemson (14%)

The “top six conference champions” rule wreaks havoc in 2020. With unranked Oregon beating USC, that means Conference Champion No. 6 is now the Sun Belt’s Coastal Carolina. Meanwhile, Oklahoma’s Big 12 title game win over Iowa State gives the Sooners a bye and sends the Cyclones to Gainesville for round one. That’s all undercard, however. The top four teams are the most dominant set of four yet, though Clemson gets a tricky draw in potentially facing a Florida team that doesn’t have Kyle Pitts et al opting out in the quarterfinals.

And yes, Alabama likely still wins it all.

---

Here are some thoughts about what these simulations tell us.

This is a massive change

In the first seven years of the four-team playoff, only two teams that ended up ranked worse than 10th in SP+ made the field — one of those two, 2020 Notre Dame, only fell out of the top 10 after playing in the CFP — while 21 of 28 teams finished the year in the top five. This club was too exclusive and hasn’t really benefited college football in terms of inclusiveness.

We’re extremely inclusive now.

Again, it’s conceivable that the thought processes behind how the committee ranks teams will change. But based solely on the 2014-20 rankings that already exist, a 12-team field would have brought in 10 teams that ranked worse than 20th in SP+. And while that group does include five of the obligatory Group of Five teams — or, as we’ll evidently now refer to them, “sixth conference champions” — it also includes a 2016 Colorado team that finished the year 53rd and a 2014 Arizona team that finished 46th.

This is a true tournament now, for better or worse, and quite a bit of the FBS will get to experience it. In this seven-year simulation, 39 teams — exactly 30% of FBS’ current membership — reach at least one playoff.

Simulated CFP appearances by team
7: Ohio State
6: Alabama, Clemson, Oklahoma
4: Georgia, Penn State
3: Florida, Florida State, Notre Dame, Washington, Wisconsin
2: Baylor, LSU, Michigan, Michigan State, Oregon, TCU, UCF, USC
1: Arizona, Auburn, Boise State, Cincinnati, Coastal Carolina, Colorado, Houston, Indiana, Iowa, Iowa State, Kansas State, Memphis, Miami, Mississippi State, North Carolina, Ole Miss, Stanford, Texas A&M, Utah, WMU

To be sure, this structure still benefits the top dogs most of all. Ohio State now reaches the CFP in every single season, Clemson and Oklahoma are currently on streaks of six in a row, and while Alabama misses the 2019 CFP based on the committee’s original rankings, again, I doubt that happens if there’s a 12-team cutoff.

At the same time, teams like Washington and Wisconsin each make nearly half the playoffs and have decent odds of doing damage when they get there, and a conference that doesn’t have a clear dominant force at the moment — the Pac-12 — sees more than half its membership make at least one appearance.

Simulated CFP appearances by conference
Big Ten: 20
SEC: 19
ACC: 12 (including Notre Dame’s 2020 appearance)
Big 12: 12
Pac-12: 11
AAC: 5
Independents: 2 (both Notre Dame)
MAC: 1
MWC: 1
Sun Belt: 1
Conference USA: 0

Again, the top dogs benefit the most. While expansion to 12 teams finally offers genuine inclusion to all of FBS, it also means that an average of 5.6 SEC and Big Ten teams make the playoff field each year. If revenue ends up distributed by the number of playoff teams your conference has, the richest conferences continue to expand their lead on the field.

The favorites remain the favorites, but the shifting odds are noteworthy

In a 12-team tournament, you’re introducing genuine randomness to the proceedings. Using the SP+ projections above, here are the average odds of reaching the semifinals for each seed, one through 12:

1 seed: 71%
2 seed: 62%
3 seed: 56%
4 seed: 59%
5 seed: 33%
6 seed: 31%
7 seed: 29%
8 seed: 13%
9 seed: 17%
10 seed: 9%
11 seed: 13%
12 seed: 8%

There’s some noise in there, but you see a clear hierarchy, and to be sure, the top four seeds have the best odds of reaching the semis as you would expect. But those odds aren’t anywhere near 100%. Even in 2019-20, when the balance of power coagulated significantly at the top, teams like 2019 LSU and 2020 Alabama only had semifinal odds in the 70-80% range. They still have to beat one extra very good team to get there, and the more good teams you play, the more tempted fate might be to throw you a curveball.

Are conference title games pointless now?

One of the immediate concerns expressed in the aftermath of Thursday’s announcement was that you’re basically rendering the conference championship game pointless. Both teams are still going to make the field, right?

Not necessarily. In this simulation, only 16 of 32 power conference title-game losers made it — and even in the SEC and Big Ten, the runners-up only made it four of seven times each.

A lot of this, of course, is the product of a bad divisional structure. The SEC Championship Game losers that didn’t make it, for instance, were the flawed Missouri and Florida teams that won the SEC East from 2014-16. If conferences were to move to the pod structure I have espoused, (a) it would greatly increase the odds that the conference’s best teams have a chance at the conference title, which I’ve always considered an obviously good thing, but (b) it would indeed lessen the importance of the conference title game in a universe with a 12-team playoff. The winner still gets a valuable bye, but the stakes aren’t quite as high.

Yes, we could have a three-loss national champion now

But it’s unlikely.

When the subject of playoff expansion comes up, those opposed to the idea are all but guaranteed to bring up something to the effect of “Now some 9-3 team could win the national title! That sanctity of the regular season is dead!” There’s no denying that in a 12-team playoff era, we’d be handing out a lot more mulligans.

In the seven years above, 15 three-loss teams make the field of 12, about 2.1 per year. We could now have a national champion with some wacky, NFL-like 14-3 record. But within that three-loss group, only 2017 Auburn (34%) and 2020 Florida (29%) end up with a greater than 20% chance of reaching even the semifinals, and almost none would have odds higher than about 3-4% of winning the title. A Cinderella run of sorts would be possible but extremely unlikely.

The Auburn conundrum

In an average season, Auburn plays in the nation’s best division (the SEC West), draws permanent rival Georgia in cross-division play, and schedules at least one extremely high-quality non-conference opponent each year — Oregon in 2019, Washington in 2018, Clemson in 2017, etc.

The reward for that high-level scheduling: one CFP bid in seven years, per the simulation above. They made it with three losses in 2017 and narrowly missed out in 2019. The rankings above basically make it so that any power-conference team is guaranteed to make the field with one loss and has excellent odds with two losses. If we see a change in how the committee ranks teams moving forward — there’s no guarantee we will, but there’s obviously a chance — it could be in terms of how much teams are forgiven for quality losses.

A changing opt-out universe

When Florida played Oklahoma in the Cotton Bowl at the end of 2020, the Gators did so without all-world tight end Kyle Pitts, receivers Kadarius Toney and Trevon Grimes and defensive back Marco Wilson — four of probably their five or six best players. They had all opted out to begin preparation for the NFL draft.

They almost certainly don’t opt out in a universe with a 12-team playoff. Florida instead hosts Iowa State in mid-December and, with a win, heads to the Orange Bowl to play Clemson. Despite losses to Texas A&M, LSU and Alabama, they still have something to play for well into the postseason.

To the extent that the goal of playoff expansion is to simply create as many meaningful games and as much entertaining content as possible, then throwing a big party is a pretty effective way of making sure as many teams as possible remain engaged in the season for as long as possible. You can find quite a few coaches pushing for an even bigger playoff in private, and that’s part of the reason why. (Well, that and their agents being able to say “He’s made the playoff X consecutive years!” to their employers.)

This is undeniably a good thing from a consumption standpoint. Hopefully it’s a good thing for the athletes themselves, too. But that’s probably fodder for a different column altogether.