clock menu more-arrow no yes

Filed under:

The College Football Playoff committee’s ranking method, revealed at last!

New, 139 comments

We don’t need a bunch of people in a room. Just follow this formula, and you can do it, too.

NCAA Football: College Football Playoff-Selection Sunday
These people are useless.
Kevin Jairaj-USA TODAY Sports

When the Washington State Cougars were revealed to be the No. 13 team in the College Football Playoff’s latest rankings, we all were pretty pissed off. For that to happen, a 10-2 team from a major conference — one of eight Power 5 teams in the entire country to have two or fewer losses — had to be ranked behind a quartet of three-loss teams.

Which, of course, means that without some sudden change of heart from the CFP committee, WSU will be on the outside looking in with regards to the biggest bowls — the New Year’s Six.

A lot of us have been asking, “How could this possibly be? How can the committee justify ranking all those teams who have one more loss ahead of WSU?”

The method actually seems to be pretty straightforward — just follow these simple steps!


1. Order the Power 5 teams by number of losses.

Step 1

Team Wins Losses
Team Wins Losses
Alabama 12 0
Clemson 12 0
Notre Dame 12 0
Georgia 11 1
Ohio State 11 1
Oklahoma 11 1
Michigan 10 2
WSU 10 2
Florida 9 3
Kentucky 9 3
LSU 9 3
Penn State 9 3
Syracuse 9 3
Texas 9 3
Utah 9 3
Washington 9 3
West Virginia 8 3
Mississippi St. 8 4
Texas A&M 8 4

CFP committee chair Rob Mullens: “This is intuitive enough, right? Moving on ...”

2. Break the ties according to the eye test*.

*It’s nothing more than a COMPLETELY UNRELATED COINCIDENCE that it just happens to bear a distinct resemblance to preseason rankings.

Step 2

Rank Team Wins Losses AP Preseason
Rank Team Wins Losses AP Preseason
1 Alabama 12 0 1
2 Clemson 12 0 2
3 Notre Dame 12 0 12
4 Georgia 11 1 3
5 Ohio State 11 1 5
6 Oklahoma 11 1 7
7 Michigan 10 2 14
8 WSU 10 2 NR
9 Washington 9 3 6
10 Penn State 9 3 10
11 West Virginia 8 3 17
12 Texas 9 3 23
13 LSU 9 3 25
14 Florida 9 3 27
15 Utah 9 3 28
16 Kentucky 9 3 42
17 Syracuse 9 3 NR
18 Mississippi St. 8 4 18
19 Texas A&M 8 4 33

Mullens: “Listen — when there are only 12 games, it’s real hard to distinguish between teams who are this close. So we’re going to rely on our preconceived notions the ‘eye test.’ ”

3. Bump the SEC teams*.

*But only a little if the team is actually a basketball school.

Mullens: “Listen, strength of schedule is a key component of our deliberations. And look — literally everyone knows that wins in the SEC JUST MEAN MORE, as you can see from the fact that we’ve ranked seven SEC teams. (Also, losses count for less because the league is just so darned hard!)”

4. Tweak at the top based on individual results.

Mullens: “I mean, that loss to Purdue by a million was pretty putrid, amirite? Losing by a field goal to Texas is much better.”

5. Penalize teams who lost late in the season*.

*But only a teensy weensy bit if they lost to another SEC team. And be sure to apply the bonus penalty to teams who were supposed to be bad and had no business being up here in the first place.

Mullens: “WSU ... yeah, we knew all along the Cougars weren’t as good as their record. Didn’t you notice we had them No. 8 for a month? It’s not our fault you can’t take a hint. That’s on you, good sir/madam. (Plus, I’m still bitter that WSU has beaten Oregon four straight times. How does THAT happen??) Anyway, I guess they’re still better than Texas, because the Big 12 kinda sucks too. And as we already talked about, we’re not even sure Kentucky is actually fielding a football team. And WSU beat Utah, so I guess we have to keep them ahead of the Utes. We’re pretty sure Mississippi State and Texas A&M are better than WSU, but we didn’t want to raise too many eyebrows.”

6. Figure out what to do with that pesky Group of 5 team.

Mullens: “Uggggggh. Why do we have to include them? It doesn’t really matter anyway — they’re not getting into the playoff and they’re getting a NY6 no matter what. Anywho, they’d obviously lose at least two games even if they played in the Pac-12 or ACC, so we can’t put them near the top four. Just between us, they’d probably lose like five if they played in the SEC. Everyone thought they’d be pretty OK before the season, so we threw them a bone so that it looked like we seriously thought about them. (We did not seriously think about them.)”

7. Congratulate yourself on your fine work!

CFP Rankings, Pre-Championship Weekend

CFP Team Record AP Preseason
CFP Team Record AP Preseason
1 Alabama 12-0 1
2 Clemson 12-0 2
3 Notre Dame 12-0 12
4 Georgia 11-1 3
5 Oklahoma 11-1 7
6 Ohio State 11-1 5
7 Michigan 10-2 14
8 UCF 11-0 21
9 Florida 9-3 27
10 LSU 9-3 25
11 Washington 9-3 6
12 Penn State 9-3 10
13 WSU 10-2 NR
14 Texas 9-3 23
15 Kentucky 9-3 42
16 West Virginia 8-3 17
17 Utah 9-3 28
18 Mississippi St. 8-4 18
19 Texas A&M 8-4 33
20 Syracuse 9-3 NR

Mullens: “Nailed it.”


Ultimately, college football always has been — and probably always will be — about perception. This is a sport that decided its champion for the better part of a century with polls, and those habits die awfully hard.

A 12-game sample size isn’t enough to meaningfully change people’s minds when trying to distinguish between closely bunched teams, especially when there’s so little common ground from which to judge them. So the committee relies on things like “the eye test,” which is rampant with confirmation bias — “we perceived that Team A would be good, therefore whatever good they do confirms that there’s a good team in there somewhere, meanwhile, Team B was supposed to be bad, therefore whatever bad they do confirms they aren’t as good as it might have appeared. And, of course, we have to take into account the strength of the conference, which is also based largely on history and preconceived conclusions, and also our very own rankings.”

Schools such as WSU will always have a perception problem because of the conference in which they play, the program’s lack of historical prestige, and their modest recruiting rankings, all of which impact preseason perception — which, as we’ve shown, has a tremendous impact on late-season perception. (Side note: This is a huge part of the reason why it took Gameday so long to get to Pullman — really tough to get ESPN’s attention when half your home games are off the table because you started the season unranked.)

I mean, there’s no way on earth a person can objectively rank the Penn State Nittany Lions ahead of WSU when the method is quite literally about ranking teams by most deserving to least deserving of a playoff appearance. Consider the following blind resumes. How would you rank the following six teams without knowing who they are?

We put this question to all of our authors, as well as to followers on Twitter, and without giving away too much (yet), Penn State’s resume was picked last in over half the responses we received. In the other ones, it was in the bottom three with only a couple exceptions. It’s not hard to see why — of this group, they are the only ones to not have a win over a ranked team and lost three times to FBS teams with winning records.

Penn State lost on the road by 35 to the Michigan Wolverines and at home to the 7-win Michigan State Spartans. They needed overtime to beat a Sun Belt team (granted a pretty good one). Their best win is over the 8-win Iowa Hawkeyes, and it’s their only victory against a team with at least eight wins.

WSU will have bested three teams with more than 8 wins after the California Golden Bears play the Stanford Cardinal on Saturday, and one of those three teams WSU beat is a 9-win division champ (Utah Utes).

So what puts Penn State above WSU? Their close loss to a team that got blown out by the Purdue Boilermakers but later found a way to win their division by blowing out the top-ranked team in their conference? That sounds like a mess.

Maybe it’s because the Nittany Lions waxed a team that won a garbage division with five losses?

The SEC’s 5-2 record in Out Of Conference games against the dregs of the ACC is being used to prop up their conference might; does Penn State whipping 7-5 “division champ” Pitt mean the Nittany Lions get that same favor? Does that win somehow mean more than WSU beating 9-3 division champ Utah?

No, it’s none of those things. It’s because Penn State was ranked 10th in the preseason and has a brand. Fine, just say that.

Instead we get explanations from Oregon AD and CFP Chairman Rob Mullens that sound horribly conflicting. Like this response to a question about why Oklahoma’s defense hasn’t negatively impacted their ranking;

We have mentioned it [OU’s defense] in several weeks, but I think that’s balanced by the dynamic offense that they have and their ability to find a way to win riding that offense.”

So Oklahoma is given credit for having to overcome their own bad defense? Like it’s a good thing? Sure, that checks out. Good to know if you win with a defense ranked 84th in S&P+ that’s actually bonus points for difficulty.

Their answers always sound horribly conflicting because this sport is a crazy mess of transitive chaos. It doesn’t have a magical sorting hat to tell the good teams what order to stand in, and the logical methods of sorting can make even less sense than that would.

Just read Jon Wilner use the transitive property like a child connecting dots in crayon to downgrade the entire conference because UW lost to Auburn in September, and other random things that have nothing to do with what WSU did on a football field. Seriously, his sole point about anything WSU could control was that they didn’t schedule a Power 5 non-conference game. Florida — ranked NINTH in the CFP — scheduled two FCS non-conference games and came up with one less Power Five win than WSU. But sure, only OOC Power Five wins matter.

Great teams are easy for both the CFP Committee and AP to figure out and they mostly get these right. The problem is with the good teams.

This year the problem is what they’re doing to the Cougs.

By the way, here’s the results to our small sample size poll (only around 50 responses). Blind resume judgement — at least on this tiny scale of responses and with these unrevealing data points — overwhelmingly favored Wazzu and Kentucky, the only two teams below 40th in the AP Preseason Poll (and without the name brand sparkle).

Florida being the highest ranked out of this group of six is the highest differential from the poll. And well...

It doesn’t matter if they do. The committee wants them there and doesn’t have to explain itself with any accountability for clarity. Losing at home to Missouri by three touchdowns is still Top 10 material because hey, Mizzou and its four losses are Top 25 caliber now.

Why? Because Mizzou is in the SEC. And those four losses were really hard, you guys.

Maybe Mike Leach was right all along with his idea for a 64-team playoff.