clock menu more-arrow no yes mobile

Filed under:

Get to Know the Playoff Selection Rules

The College Football Playoff Selection Committee recently announced that it has finalized its protocol and procedures - we'll take a look at them to understand how this most critical element of the college football season will work.

The Lombardi Trophy meets the Olympic Torch.
The Lombardi Trophy meets the Olympic Torch.
Kevin Jairaj-USA TODAY Sports

We'll get our first sense this year of the wisdom of forsaking the devil you know in the college football postseason as the playoff selection committee takes the place of the BCS in determining who will play for the national championship. The basic idea is that a small, carefully chosen, committed group of experts will arrive at a consensus ranking of teams based on all available information, and the top four teams will enter a two-round playoff for the national championship. The committee recently sent out a press release detailing the final version of its protocol and procedures - let's take a look at the details so we understand what we're in for. Note that all quoted text is taken from the College Football Playoff press release of August 14th.

What's Their Goal?

The committee’s task will be to select the best teams, rank the teams for inclusion in the playoff and selected other bowl games and, then assign the teams to sites.

Note that the mission statement says they will "rank the teams for inclusion in the playoff," not "rank the teams for inclusion of the four best in the playoff." Hmm... Building in some flexibility already? It's also interesting that the committee will assign the teams to "selected other bowl games," which we'll get into more below. Spoiler: I like what they're doing.

What's In the Hot Dog?

The committee will select the teams using a process that distinguishes among otherwise comparable teams by considering:
  • Conference championships won,
  • Strength of schedule,
  • Head-to-head competition,
  • Comparative outcomes of common opponents (without incenting margin of victory)
  • Other relevant factors such as key injuries that may have affected a team’s performance during the season or likely will affect its postseason performance.

Nothing innovative about the first three criteria, but then it gets interesting. I don't think margin of victory is an absolute measure, but I don't think you can fully discount it either; that's a discussion in itself, though, and not one to enter into here. I find the last criterion especially fascinating. It sounds reasonable and I think it may even rise to the level of common sense; imagine, however, if FSU runs the table in the regular season and then wins the ACC championship game but Jameis Winston gets injured and will definitely miss the bowl game.

Would the committee leave FSU out of the playoff? Would it drop them from the #1 seed to the #4, resulting in a less-favorable semifinal matchup at a less-favorable venue? What would be the hue and cry if they did? The members no doubt wouldn't be bothered by it, but would the powers that be behind the committee respond to pressure to change the protocol?

The voting process generally will include seven rounds of ballots through which the committee members first will select a small pool of teams to be evaluated, then will rank those teams, with the top-ranked teams being placed in the rankings in groups of three or four. Individual ballots will be compiled into a composite ranking. Each committee member will independently evaluate an immense amount of information during the process. This evaluation will lead to individual qualitative and quantitative opinions that will inform each member’s votes.

There's actually a lot more detail on the voting protocol at the CFP site, which I'll try to summarize here as best I can. Each of the 13 voters will make a list of his or her top 25 teams, in no particular order. Any team appearing on at least three voters' lists will be included in the general vote. From that common list, each voter will select his or her six best teams; from that vote the top three teams will be broken out, and then the vote-for-six/choose-three process is done twice more for teams 4 through 9, and then it switches to a vote-for-eight/choose-four process for teams 10 through 25. At the end, the committee will have a common top 25 which will then be publicized. The votes are all conducted via secret ballot.

The committee will meet in person weekly beginning at mid-season to produce interim rankings before selection weekend.

The first meeting will be after Week 9, on October 27th and 28th. Makes a lot more sense than having so much ride on the silly pre-season AP and Coaches' polls.

There will not be one single metric to assist the committee. Rather, the committee will consider a wide variety of data and information.

Again, more detail on this is available on the above-linked voting protocol page at the CFP site, but it's worth noting that this doesn't mean there's no standard for the data they'll consider. The playoff has retained a company called SportSource Analytics to provide deep-dive data over the course of the season. They also emphasize that they will not use any single metric analogous to the RPI used for college basketball tournament seeding.

What Will the Actual Playoff Look Like?

There shall be no limit on the number of teams that may participate from one conference in the playoff semifinals and the associated bowl games.

Theoretically, we could have an all-SEC playoff; given that ESPN has no control over the selection process, though, that seems unlikely. In the abstract I don't have a problem with this rule - the best teams are the best, no matter where they play.

When assigning teams to sites, the committee will place the top two seeds at the most advantageous sites, weighing criteria such as convenience of travel for its fans, home-crowd advantage or disadvantage and general familiarity with the host city and its stadium. Preference will go to the No. 1 seed.

This is the next best thing to hosting the semifinals on the higher seed's campus (which I'd prefer, but I understand the bowl game economics are too powerful to ignore). I really like that they actually made "convenience of travel for its fans" a consideration. The bowls that make up the rotating pool for the semifinal games are the Rose, Sugar, Orange, Fiesta, Peach, and Cotton. In years when they don't host a semifinal - like this one, as the first semifinal hosts are the Rose and Sugar - the Fiesta, Peach, and Cotton, which no longer have any contractually pre-determined participants, will have both participants decided by the committee, with preference going to conference champs displaced by the semifinal game and the top "non-contract" (non-AQ, in BCS terminology) champion.

The two semifinals and four other premier games will be played in consecutive triple headers on New Year's Eve and New Year's Day. A return to the compelling football bonanza of years past - I love it. The national championship game will generally be played on the first or second Monday after the semifinal, depending on how the dates fall; never less than seven days between games, nor more than 13.

The committee will also try to avoid regular season rematches in bowl games, and will try to avoid placing the same team or conference to the same bowl game over and over.

What About Conflicts of Interest?

If a committee member or an immediate family member, e.g., spouse, sibling or child, (a) is compensated by a school, (b) provides professional services for a school, or (c) is on the coaching staff or administrative staff at a school or is a football student-athlete at a school, that member will be recused. Such compensation shall include not only direct employment, but also current paid consulting arrangements, deferred compensation (e.g., contract payments continuing after employment has ended) or other benefits.

A valid question, to be sure. We'll review the members of the committee, the pre-defined recusals, and what that means for the voting process in a future post. The committee at least provides for an acknowledgement of conflicts of interest, though; beyond that, we have to trust in the reputation and the gravitas of the members. Of course, one shouldn't mistake an ability to keep a stoic expression with gravitas, but we'll cover that later.

So What Does It All Mean?

Much too early to tell, but I'm cautiously optimistic that this will be a significant improvement. I like the limited-scope playoff, I love the return to the New Year's smorgasbord, I like the latitude that the committee has in making their evaluations, I like how late in the season the first ranking will be published. Let's hope the system can live up to its potential.