We should expand the college football playoff from four teams to eight. This would help ensure we don’t exclude teams that have a legitimate chance of winning the championship, while avoiding the addition of too many games to the schedule.

~~~

I’ve been waiting for a college football playoff for many years. After watching so many seasons with bowl games that yielded multiple potential claimants to the national title, I’m relieved the playoff has arrived at last. It’s a good step forward, but it’s still not yet where it should be.

The goal of the playoff is to determine the best team in college football. As with other playoff systems, such as the National Football League or college basketball, a good approach is to have the top teams play each other, with the winners advancing until only one team remains. Unlike previous college football systems (the bowl system and the Bowl Championship Series), the current four team playoff achieves this and is therefore a good step forward. However, some key questions I consider are: how many teams should be in the playoff, and what should determine this?

I think there are two primary criteria. The first is the number of rounds of a playoff that can feasibly be completed given the number of games college football teams already play in the regular season and the logistics of scheduling around Christmas and New Year’s. How many extra games should teams be expected to play? I think for college students one or two extra games would be fine, and isn’t significantly different than what has been done for years. I worry that four would be too many—most colleges play between eight and twelve games in a season, so four additional would be a significant increase. I think three games is borderline.

The second criterion is the likelihood of having the right teams in the playoff. I find it helpful to think about this in the opposite way: what’s the likelihood of excluding the eventual potential champion from the playoff? I’m concerned that starting with a group of only four teams is overly limiting—there’s a significant risk that the best team is not included in the playoff. I don’t think the goal of the playoff should be to enable Cinderella to dance at the ball and potentially steal the show like the 64-plus team college basketball tournament. But I think the goal should be to minimize the likelihood that the best team doesn’t get a chance to compete for the championship. I think the 2014 playoff was a good example of this: Ohio State won the championship, and there were many people who didn’t think Ohio State deserved to be in the playoff at all. Meanwhile Baylor and a few other teams with very compelling seasons (perhaps better than Ohio State’s) were excluded. Did we exclude a team that could legitimately have won the championship? Selecting such a small group of teams also makes the system overly arbitrary and dependent on the selection committee.

I think we can do better. While a four-team playoff is a good step, it’s too small. The potential alternatives are eight, twelve, or sixteen-team playoffs. In considering these alternatives, I think eight is the right number. If we select the top eight teams it’s highly unlikely that we are excluding the potential national champion. With the four-team structure there was significant debate about who the fourth team should have been. I think it’s quite possible that the one or two other teams who were not included could have had a legitimate chance of winning the title based on what they had shown during the season. But if we took the top eight teams, while there would still be debate about who the eighth team should be, I would worry less about the consequences of this decision. If a team was ranked ninth or tenth at the end of the year, how likely is it that they’re the best team in the country, and that therefore it’s a mistake to exclude them? I think it’s quite unlikely. So I think a playoff with eight teams gives us an excellent chance of identifying the best team.

An eight-team structure is simpler than a twelve-team structure which would offer byes to the top four seeds. This would also require the champion to play four extra games if they weren’t one of the top seeds, and  this is problematic. The sixteen-team format is also problematic because it would require four incremental games, and is unnecessary because teams ranked so low at the end of the year can be safely excluded. It’s also important to not undermine the importance of the regular season—it should not be the only thing that counts (otherwise we could just pick the top two teams), but teams should not be able to compete for the national title if fifteen other teams did better than they did during the regular season.

I hope the system will evolve in this direction, and I think it will. I can live with the four-team playoff (for now), and am much happier than I’ve been in the past. But in a few years I expect eight will be the solution everyone calls for, and I look forward to watching those playoff games!