I think I have a simple solution to the problem of deciding a national champion in College Football. Take the winners of all 11 conferences, and put them in a playoff. The tournament would last 4 rounds, and the top 5 teams would get a bye for the first round. This has the advantage of giving every team the chance to play their way a national title. If you win your conference, you get a shot. If you don’t win your conference, then it’s your own fault and you can’t blame voting or computers or anything else. The regular season remains incredibly important, which is one of the nicer aspects of the current system, and we get a meaningful championship where all 120 teams have a shot at it each year. To test this new approach, let’s consider the worst cases possible with the current system in terms of being able to find a fair solution.

• Three (or more) major conference champions are undefeated. Let’s say the winners of the PAC-10, the Big 12, and the SEC are all undefeated. In the current system, only two of them get a chance to win the title. The third is left outside looking in, even though they never lost a game. Going to my proposed system clearly solves this problem.
• There are no undefeated teams, but three (or more) contenders only have one loss. Again, the current system tries to pick the best two subjectively, but that decides things off the field. Being able to make your case to the media and on TV becomes a coaching skill. With my proposed system, it’s settled on the field. Note that you may run into a case where a team is considered a top contender by the polls but loses their conference. I say, tough shit. They should have won their conference.
• The only undefeated team is from a smaller conference, and the major contenders (according to the polls) all have one or more losses. The current system tries to decide who is better – the major conference contenders with losses, or the undefeated minor conference team. This is not where the debate belongs. Teams from the smaller conferences that have great years deserve a shot to at least play their way into the national title game.

All that being said, there are a few problems with my proposal, but I think I have answers to those as well. The first one is, at least in theory, easy to solve. Right now, there are just 3 teams that don’t belong to any conference, the most famous being Notre Dame. The simple solution is to force all teams into conferences. With 120 teams, there’s no way to fairly evaluate the best teams if you don’t divide them into groups and have them play each other. Having “special” teams outside those groups just doesn’t work.

The next two problems are kind of related. First, what do we do with all the current Bowl Games? Second, how can we make sure that students from the competing universities can still have enough time and afford to attend the games? If we make each playoff game a different bowl, the average student will only be able to make it to one (or two at most), and then they get into the case where they have to either go to the less meaningful early games or try and guess how far their team will make it. Again, I think I have a solution. The first three rounds of the tournament should happen after the regular season during the early bowl season. These games should be home-away games, with the higher ranked team getting the home-team advantage. The bowl system stays intact, but six bowls agree to participate in the new system. Each of those six bowls takes turns hosting the national championship game, with the other 5 hosting the remaining 9 teams from the tournament, as well as one at-large team from outside the tournament. The 11 conference champions and those 6 bowl games are the only special case here – the rest of the bowls can pick teams to participate however they wish.

Finally, there’s the matter of determining rankings, which can be tremendously important because the higher ranked teams get bye weeks as well as home-field advantage. It’s tempting to use a ranking system similar to the BCS to determine the higher ranked teams, which isn’t terrible, but I’d like to stick with the philosophy of keeping things deterministic. Since the teams are each representing a specific conference, I would suggest we rank the teams based on the past performance of their conferences, so that the average finish of a conference over the past 10 years will determine the ranking of each team in the playoff.

Alright – that’s my idea in all its glory. I’ve mentioned it to a couple of people at various points, and now I’ve written it down. Let me know what you think.