The BCS provides a wonderful product. The postseason is filled with meaningful games, and half the teams competing get to finish their season on a high note with a bowl victory. As a Boise State fan, I look back with pride and a sense of accomplishment on our Fiesta Bowl wins to close out the 2006 and 2009 seasons. They were each a great game, and a great way to end an amazing season. Had there been a playoff in either year, there's a good chance that even had Boise State won a matchup in an early round, it would have been eventually followed by a loss to end the season.
I was reading this article on the Washington Post site/blog, and there was a comment by Bill Hancock, executive director of the BCS. I thought I'd offer this rebuttal of his comment. It is posted here for criticism/comments before I register on the Post site and post it there. How does it read? Is there anything else I should say, or something that doesn't come through clearly?
This is what the BCS system (and bowl system in general) does well. It provides a meaningful series of games to reward teams who finish with a winning record. For teams like SMU or Idaho this year--where winning records have been hard to come by recently--their wins in bowl games are all the more special. Teams like Idaho or SMU, despite their historic seasons, wouldn't have a chance at playing in a playoff.
While the BCS and bowl system excels at providing a great set of football games (and maybe a few clunkers) to end the year, it does not do well at determining a national champion.
When BCS supporters say that "every single week of the college football season matters," I ask which week eliminated TCU or Boise State or Cincinnati from contention? The answer is that they were eliminated from contention before the season started--that the national championship game was not determined by this year's games, but by this year's games and the result of so many past seasons. With Boise State's win in the Fiesta Bowl, they will probably be ranked high in the preseason polls. If things break their way (they finish the season undefeated--again--AND the teams ranked above them and closely below them lose a game), they might play in the BCS championship game next year. BCS supporters will point to this and say that the system works. I claim that even if such events happen, this isn't proof that anybody can win. Instead, this would be a culmination of 3 years of work on Boise State's part. Even in such a case, the championship game is determined by that year's games and the result of the previous two seasons. Even if I were to consider the BCS championship game as a legitimate championship, it cannot be said to be an accurate determination of *that year's* best team.
Mr. Hancock said that in the twelve years the BCS has been in existence, it has matched up the top two teams by the AP poll nine times. He sees this as a measure of success. Even using a similar metric (the AP poll vs. the BCS formula combining several polls), I see a 25% failure rate in identifying the top two teams. If a different metric were used--one that did not involve a popularity contest (which, again, is based in large part on past accomplishments)--this failure rate would be even larger. There is too much disagreement about which teams are the top two teams, and too many ways to compare teams to reliably pick just two teams as the best two in a given year.
Furthermore, Mr. Hancock claims that any way of selecting teams for a post-season tournament would have to rely on the same tools that BCS opponents disparage--the polls. This simply is not true. One easy way to select participants in a post-season tournament is to invite all conference winners. This will keep the conference regular season just as important, because any conference loss could eliminate a team from contention. It may also cut down on the tendency for schools to schedule "cupcake" teams for non-conference games--if a loss to a good team out of conference won't prevent a team from winning a national title, they might be more willing to schedule tougher teams for experience. Inviting all conference champions (however the conferences determine champions) will provide a pool of 11 teams. A tournament may want to expand beyond 11 teams for several reasons: to give independent teams a chance to compete; to invite other strong teams who did not win their conference; or to have a rounder number of teams (12 or 16) playing to make scheduling the tournament easier. Certainly, some outside system like polls may be used to seed teams or to invite the extra teams, but if a team does not qualify for the tournament, it would be because of something that happened that season, and not previous seasons. (It may be possible that two Big 10 teams finish their conference undefeated, but this scenario and that of one of the three independent teams going undefeated are the only two I can think of where a team might not play in the tournament despite an undefeated record. Expanding the field to 12 or 16 teams will easily take care of these situations.)
Opponents of a football tournament say that it would disrupt class too much. I have some sympathy for this, especially as early rounds may be scheduled at about the same time as semester exams at many schools. However, the regular season is equally disrupting at many schools. Boise State often plays Wednesday night or Thursday night games. These are class nights, and when the team has to travel, they will miss two or three days of class. Tournament games on Saturdays, especially games that happen in an intersession break when the students aren't attending class, would be less disruptive to their class schedules than long road trips and mid-week games. I see no reason why a postseason tournament would be worse for students than mid-week games or the NCAA basketball tournament, which takes place during the week in the middle of the semester. Successful student-athletes have managed these conflicts, and would be able to do the same with a postseason tournament.
There are too many teams in the BCS division of college football, and too few games played, to be able to accurately know the top teams. A postseason tournament that includes the conference champion for each conference is the only reliable way to determine the NCAA football champion.
(Edited to correct three spelling mistakes.)