An Assessment of College Football's BCS System

Michael Francis - www.college-football-blog.org

The 2008/2009 season of college football is over and done. The Florida Gators have been crowned the National Champions. Several other teams that went to BCS bowls still believe they should be holding that trophy, and the state of Utah is suing the BCS for unfair practices after they completed their second undefeated season without invitation to the big dance. But how did the BCS actually do this year? Let's take a look.

Whether you agree with the BCS ranking system or not, I would maintain that few people would argue that the best team in the nation could not be found within the top ten of the BCS at the end of the year. The major exception? TCU, and they finished 11th. Here's what the top eleven looked like prior to bowl season:

1 - Oklahoma
2 - Flordia
3 - Texas
4 - Alabama
5 - USC
6 - Utah
7 - Texas Tech
8 - Penn State
9 - Boise State
10 - Ohio State
11 - TCU

So, if the top eleven teams in the nation as chosen by the BCS seems pretty accurate, then perhaps the system is fair, at least as far as the regular season goes. It is extremely difficult to argue that the system is inaccurate in ranking teams that have never played, a case Utah will be attempting to make, but mostly everyone, including those in favor of a playoff, can agree that all the major players appear on this list. Now, if the BCS is reasonably accurate prior to bowl season, let's take a look at the bowl season itself.

During this bowl season, there were five BCS bowls, five bowls containing two BCS ranked opponents and five bowls containing one BCS ranked opponent. Of the five BCS bowls, three were upsets, at least according to the BCS rankings. Florida (2) defeated Oklahoma (1) for the National Championship, undefeated Utah (6) won out over long time number one ranked Alabama (4) and Virginia Tech (19) emerged from a chaotic ACC conference to beat Cincinnati (12).

Of the three bowl games consisting of two BCS ranked teams, there were also three upsets. TCU (11) and their impressive defense won out against then undefeated Boise State (9), a much improved Oregon (17) team took out the highly regarded Big 12 power Oklahoma State (12) and Ole Miss (25), this season's giant slayer toppled yet another monster in Texas Tech (7).

Of the five bowl games that showcased an unranked team against a BCS ranked opponent, ALL FIVE were upsets. Arizona over BYU (16), Oregon State over Pittsburgh (20), Vanderbilt over Boston College (24), LSU over Georgia Tech (14) and Tulsa over Ball State (22) all displayed the abilities of unranked teams.

In case you're not keeping track, this means that of the 15 games involving BCS ranked teams, only four of them went as the rankings would suggest. Of those four games, two were one possession games decided in the final minutes. Now, the question is, with all these harrowing statistics, does this mean that the BCS is a bust?

The simple answer is yes. The BCS system is meant to rank teams, just like the march madness rankings in college basketball. When a higher ranked team loses to a lower ranked team, fans are surprised, excited and upset. Last year in the march madness tournament, all number one seeds advanced to the Final Four, demonstrating what many called a successful ranking system. In stark contrast, the BCS this year predicted less than a third of the bowls involving ranked opponents correctly. A majority of upsets generally indicates the system to rank the teams is flawed.

But this is not the only answer. When considering the BCS system closer, it is possible to argue that the results of these bowl games in no way reflect on the performance of the BCS system's ranking methods. The BCS is built to rank teams with respect to their body of work thus far, it is not meant to predict the ability of the teams overall. This is a key factor that I believe most people fail to keep in mind when discussing the BCS. Big East champion Cincinnati and ACC champion Virginia Tech, for example, found themselves ranked 12 and 19 respectively, despite winning their conferences. This is because the strength of their opponents and perceived power of their conferences are not high enough to merit top ten rankings. Similarly, this is why Utah and Boise State found themselves playing for trophies other than the National Championship, despite having been undefeated at the end of the season. Their perfect regular seasons were not as impressive as one loss records in stronger conferences.

As such, it can be said that the BCS accurately ranked teams prior to bowl season based on their body of work. No doubt, these rankings would look differently after the bowl games completed, if the BCS was computed again. If teams, such as Utah and Boise, believe they should be higher in the rankings, they need to begin scheduling more difficult opponents, a concept which both teams perhaps have taken to heart as Oregon appears on both schedules in the upcoming year. After all, even with the supposed success of the march madness ranking a year ago, this was the first time in the history of the tournament that it has occurred.

As a final thought on this subject, I would like to take a quick glimpse at the components of the BCS ranking, as the supposed failure of this system is often blamed on the inclusion of computer rankings. I for one am a proponent of computers and technology in general, so I'd like to defend them given the opportunity. So, if the top 25 teams were ranked solely based on the USA Today Coaches Poll, there would have been just as many upsets. The Coaches Poll correctly predicts Oregon winning out over Oklahoma State, but incorrectly believe Northwestern would beat Missouri. If the rankings were instead based on the Harris Poll, a conglomeration of media and other individuals with college football associations, the results again, would remain the same. The Harris Poll correctly chooses Florida as the National Champion, but again fails with respect to Northwestern. If, however, the rankings were created solely based on the computer averages, the number of upsets would actually decrease, by one. The computers, in turns out, are the only ones that have faith in undefeated Utah. Perhaps the state of Utah should be arguing against the people's involvement in the BCS rankings, and not the technical system they're based on.