by Bryant Garrison & John Vincent, University of Alabama

The Bowl Championship Series

The top level of College football in the United States produces a national champion through a complicated computer and voting system called The Bowl Championship Series ( BCS ) (Wikipedia, 2009) .

The BCS is a selection system designed to give the top two teams in the NCAA Football Bowl Subdivision (FBS) an opportunity to compete in a "national championship game" (Wikipedia, 2009). This championship is intended as a substitute for a playoff system since the NCAA does not formally determine a champion in this category. The BCS relies on a combination of polls and computer selection methods to determine relative team rankings and to narrow the field to two teams to play in the BCS national championship game (Wikipedia, 2009). The winner of this game is crowned the BCS national champion, and is guaranteed at least a share of the national championship (Wikipedia, 2009). The system also selects matchups for the other prestigious BCS bowl games . The ten teams selected include the conference champion from each of the six BCS conferences (SEC, ACC, Big10, BigXII, Pac10, Big East) plus four others ("at-large" selections) (Wikipedia, 2009). The BCS was created by formal agreement among six conferences, and has evolved to allow other conferences to participate to a lesser degree (Wikipedia, 2009). This is an example of using a voting system to generate a complete ordered list of winners from both human and computer-constructed votes. Obtaining a fair ranking system is a difficult mathematical problem and numerous algorithms have been proposed for ranking college football teams in particular (Wikipedia, 2009). The BCS formula calculates the top 25 teams in poll format. After combining a number of factors, a final point total is created and the teams that receive the 25 lowest scores are ranked in descending order. The factors are:

• Poll average: Both the AP and ESPN- USA Today coaches' polls are averaged to make a number which is the poll average (Fans Only, 2008).
• Computer average: An average of the rankings of a team in three different computer polls are gathered ( Jeff Sagarin / USA Today , Anderson-Hester/ Seattle Times , and New York Times ), with a 50% adjusted maximum deviation factor. (For example, if the computers rank a team third, fifth, and twelfth, the poll which ranks the team twelfth will be adjusted to rank the team sixth) (Fans Only, 2008).
• Strength of Schedule: This is a team's NCAA rank in strength of schedule divided by 25. A team's strength of schedule is calculated by win/loss record of opponents (66.6%) and cumulative win/loss record of team's opponents' opponents (33.3%). The team that plays the toughest schedule is given .04 points, second toughest .08 points, and so on (Fans Only, 2008).
• Losses: One point is added for every loss the team suffers during the season. All games are counted, including Kickoff Classics and conference title games (Fans Only, 2008).

Problems with the BCS

At the end of the 2008 college football season, the University of Florida, nicknamed the Gators, was crowned champions of college football for the second time in three years. Although the Gators won the BCS title game, it is open to debate whether they were the best team. It is questionable whether the two best teams played for the title. This is because the BCS formula described above produced a championship game whose contestants ranking was based not only on their performances on the field, but also on how much the computerized BCS system de-emphasized each team's one loss based on when the loss occurred. The University of Florida and Oklahoma University, who competed in the national championship game, only had one loss at the end of the regular season, but so did seven other BCS teams.

Several universities such as Alabama, Texas, Southern California, Pennsylvania State, Texas Tech and Boise State all finished the regular season with one loss, and Utah even managed to go undefeated. This makes is difficult to evaluate which one loss team is better than the other because one-loss teams get rewarded for losing early in the season, whereas one-loss teams losing in the later half of the season get punished. For example, Texas defeated Oklahoma 45-35 on October 11, 2008 in a head to head match up on a neutral field in Dallas, Texas (ESPN, 2009). Three weeks later, Texas lost to Texas Tech, who, in turn, eventually lost to Oklahoma. Thus, the Big XII conference had three teams in the same division (Big XII South) with only one conference loss, and just one loss overall. So the team that would eventually go to the Big XII title was determined not only by the play on the field, but who was ranked higher in the BCS standings. Since Oklahoma had lost early in the season and did not lose again, their BCS ranking was higher than Texas, and they were selected to play in the Big XII championship. Texas, on the other hand, was punished for having their one loss so late in the season and was kept out of the Big XII championship even though they had beaten Oklahoma earlier in the year (Fox Sports, 2009).

The Southeastern Conference (SEC) provides another illustration of how the BCS formula operates. The University of Alabama was the only team in a BCS conference to go the entire regular season without a loss. They were ranked #1 in the nation before playing the University of Florida in the SEC championship game. The University of Florida was ranked by some polls as the #2 team in the country, even though they had suffered a loss at home to the University of Mississippi in the fifth week of the season. The University of Florida eventually beat the University of Alabama in the SEC Championship game and was rewarded with the right to play in the national championship game against the University of Oklahoma. The University of Alabama, with only one loss all season, was given the consolation prize of playing The University of Utah in the Sugar Bowl. Utah was undefeated during the regular season, but because they play in the Mountain West Conference, which is considered a minor conference, they were not given the opportunity to play in the national championship game. In addition, there were three universities that had only lost one game each all season. Thus, it could be argued that they had just as much right to play in the national championship game as anyone in the country (ESPN, 2009). However, since these other teams suffered their loss in the latter part of the season, their dreams of a national championship were dashed. One of these one-loss teams did lose early in the year, but in the end were left out for a different reason. The University of Southern California (USC) lost to Oregon State University in the fourth week of the season. However, they were eliminated from contention for the national championship because they played in what most experts considered to be the fifth, or at best, fourth best conference in the nation. Like other teams before them, USC lost a game during the regular season, but managed to play extremely well and win out through the rest of the year including their seventh straight PAC-10 conference title. However, the BCS formula punished USC for their loss because of the perceived inferior quality of the other teams in the Pac-10 conference.

Can the BCS be fixed?

Did the two best teams play in the National Championship game? Many expert commentators and college football pundits question this. Could the BCS formula be changed so college football could guarantee that the two best teams play for the national championship? Could a new system be devised to facilitate this? Given that the current BCS contract does not run out until after the 2013 season, it would appear that any drastic changes toward a playoff system in the foreseeable future is not going to be possible. However, even if the BCS formula could not be completely replaced it is possible that it could be modified. Below are some suggestions:

1. Let the fans, see the polls - Neither the coaches' poll nor the Harris poll reveals its weekly voting results until the last week of the regular season, when the final BCS standings are released. The Harris poll is available for fans to see, but the coaches' poll is not available until the last week of the regular season. For many years the coaches have resisted making their ballots public during the regular season. However, under media pressure, they are now made to show the final coaches poll, which reveals the BCS national champion (BCS, 2008).

2. Do not use Tie-Breakers - The BCS formula was not made to determine anything but the top two teams for the BCS title game. An alternative would be for the each individual conference commissioner to devise a tiebreaker for his or her own conference. This would mean that the BCS formula would not produce a tiebreaker (BCS, 2008)

3. Bring back Margin of Victory – After the 2001 season, the margin of victory from each team was no longer considered. This was designed to discourage teams from running up the score. However a disadvantage with this is that the computers do not watch the games and cannot distinguish between a one-point loss and a 50-point loss. The first fallout from this decision took place in 2003, when the University of Oklahoma was defeated by 28 points in the Big 12 championship game, yet still stayed No. 1 in the BCS standings because the computers were unable to distinguish a one-point loss from a 28-point rout (BCS, 2008).

4. Adjust the computer ratings - A problem with the BCS formula is that it mixes ordinal numbers with percentages. In the human polls the formula counts actual votes for teams. However, the computer uses rankings only. So when a team falls from second to third in the computers, the consequences are all but disastrous; but when a team falls from second to third in the polls, it could be just about a dead-heat, as far as the BCS standings are concerned (BCS, 2008).

5. Create the “PLUS-ONE” Formula: If college football is not going to have a playoff system to determine a champion, it could implement a PLUS-one system. This would allow for the top four teams in the BCS rankings to play each other in a national semi-final and a national championship game. At the end of the regular season when all the conference champions are determined, the top four teams in the BCS would get to play for the national championship. The number one ranked team would play the fourth ranked team, and the number two and three teams would play each other. The winner of both games would play each other for the National Championship. To get into the top four in the BCS, you would have to play very well throughout the season. The plus one system only adds one more game for just two teams, so for the vast majority of college football teams (117 out of 119 currently in the bowl subdivision) the schedule would not change. The biggest advantage to the plus one system is it lets in the elite without worrying too much about a plucky underdog ruining the season of the best team in the country. For example, from this past 2008 fall season, in a New Year 's Day bowl game, perhaps the Fiesta Bowl, the University of Oklahoma would have played the University of Alabama since the Oklahoma Sooners were ranked number one in the BCS rankings and the Alabama Crimson Tide finished fourth. Then, in another bowl game, such as the Orange Bowl, the Florida Gators would have played the University of Texas. A week later, the winners would have played off for the national title. Although this may not be a perfect solution, it does address the problem created by the fact that in only four years out of the 12 that the BCS has been in existence has there been just two undefeated teams at the end of the season. The PLUS-one system would provide an opportunity for two more teams to play for a national championship instead of the two who just happen to be number one and two at the end of the year because of the BCS formula.

Conclusion:

In college football it seems that the concept of a "national champion" which has been decided on the field remains an elusive almost mythical concept. However, do most people really want a system that produces a "true" national champion on the field? It can be argued that the current bowl system and BCS formula has its advantages even though we do not get to see the one-and-done "playoff" concept.

References

BCS website (2008). BCS standings . Retrieved on 04/09/09, from http://www.bcsfootball.org/bcsfb/standings

ESPN website (2009). Lobbyist speaks about changing the BCS. Retrieved on 04/09/09 from, http://myespn.go.com/blogs/others/0-4-382/Lobbyist-speaks-about-changing-the-BCS.html

Fans Only website (2008). New formula for Football Championship announced: Top bowl game to be chosen by polls, computers, formulas . Retrieved on 04/09/09 from, http://www.fansonly.com

Fox Sports website (2009). BCS system broken and widely unfair . Retrieved on 05/10/09 from http://msn.foxsports.com/cfb/story/9566868/BCS-system-broken-and-widely-unfair-

Wikipedia website (2009). Bowl Championship Series . Retrieved on 04/27/09 from http://en.wikipedia.org/wiki/Bowl_Championship_Series

What do you think?

Questions for discussion: if you want to answer the questions, please use the hereunder comment system !

1. Is the BCS good for college football? If so, why?
2. What changes would you make to the current BCS system?
3. Are you in favor of a play-off system?
4. Which solution proposed in this article would you prefer to see implemented?
5. Do you have any alternative ideas?