The year-end holidays season means different things to different people. For fans of American college football, late December and early January represent an opportunity to overindulge in a seemingly endless series of sponsored bowl games designed — in theory at least — to culminate in one team being crowned as the national champion for the season.
To remove regional biases and other flavors of favoritism from determining a national champion, several years ago the college bowls turned to a computer-based ranking system known as the BCS aka Bowl Championship Series. At the end of the season, the top two-ranked teams are paired together in a designated bowl game, the winner by definition becoming the best for the just-concluded college football season.
Alas, the BCS method of determination has generated its own critics over the past years, but none more than this season. In a bewildering outcome, the apparent top-ranked team after the final regular season games was excluded by the BCS algorithm, which instead has pitted what most consider the second- and third-rated teams [PDF: 8kb] to play each other on January 4, 2004 for the right to be called the national champion.
Many sportswriters, coaches and fans are again calling on the college football powers that be to change, if not scrap, the BCS methodology and revert to a system that builds in some degree of common sense rather than merely a mathematical formula. In response, the human wiz behind the BCS formula — Wes Colley — has come forward recently to explain and defend the current approach. He also has posted a detailed explanation titled ‘Colley’s Bias Free College Football Ranking Method: The Colley
Matrix Explained‘ [PDF: 168kb] on his ‘Colley’s Bias Free Matrix Rankings’ Web site.
According to the Colley introduction:
‘The method is based on very simple statistical principles, and uses only Div. I-A wins and losses as input — margin of victory does not matter. The scheme adjusts effectively for strength of schedule, in a way that is free of bias toward conference, tradition, or region. Comparison of rankings produced by this method to those produced by the press polls shows that despite its simplicity, the scheme produces common sense results.’
Fans of the University of Southern California, rated best in the country in season-ending polls as determined by sportswriters and coaches but not by the BCS, aren’t buying it.