Here is a little refresher lesion on how the BCS rankings work. The BCS is broken up into three "separate" components: Coaches Poll, Harris Poll and the Computer rankings. When all three of these polls are published they are done so with an average ranking of one to twenty-five. These rankings are misleading as the 2nd place team is not a set number of points behind the 1st place team and the 3rd place team is not a set number of points behind the 2nd place team. Instead all three of these systems are based off of a percentage score, kind of like a midterm. In fact the easiest way to think about the BCS is to imagine it as a class with three graded tests and all 120ish BCS schools as students in this class. These three graded are averaged together for your total grade.
The Coaches Poll is made up of 59 FBS coaches who are members of the American Football Coaches Association. You can find the list of coaches in this year's Coaches Poll at the link at the bottom of this blog. Each week all 59 coaches get a ballot were they rank what they think the top 25 teams are. For each 1st place vote a team gets they receive 25 points (24 points for 2nd place, 23 points for 3rd place... all way way down to 1 point for 25th place). The BCS then take each team's total points and divides it by the total number of points possible to get a "grade". This year the total possible points are 1475 (59*25). That means that each coach can influence a team's BCS percentage by 0.59% or by 0.022% for each ranking he gives a team. For example if a coach ranks BSU 3rd instead of 4th then BSU's BCS ranking would move up by 0.022%.
The Harris Poll rankings/grades works the exact same way that the Coaches Poll does. Each Harris Poll members gets a ballot to fill out, the rankings are then inverted on a 25 point system and then each team is assigned a percentage based on the total number of points possible. The number of participants and who participates in the Harris Poll each year changes, but there are usually around 120 voters in the Harris Poll. Each conference and the three independent schools all submit a list of former players, administrators and current/former members of the media (sorry OBNUG no bloggers) that they would like in the Harris Poll (last year there was 300 nominations). The Harris Poll committee then randomly selects about 120 of these people to be Harris Poll voters. This means that each Harris Poll member can influence a team's BCS percentage by 0.27% or by 0.011% for each ranking he gives a team. See example above. The Harris Poll does not come out until about week six and as far as I know they have not released the Harris Poll members for 2011 yet.
Some smart guys came up with what they think is the best way to rank college football teams and who am I to judge? I am going to post a link to the computer rankings at the bottom of the page are you can read more about why they think their ranking are the best from their site. Just like the two polls the computer rankings do their magic and rank each team. There are six computer rankings, but each team's highest and lowest computer rankings are thrown out and the remaining four are used in the BCS. The BCS then takes the top 25 teams in each computer rankings, the rankings are then inverted on a 25 point system and then each team is assigned a percentage based on the total number of points possible. This means that each computer can influence a team's score by either 0% (if it gets thrown out) or by 8.3%. This also means that for every ranking a computer rises or drops a team; then that teams final BCS percentage rises or falls by 0.33%.
Wrapping it up.
The BCS then takes the three scores/test and then averages them together for a final BCS score.
Coaches Poll Members