
What It Is: The AP Poll is a ranking that is compiled weekly by the Associated Press. Sixtyfive sports writers
who are members of the Associated Press rank twentyfive teams of their own
choosing. Using each writers’ rankings, a point total is assigned to each
team. A ranking (or vote) for first place gives a team 25 points. A ranking/vote for 25th place gives a team 1 point. The 25 teams that receive the most points are considered ranked.
Why It’s Good: The AP Poll is the longest running national ranking system. It has the most credibility of any system that rates/ranks college
football, as many schools only claim AP Championships. Sports writers are generally considered more knowledgeable about the different football teams than the average college football fan. Also, the different AP writers who vote for the poll publicly disclose their votes, so there is no secrecy in the compilation of the poll.
Why It Sucks: The first problem is that many writers are only able to watch the teams that they cover. Therefore, they can’t be informed enough to vote on every team in the nation.
Secondly, the voting of the writers tends to take most recent performance into account more than games that occur early in the season.
As such, teams that lose a game late in the season end up ranked below teams that lose a game early in the season.
Third, voting on college football provides rankings that make the news. Many news organizations have ethics rules that generally would consider this unethical since the media should report the news, not make it. This is part of the explanation why the AP organization refused to be the sole factor in determining the BCS championship.
Lastly, some of the voters are idiots.
BCS Consideration: The number of points accumulated by each team are divided by 1625. This number represents the maximum number of points a team can receive (65 votes for #1). This percentage is 1/3 of the final BCS total.
What It Is: The Coaches' Poll is a ranking that is compiled weekly by USA Today.
Much like the Associated Press Poll, sixtyone coaches of Division 1A teams rank twentyfive teams. Using each coaches’ rankings, a point total is
assigned to each team. A ranking (or vote) for first place gives a team 25 points. A ranking/vote for 25th place gives a team 1 point. The 25 teams that receive the most points are considered ranked.
Why It’s Good: Coaches are considered the most knowledgeable about what teams are the best in college football. Also, the Coaches poll has a decent amount of credibility. Although not as credible as the AP poll, it is used by ESPN/ABC to display the rankings of
games. The Coaches' poll is as highly publicized as the AP poll.
Why It Sucks: Although the coaches are considered knowledgeable about teams, they are busy running their teams on
Saturday, and therefore unable to see games on Saturdays. However, their votes are called in late Saturday night.
Another problem is that the coaches’ votes are kept private. Therefore, there is no oversight on how the coaches vote.
Anything that involves as much money as college football, there is the potential for conspiracy if things are kept secret.
As is the case with the AP Poll to place more emphasis on later games than early games allowing teams that lose early to be ranked higher than teams that lose late. Finally, due to the arrangement with the BCS, the coaches are required to vote the winner of the BCS Championship Game #1 at the end of the season. Forcing a coach’s vote isn’t a good thing.
BCS Consideration: The number of points accumulated by each team are divided by 1525. This number represents the maximum number of points a team can receive (61 votes for #1). This percentage is 1/3 of the final BCS total.
What They Are: The Computer polls are rankings created using different mathematical formulas.
While anyone with an interest and a slide rule can compile their own rankings, certain rankings have gained a measure of national recognition. Some of the recognized rankings are compiled by AndersonHester, Richard Billingsley, Wes Colley, Ken Massey, Jeff Sagarin, Peter Wolfe, the New York Times, and the Dunkel family. Each
computer ranking system follows its own rules, and its own set of data to produce the results. The purpose of the rankings also differs for each
system. Some try to find a result that is similar to the Human Polls (AP and Coaches). Others try to produce rankings that will find the best team regardless of conference alignment or division.
Why They’re Good: First, computer rankings follow a standard set of mathematical rules that are defined long before the games take place.
Second, the same rules are applied to every team so that bias towards teams is eliminated. Computer polls also tend to take the entire season into consideration which eliminates the lose early, win later bias of the human polls.
Why They Suck: Most people neither trust nor understand computers, and therefore the computer polls do not have much credibility. The general attitude is the human polls know what they’re doing, and the computers do
not as evidenced by the “tweaking” of the BCS over the last few years.
Another knock on the computer polls are that unusual win/loss scenarios produced by playing the football season may produce
unexpected results when applying a mathematical formula. The mathematical formulas themselves have also been manipulated by the BCS since it was declared that no computer rankings will be used if they take into account Margin of Victory (the number of points by which a team wins). The Dunkel Index dropped out of the rankings, and others changed their formulas to meet the BCS demands.
BCS Consideration: The Computer Rankings by AndersonHester, Richard Billingsley, Wes Colley, Ken Massey, Jeff Sagarin, and Peter Wolfe are used. The lowest and highest ranking are dropped and the remain four are used total points like the human polls. (25 points for 1st place, 1 point for 25th place). The total is divided by 100 to produce a percentage. The percentage is 1/3 of the BCS total.
