Tracking the Top 25

Well, the regular season and conference championship games have concluded and I've come up with the final computer rankings for the Top 25. Yes, I've expanded it again adding another five teams. I still don't think the formula is quite right. For this one, I went with the winning percentage bonus because I think that needs to be in there. But, I still think I might be overweighting losses, but I am not 100% sure. I am going to run some tests on various formulas during the offseason to determine which one is best for next year. And who knows, maybe I can rank every Division I-A team then.

And without further ado (More notes at the end)...

1 Ohio State 12-0
2 Florida 12-1
3 Michigan 11-1
4 Boise State 12-0
5 LSU 10-2
6 Louisville 11-1
7 Oklahoma 11-2
8 Wisconsin 11-1
9 Virginia Tech 10-2
10 USC 10-2
11 Notre Dame 10-2
12 Wake Forest 11-2
13 Rutgers 10-2
14 Arkansas 10-3
15 Tennessee 9-3
16 Auburn 10-2
17 West Virginia 10-2
18 BYU 10-2
19 Texas 9-3
20 Boston College 9-3
21 Houston 10-3
22 California 9-3
23 TCU 10-2
24 Pennsylvania State 8-4
25 Nebraska 9-4

Conference Breakdown (lowest ranked in parenthesis)
SEC - 5 (16)
Big Ten - 4 (24)
Big East - 3 (17)
Big XII - 3 (25)
ACC - 3 (20)
MWC - 2 (23)
Pac 10 - 2 (22)
Independent - 1 (11)
WAC - 1 (4)
Conference USA - 1 (21)


I think the rankings are pretty accurate. Florida deserves a shot at the national title after playing the toughest schedule in the country and winning the best overall conference in the nation. Michigan's schedule was not as tough as Florida's. The Gators showed they can win games against the best competition (six games against top 25 teams and five games against top 25 defenses). Actually, the largest gap in my formula between any two teams is between Florida and Michigan. I created the formula with no idea how things would play out - so the rankings are based solely on a team's quality of competition and ability to beat that competition. By this formula, Florida's tougher schedule and ability to win games down the stretch make them the second best team in the country today.

To give you an idea of how close these rankings can get: The difference between Wake Forest and Rutgers is about .002. Houston and California are separated by about .003. Compare that to Florida which is ahead of Michigan by about .55.

Using these rankings, here is how I'd have the five BCS Bowl Games:
BCS Championship: 1 Ohio State vs. 2 Florida
Rose Bowl: 3 Michigan vs. 10 USC
Fiesta Bowl: 4 Boise State vs. 7 Oklahoma
Sugar Bowl: 5 LSU vs. 11 Notre Dame
Orange Bowl: 6 Louisville vs. 12 Wake Forest

Just a refresher on how I am calculating the rankings...
Teams get points for wins based on when they play the game (diminishing returns for wins earlier in the year to get a true representation about how a team is playing right now), what their opponent was ranked at the time of the game, where the game was played and their opponents winning percentage the day of the rankings are calculated.

Teams lose points for the opposite reasons. They lose points based on when the game was played (losses late in the season hurt more than losses early on), where their opponent was ranked at the time, where the game was played and their opponents winning percentage the day the rankings are calculated.

Bonuses are added for the team's winning percentage.

The rankings are supposed to represent who is playing the best football at the time of the rankings while still taking into consideration a full seasons worth of work.