But neither team was even listed in a poll listed on the NCAA Web site. NCAA.com started a new set of power rankings this year, ones that solely take into account statistics.
Here are this week's top 15 teams:
1) Georgia Tech (4-0, 1956.88)
2) Wisconsin (4-0, 1819.64)
3) Alabama (4-0, 1657.25)
4) Michigan State (3-1, 1460.59)
5) Florida (4-0, 1450.42)
6) South Florida (4-0, 1332.52)
7) Virginia Tech (4-0, 1149.24)
8) Stanford (3-0, 1109.77)
9) Oklahoma State (4-0, 1103.61)
10) Missouri (2-2, 1073.46)
11) Baylor (3-0, 1057.29)
12) Oregon (3-1, 1039.20)
13) Georgia (2-2, 1014.35)
14) Florida State (2-2, 986.64)
15) California (3-1, 955.97)
Hey, I'm all for adding more polls that look at things in original ways. But this poll also shows why there's a need for human logic at times. I mean, check out spots No. 10 and No. 14. The poll doesn't have the aforementioned undefeated Oklahoma. But it does contain a pair of two-loss teams who have lost to the Sooners this season. Florida State even lost to Oklahoma in Tallahassee.
For what it's worth, here's the site's description of the poll:
"The NCAA.com Power Rankings are based solely on team statistics. The formula includes current stats for both offense and defense, taking into account teams' ability to produce positive results on both ends of the field. A strength-of-schedule factor is also included to reward teams that play more BCS opponents, and gives a disadvantage to teams that play a high number of non-BCS and FCS teams. The same SOS factor carries across the whole season for a team, meaning the weekly shift in ranking is accounted for completely by that week's statistics."
That's why Georgia Tech, which has produced mind-bending offensive stats, is sitting at No. 1. Although if the strength of schedule component was as important as mentioned, it's hard to see how Oklahoma wouldn't make it with wins over Missouri and Florida State, two teams in the Power Rankings' top-15. Texas was in the top 15 (at No. 13) after its win over UCLA, but was passed up after the bye week.
Count this one as another poll that has an interesting concept, but mediocre (at best) execution. Most of the time when people attempt to put together a formula, they back-cast it. That is to say that they come up with the formula, try it on previous seasons to see if the results are accurate, then hone it to perfection, or at least as close to perfection as it can get.
While there isn't any indication whether the creators of the poll did just that, the early results would certainly indicate that tweaking is needed.