Well, if there was ever a logic to completely scrub preseason logic with high-concentrate brain bleach, last week was it. Most of the SEC went as planned, but a few teams gave us plenty of 'discussion material'. Not everybody was ready for the opening of the season. My ballot is here, and after the jump I'll include the ballot exactly as our computer ranking system spits it out (along with explanations).
Just remember, I place only nominally more value in a week 1 ballot than a preseason ballot. Feel free to disagree mightily.
And, for comparison, here's the computer rankings. As a quick reminder, the scores in each individual columns are the number of standard deviations that a team's performance deviates from the mean. Positive is good in all cases. The 'SCORE' column is a weighted sum, where the columns to the left are weighted more heavily than the columns to the right. (And for those keeping score from last season, I haven't changed the weights yet. I know they're wonky, but I'm waiting until all D1-A teams play a game before tinkering, as the empty references screw up the computer calculations.)
| SEC | Team | WL | SOS | PED | RD | 3DO | TD | PEO | OPPG | TO | 3DD | SCORE |
| 1 | Alabama | 0.694 | 1.552 | 1.118 | 0.547 | 1.388 | 1.216 | 1.100 | 1.156 | 1.629 | 1.787 | 100.993 |
| 2 | Mississippi St. | 0.694 | 1.814 | 0.139 | 1.084 | 0.523 | 0.722 | 2.705 | 0.893 | 1.443 | 1.416 | 95.535 |
| 3 | Georgia | 0.694 | 1.328 | 1.296 | 1.386 | 0.337 | 1.589 | 0.255 | 0.893 | -0.184 | 1.061 | 90.610 |
| 4 | Arkansas | 0.694 | 1.267 | 0.068 | 0.659 | 1.316 | 1.120 | 1.057 | 1.156 | 1.019 | 0.541 | 82.267 |
| 5 | Tennessee | 0.694 | 1.028 | 1.464 | 0.916 | -0.533 | 1.478 | -0.422 | 1.354 | 1.172 | 1.519 | 77.237 |
| 6 | Auburn | 0.694 | 1.117 | -0.271 | 1.062 | 0.523 | -0.304 | 2.580 | -0.358 | 1.773 | 0.468 | 68.795 |
| 7 | Florida | 0.694 | 1.153 | 0.944 | 1.498 | -1.061 | 0.921 | -0.121 | 0.564 | -1.582 | 0.541 | 60.984 |
| 8 | South Carolina | 0.694 | 1.476 | -0.021 | 0.793 | -0.110 | -0.606 | -0.085 | 0.498 | 0.426 | -1.163 | 47.022 |
| 9 | Kentucky | 0.694 | 0.728 | 0.919 | -0.583 | 0.280 | 0.086 | 0.047 | 0.300 | 0.570 | 0.541 | 45.677 |
| 10 | LSU | 0.694 | 1.417 | -0.927 | 1.274 | -0.004 | -0.860 | 0.341 | -0.227 | -0.726 | 0.747 | 39.459 |
| 11 | Mississippi | -1.428 | 0.536 | -0.902 | 0.390 | -0.004 | -0.216 | 1.475 | -1.873 | 0.680 | -0.797 | -51.189 |
| 12 | Vanderbilt | -1.428 | 1.047 | -2.525 | -0.057 | -0.695 | -0.296 | -0.331 | -0.161 | 0.282 | 0.272 | -78.365 |
Alright, the explanations.
First, the SoS is a little askew at this point, thanks to Boise State and Virginia Tech. Sagarin won't update his preseason rankings until those games are finished (which makes sense), but the ballots are due this evening. Sooo.... they're still the preseason SoS estimates which, for this time in the season, is close enough for me.
Now for the edits I made. First was to bump Auburn ahead of Tennessee. This was done solely due to Auburn being in their second year with their head coach and Tennessee being brand new. Next was bumping Florida down: South Carolina played a much better game than Florida (as did Kentucky), so they rightly move up. Arguments can be freely made between Florida and LSU, but I just wasn't inclined to worry about them right now. No big deal though; both teams have 11 games to atone for their opening weekend miscues. Vandy went ahead of Mississippi for two reasons: one was that they played a better team (and still very nearly won), but the other was that Mississippi's failure to close was, in my eyes, the more egregious weakness of the two teams.
One final note: if you're wondering why teams that won have a lower absolute value to their W/L than teams that lost, that's because of the sheer number of 1-A vs. 1-AA games. Most 1-A teams won, so the average W/L is skewed positively, which means that 1-0 is closer to the mean than 0-1. It makes some individual conferences look odd, but it works out over the course of the whole 1-A, which was what this was intended to do.