Last week, I introduced this method of projecting game scores. If you want to see exactly how it works, hit that link. If you want to see how it did last week, scroll down to the bottom of the post.. Keep in mind that since these work off of season stats, they should get better over time. Also a reminder: for now, I consider absolute projections to be generally more reliable than percentage projections, though the percentage ones weren't bad last week.
Here's this week's projections in order of time of kickoff.
GEORGIA at VANDERBILT
| Team | ScorOff | OppDefEff | ProjScor | ScorDef | OppOffEff | ProjScor |
| Georgia | 25.83 | -1.57 | 24.26 | 30.67 | -12.10 | 18.56 |
| Vanderbilt | 13.60 | 2.05 | 15.65 | 18.80 | 4.63 | 23.43 |
| Georgia | 23.84 | Vanderbilt | 17.10 |
This method is projecting a headache-saving win for Georgia, but not by a particularly wide margin. The real story with these two teams is that UGA allows more points than its opponents score on average, and Vanderbilt scores a whopping 12 points below what its opponents allow on average. Between those two, something's gotta give.
ABSOLUTE PROJECTION: Georgia 24 - Vanderbilt 17
PERCENTAGE PROJECTION: Georgia 23 - Vanderbilt 15
----
ARKANSAS at FLORIDA
| Team | ScorOff | OppDefEff | ProjScor | ScorDef | OppOffEff | ProjScor |
| Arkansas | 34.75 | -18.39 | 16.36 | 32.25 | 11.75 | 44.00 |
| Florida | 33.25 | -2.18 | 31.07 | 7.25 | 10.40 | 17.65 |
| Arkansas | 17.01 | Florida | 37.53 |
Florida's monster defensive stats ensure that this one projects to a blowout, but I don't know how much I believe it. For one thing, Arkansas' defense has looked pretty good the past two weeks, and second of all, Florida's conservative offense this year hasn't been scoring as much as last year's offense did.
ABSOLUTE PROJECTION: Florida 38 - Arkansas 17
PERCENTAGE PROJECTION: Florida 41 - Arkansas 10
KENTUCKY at AUBURN
| Team | ScorOff | OppDefEff | ProjScor | ScorDef | OppOffEff | ProjScor |
| Kentucky | 25.20 | 0.06 | 25.26 | 26.80 | 10.83 | 37.63 |
| Auburn | 38.33 | 1.96 | 40.29 | 27.17 | 3.64 | 30.81 |
| Kentucky | 28.03 | Auburn | 38.96 |
This is another projection I'm a bit skeptical of, as Auburn's offense didn't look so hot for most of last week and Kentucky has a better defense than Arkansas does. If we see the Tigers score a lot fewer than 39 points, it will go to show why it takes a while into the season for this method to work well. I'll be very interested to see how Kentucky's offense does, since Auburn's defense hasn't been quite so lock down, but UK is out its starting quarterback.
ABSOLUTE PROJECTION: Auburn 39 - Kentucky 28
PERCENTAGE PROJECTION: Auburn 39 - Kentucky 29
----
SOUTH CAROLINA at ALABAMA
| Team | ScorOff | OppDefEff | ProjScor | ScorDef | OppOffEff | ProjScor |
| South Carolina | 25.20 | -14.91 | 10.29 | 19.20 | 10.48 | 29.68 |
| Alabama | 37.00 | -3.71 | 33.29 | 12.50 | -2.54 | 9.96 |
| SC | 10.12 | Bama | 31.48 |
South Carolina seems to be a fairly popular pick for giving Alabama a run for its money this weekend, but I'm not sure why. I guess it's because of the Gamecocks' 5-1 record, but they lost to Georgia and thanks to Ole Miss' collapse, they haven't beaten anyone of note. Did I mention they're going on the road? I agree with the projection in not really giving South Carolina much of a shot.
ABSOLUTE PROJECTION: Alabama 31 - South Carolina 10
PERCENTAGE PROJECTION: Alabama 29 - South Carolina 11
----
LAST WEEK: 4-1
ALABAMA vs. OLE MISS
Absolute Projection: Bama win, 29-16; Percentage Projection: Bama win, 21-17; Actual Score: Bama win, 22-3
The projected score for Ole Miss was always going to be too high because of the 45 points they put up on Memphis. If you take that game out, Ole Miss's absolute projection score drops to 9 points. That said, the absolute projection predicted an easy Bama win, and that's what happened. The percentage got really close to the Tide's score.
FLORIDA at LSU
Absolute Projection: UF win 32-9; Percentage Projection: UF win 22-11; Actual Score: UF win 13-3
The projected score for Florida was bound to be high in retrospect, given the overly conservative game plan from Urban Meyer. In the percentage projection's favor, Florida could have had 23 points without a strategy change, and if you take out the 56 points against Troy, it comes out to 18 points (which is fairly close). Either way, it projected a dismal score for LSU, and while both came out too high, LSU wasn't a danger to score much.
AUBURN at ARKANSAS
Absolute Projection: AU win 46-32; Percentage Projection: AU win 44-34; Actual Score: Arkansas win, 44-23.
This one was way off. Very little in either team's history this season indicated the result that we got was coming, except for maybe Arkansas' offensive outbursts against everyone but Alabama. It was wrong, but I think a lot of other people were too on this one.
GEORGIA at TENNESSEE
Absolute Projection: UT win 31-21; Percentage Projection: UT win 27-25; Actual Score: UT win, 45-19
I'm actually pretty proud of this one. Tennessee managed to run up a big score as Georgia collapsed, but a 31-21 Tennessee win was not anywhere close to what I saw anyone else predicting. Not only was it just two points off of UGA's score, but it predicted a rather high score for the Vols that drew at least two sets of skeptical eyes (though one set was my own).
KENTUCKY at SOUTH CAROLINA
Absolute Projection: SC win 25-18; Percentage Projection: SC win 22-21; Actual Score: SC win, 28-26
This one wasn't particularly accurate, but at least both picked out the correct winner.