No doubt my curiosity about this is fueled by being a homer, but if you look at the current Sagarin ratings, you’ll find that Georgia and Arkansas have identical 7-5 records, Georgia (5) sports a better SOS number than does Arkansas (13), Georgia has a better record against Sagarin’s top ten teams and top thirty teams than the Hogs do… and Georgia beat Arkansas head-to-head. In Fayetteville, to boot.
So why, exactly, does Sagarin have Arkansas ranked 16th, a whopping twelve spots ahead of Georgia? An inquiring mind wants to know.
Get The Picture 
Margin of victory.
LikeLike
Sagarin actually uses two algorithms, which he combines for the overall “Rating”.
In the ELO-Chess algorithm (which does not take into account margin of victory), Georgia is 17th and Arkansas is 19th. (This is the ranking that Sagarin is mandated to submit for the BCS, since it doesn’t use score margin).
In his “Predictor” algorithm (which DOES use margin), Arkansas is 17th and Georgia is 33rd. I suspect this is because Arkansas has a better average scoring margin than the Dawgs, albeit against a weaker schedule.
DH
LikeLike
http://en.wikipedia.org/wiki/Jeff_Sagarin
The big difference is the Predictor or Pure Points or margin of victory. UGA is ranked above ARK in ELO_CHESS which is the only factor the BCS takes into consideration, correct?
LikeLike
Blah, blah, blah. Algorithm? Al Gore Rhythm? Sagarin is like a social scientist: he takes his feelings and his hunches and puts them in a computer and thinks that makes his results more scientific, at least to the public. After all, his results come from a computer! If you discount his unnatural love for certain schools (the ones he coached at) Lou Holtz is right as often as Sagarin.
LikeLike
Algorithms? Who knew?
I always thought the Sagarin Ratings were done by Craig Sagar as in Craig Sagar’s done been trying to figure something out in the time it takes me to complete my Mumme Ballot which is between 0 and 60 seconds…
LikeLike
Chuck, Sagarin’s rankings are just computer rankings, but they are fairly accurate. I’ve tracked the predictions each week on DawgRun, and they finished the season 79-18 (.814). Not perfect, but not shabby, either.
LikeLike