Sagarin’s computer is still on drugs.

The week before, 46th.  Then Georgia beats a one-loss Sun Belt team and is now… wait for it…

Seventh.

Sagarin has Notre Dame tenth.  Where Georgia will stand if it wins Saturday night is something only Gawd and the computer’s dealer knows.

Advertisements

9 Comments

Filed under Georgia Football, Stats Geek!

9 responses to “Sagarin’s computer is still on drugs.

  1. There had to be a data error of some kind when last week was done. Based on his rankings, I would still be more than happy to take Georgia -4 1/2.

    Like

  2. dawgfan1995

    My guess is that if Georgia beats Notre Dame, Sagarin’s computer will drop the Dawgs to about 100 or so because that makes as much sense as everything else his computing has spit out so far.

    Like

  3. Timphd

    Oregon is ahead of Auburn, despite losing to them. That is why I hate computer rankings. One of the reasons.

    Like

    • Whiskeydawg

      I didn’t know anyone was still using VisiCalc. Got to watch that garbage in, garbage out stuff.

      Like

    • Macallanlover

      I would probably rank Oregon ahead of AU too. Everyone acts like a better team never loses. I understand the wanting every result to mean something truly definitive. And it does, it is an example of one possible result. I realize most want a school that beats another to always be ranked above them, but it often doesn’t as we get near the end f the season and we have a larger view of each team. Winning a game doesn’t mean it would be that way all the time, or even the majority of the time. You might rank AU a head of Oregon in a poll and I would not complain at this point, but let AU lose 3 games and Oregon finish with that one loss in December and I would expect Oregon to the higher ranked team. And if they played again next week, I would take Oregon in a bet with you.

      Sagarin gets better as more games are played but I never think his power rankings are the best, and his may be among the worst for computer power polls. Computers all have a human bias at the “garbage in” phase, impact of that diminishes each week as more games provide interactions of results between the teams’ actual performance.

      Like

  4. There was a data error. Had to be.

    Like

    • Dawgoholic

      Actually it sort of makes sense if the ranking is based on this season only and largely based on the quality of your schedule and your opponent’s schedules as to wins and losses. Such a ranking will take several weeks to sort itself out as there simply isn’t enough data for anything reliable.

      Now if the ranking included ypp and the like it would be accurate much sooner.

      Like

  5. FlyingPeakDawg

    If faced with the choice between Sagarin or 1000 monkeys with typewriters to create something that makes sense, I’m choosing the monkeys every time. Kind of like what we have now with the CFB committee when you think about it.

    Like

  6. Bryon Zazueta

    SOS is factored in which is why UGA was so low. The ratings still work better than any other handicappers that promise a 70-80% win rate which is BS!!
    It is only on drugs if you don’t know how to use it otherwise it is golden.

    Like