Ken, I realize the polls here in GA are close. I wonder how accurate they are, though. Republicans are motivated in this election, & Dems are not. To what degree are the polls measuring that? Every time someone says they support Perdue or Nunn, is the next question, ‘How likely are you to vote?’? With no Obama on the ballot, I’d be surprised if the Dem turnout is that impressive. Of course there’s always voter fraud. So we shall see.
There is nothing more studied or labored-upon by pollsters than the turnout / likely-voter adjustments they make to the raw data. With a bit of noise, the adjustments get more accurate over time. You should assume that the polls are accurately measuring motivation, to the extent it impacts turnout, absent good reason to believe otherwise.
Also, as a general matter, the Republican turnout advantage in the South, especially in “low-motivation” situations, has been badly eroded by the Democrats GOTV operation: terrific technology and personnel, aided by a brace of voting options (absentee as of right, early voting, Sunday voting, etc.)