Get your popcorn ready on election night… Again I will insist that an Obama win is now almost certain… Up to 95% according to the models I use. The question is becoming how big will this win be ??? I mentioned in my last Full Electoral College Breakdown how a landslide win (375+ EC votes) was possible, what if I told you that states that many have written off got Obama to that mark ??? After the primaries I had indicated due to high African American turnout that I though states in the southeast were in play for Obama, particularly MS and GA with the Bob Barr effect perhaps much more the reason in GA. I had gotten away from that idea after the “Palin bounce” but recent polling and a new study about the Bradley effect is making me re-think that these are indeed Safe McCain states. I’ll get to that study in a second but first lets start with some numbers… We’ll begin in Mississippi…
Polls of this state have been sparse due to the fact its a safe Republican state…. In fact since 9/1 there have only been 3 polls done here, one by ARG who’s accuracy I question… on 9/9 Research 2000 did a poll indicating a +12 McCain win. This was at the height of the Palin bounce for the red team… But by 9/30 the last poll done in MS the VERY Republican tilted Rasmussen had MS in the McCain column at +8.
Since that time there has been no more polling in MS and the more trusted polls like Survery USA haven’t been there at all.
My thoughts on MS at the end of the primary were that with a 35% voting public that is African American that in theory Obama would need to only get 20% of the white vote to win this state. Now comes the kicker… a recent study on the Bradley effect shows that in some parts of the country there is a reverse Bradley effect…
The Bradley effect–a black politician doing well in the polls but badly in the actual election–has been much discussed but there is little data on how it would work this year. Now a team of researchers from the University of Washington has analyzed the data from the 32 Democratic primaries this year and come to a surprising conclusion. The effect existed in California, New Hampshire, and Rhode Island, but a reverse effect existed in 12 other states, largely in the Southeast. Obama did better in the elections there than in the polls leading up to them. Researchers speculate that some people may tend to give pollsters the “socially correct” answer and in the Southeast, supporting the white person is socially correct.
Hmmm is the only thought that comes to mind on this issue… now on to Georgia…
Where along with a high African American turnout likely also come these to “B” words… “Bob Barr”
This state has been polled more and by other companies first I’ll start with Survery USA 9/15 poll showed McCain up +16, but by 9/29 that lead had been cut in half by Survery USA to +8 most polls this month have been in the McCain +7-+8 range with an error % of 5 %… add in Bob Barr stealing votes from McCain especially if he alienates more true conservatives and this possible reverse Bradley effect and Election Night could be very interesting indeed. On 10/9 there was a McCain plus 3 poll for GA as well.
Organizing for Change 