Last Tuesday’s kos post noted that the early voting in Georgia had a high proportion of African American voters (37%), even though that group makes up only 29% of the state’s registered voters. Since then, this has dropped to 35.9.
But this number is still impressive, and there are several reasons why Georgia might turn out to be the key to Obama’s election.
First, the numbers. According to the Georgia Secretary of State, as of Friday, there were already 635,798 votes cast, 228,185 by African Americans (35.9%). In 2004, approximately 825,000 African Americans voted. Consider this in light of the following, from the Atlanta Journal-Constitution a few months back:
Over the past two decades, Democrats in Georgia have believed they needed to win 90 percent of the black vote and about 37 percent of the white vote to win statewide. That assumes African-Americans make up 25 percent of the total electorate casting ballots, as they did in the 2004 presidential election. About 72 percent of all registered black voters cast a ballot that year.
I think we can assume that Obama will win at least 90% of the black vote. Also, I would expect Bob Barr (former Georgia Congressman) to collect more votes than Michael Badnarik’s 0.56% in 2004 for the Libertarian Party. So Obama will likely need less than the 37% of white votes thought in the past to be necessary. He certainly won’t if he keeps up the current pace, which far exceeds the 25% of the voters in 2004.
Consider also the likely effect of early voting. The overall race is likely to narrow by November 4th, but as Obama banks early votes, that will not matter.
So assume a scenario where McCain is able to narrow the race enough to start taking back states like Nevada, Colorado, Ohio, Florida, New Hampshire, New Mexico, Missouri, North Carolina, and Virginia. A reasonable assumption in such a case might be that if McCain is winning so many of these other states, he’ll certainly win Georgia too. But the combination of the early voting and the large African American population (plus Barr’s presence on the ballot) may result in a surprise.
The African American turnout likely isn’t helping Senator Chambliss keep is seat, either.
Those Secretary of State numbers bear watching. (btw, I believe that these race data are being kept because of past violations of the Voting Rights Act.)
[Update] I just noticed this AP article that came out today, with somewhat old GA numbers, discussing the impact of early voting in several key states.