By Zen Prise from The Real Right:
The Deep South boasts the largest concentrations of blacks in the country, so shifts in turnout can have a significant effect on election results. Based on a black voter participate rate of 90%, Obama would only need a very small fraction of white voters in Georgia, Mississippi and Louisiana.
Don’t believe me? Well I come armed with data to support my claim, so keep an open mind.
My entire premise is based on very high turnout rates among black voters. The Deep South boasts the largest concentrations of blacks in the country, so shifts in turnout can have a significant effect on election results. The model I’ve built assumes that black voters participate at a rate of 90%, a dramatic increase to be sure, while white voter turnout only increases to about 65%.
To get started, let’s look at the demographics of the Deep South:
Now let’s apply our voter turnout model to these numbers to see what the demographics of the voting population looks like.
The third column is the percentage of white voters that Obama needs to get in order to win. Let’s compare these numbers to John Kerry’s totals among white voters in 2004:
So right off the bat, it’s clear that Obama needs far fewer white votes than Kerry. Given some of the racial factors at play, that’s a good thing for Obama. But how realistic is it for Obama to actually get the white vote percentages that he needs? Luckily, the Democratic primaries were very competitive this year, so we have some good data on white support for Obama in the Deep South.
There has been a lot of confusion about these numbers. The problem is people are comparing apples and walruses, so to speak. The numbers in the table are the absolute points that Obama will get from his share of the white vote. Add that to the black vote margin to get his projected outcome.
Here are the white vote share % you're looking for: GA (23.4%),SC (13.45%), LA (10.21%),MS (14.06%) and AL (13.03%). My next post (linked at the bottom of the diary) basically predicts that Obama will win GA, LA and MS, but lose SC and AL
The columns here are a bit wordy, so let me explain my assumptions and calculations. The first column is a repeat of the final column from the previous table. The second column simply calculates the % of white voters that supported Kerry that Obama needs to win over. The third column is Obama’s white vote percentages in the Democratic primary.
The fourth column is really important. I assumed that the pool of white Democratic primary voters roughly maps to the white voters that supported Kerry in ’04. Using that assumption, we can predict that Obama starts off with 15% of the white vote in Georgia, for example. In other words, Obama would expect to get 43% of the 34% of white voters that Kerry got. He only needs 11.49% (as a percentage of the total electorate) of the non-black vote to win, so he already projects to win the state.
The final column represents this percentage as a percentage of the white votes cast for other candidates (Clinton and Edwards) in this year’s primary. As you can see, Obama already has all the support he needs to win MS, LA and GA. In South Carolina, Obama needs about 4.14% to win. That works out to 16% of Clinton’s white voters. In case it’s not clear, the reason for doing this is because it is unlikely that white support for Obama will come from the pool of white voters that supported George W. Bush.
There are basically two assumptions that this model is based on that can be challenged. First and most obvious are the turnout assumptions. The second is the assumption that the voters in the Democratic primaries map proportionatel to Kerry voters. ( I guess one other assumption built into the model worth mentioning is that the state level demographics cited in the beginning are identical to the demographics of the voting age population of each state. That could swing the numbers in either direction by a couple of points.)
In case you’re wondering, even if you’re conservative and assume a 75% turnout among blacks, Obama still needs less than 50% of Clinton supporters to shift to Obama in order for him to win LA, MS and GA. South Carolina comes in at 33% and Alabama at about 52%.
Even though (with the exception of Georgia) no polling outfit is giving Obama much of a chance in the South (nor for that matter is Obama actively campaigning there), I’m willing to go out on a limb and say that Obama will almost certainly win Mississippi, Louisiana * and Georgia, where the Bob Barr effect will also be at play. Obama has a good shot in South Carolina as well. Alabama is the only long shot in the group.
Another important point is that this demographic tidal wave will probably take down Sen. Chambliss in Georgia, and Sen. Wicker in MS, assuming Obama voters vote for Dems in the down ballot races as well.
So to all you Obama supporters in the heart of the old Confederacy, your vote really matters. If you or someone you know thought you couldn’t paint your state blue this Tuesday, show them this article for motivation.
*
The demographic breakdown is from 2005, before Hurricane Katrina. Any changes in the size of the black population as a percentage of the total population of Louisiana would affect Obama’s chances of victory.
I've created a projection of Obama's white voter share in the South to go along with the black turnout adjustment in this article. Check it out!