2016 Georgia Presidential and U.S. Senate Vote Estimates
After taking a look at the Georgia House and Senate districts, I wanted to take just a few moments to discuss the model predictions for the presidential and senate races.
For a review of the methodology go here:
For the State Senate estimates go here:
For the State House estimates go here:
Turnout Estimate
As a reminder, I’ve used four different turnout scenarios:
- 2008 Turnout
- Average of 2008 and 2012
- 2012 Turnout
- Adjusted Turnout (Based on changes in early vote electorate)
Here’s a chart of those turnout estimates:
In 2012, the electorate was just over 61% White and 30% Black. If the districts match 2008 or 2012 turnout, the White share of the electorate would be down about 2% and the Black share of the electorate would be up just slightly. This goes in the face of the recent polls and the makeup of the early voting electorate. As I mentioned in the methodology post, the registered voter population in 2012 was 59.15% White and White turnout was a little over 75%. This year, the White registered population has decreased about 2.5% to 56.74%. I’ve crunched some numbers, and to match the 2012 electorate of 61% White and 30% Black it would take 2004 level turnout for both demographic groups (80% White and 70% Black) and just under 2012 turnout from other demographic groups. It’s obviously plausible that this scenario occurs, but we only have a little while longer to find out if that’s the case.
It’s difficult to present every plausible scenario, so I’ll continue to just give the four scenarios I discussed before.
Presidential Estimates
The biggest question on the totals of both the presidential and senate races is what happens with the polling numbers for the third-party candidates. It’s unlikely that either Johnson or Buckley reach the current polling averages in the final tallies.
A few scenarios are possible. First, I believe that Johnson will likely receive around 3–5% of the total vote, so I’ve accounted for the top and bottom scenarios there. The next question is where does the surplus polling total go? Does it split evenly? Or does it push more heavily towards Trump?
I’ve accounted for a few different scenarios in the table below. In all the scenarios, the undecided vote is split evenly between the candidates. Beyond this, there are three options for the surplus Johnson vote (the polling numbers above the 3% or 5% expected vote total):
- Split evenly between the candidates
- 2-to-1 to Trump
- All to Trump
Each of those scenarios is accounted for within each turnout scenario, and I’ve highlighted the projected winner under each in green.
I’ll let the numbers mostly speak for themselves. I do believe that we will end up close to 2012 turnout numbers when all is said and done. I don’t think it will just be close to the overall 73% turnout in 2012 but also similar among the demographic group 2012 turnout. I also feel that Johnson will end up with about 3% of the total vote. I do also tend to think that the excess Johnson vote will lean towards Trump. Given that, Clinton needs to hope for those votes (the 3.74% above the final 3%) to split less than 2-to-1 to Trump. At the 2-to-1 to Trump rate, the model estimates a 0.26% victory for Trump.
Senate Estimates
I’ve done the same for the U.S. Senate race as I did above for the presidential numbers. As you can, this race is much less exciting. It’s not even looking there’s much of any chance for a runoff here. Barksdale would have to hope for a Buckley performance at or better than his ~7% average and to then take almost all of the undecideds. An unlikely scenario to say the least. I expect Isakson to clear the 50% threshold. The question is really just by how much.
— Following the POLITICS of today, for the generations of TOMORROW —
Politics for Tomorrow is a publication focused on progressive politics both nationally and in the State of Georgia.
Follow Politics for Tomorrow on Twitter here.
Follow Austin Wagner on Twitter here.
