The October Update
A lot has happened since my last update, but we’re essentially still in the same spot: Trump continues to prove he’s unfit for the presidency, Georgia’s electoral votes are in play, Jim Barksdale’s campaign (or lack thereof) has him down double digits even though Isakson continues to support Trump, and the Georgia Democrats don’t have enough candidates to win the statehouse.
While everyone is talking about the Senate race and the Presidential race, I’d like to continue discussing the Georgia General Assembly races. We haven’t seen much discussion as of late, but WABE just had a piece on the state of the race. They said it was “unlikely” for the Georgia Democrats to take back the House, but it appears no one did the math. Just a quick reminder: there are not enough Democrats on the ballot to take back either chamber.
The WABE piece didn’t get into the details, but that’s what I’m here for. Georgia was blessed with multiple polls dropping last week. I’ve taken the time to go through those polls and update my calculations based on the new polls. I was only able to use the AJC poll and the Landmark Communications poll. These were the only polls to include detailed enough crosstabs to extrapolate to the demographic calculations. I also updated the registered voter numbers as of October 1st.
A few notes on my assumptions here. I’ve weighted the polls more heavily to the AJC numbers because the polling company is rated higher than Landmark Communications. I took the numbers from both and came up with averages for the presidential race and the senate race. The idea here is to come up with a simple projection for the state house and state senate races, so I made assumptions as to where those numbers will fall compared to the top of the ticket races. Barksdale has remained virtually unknown against a very popular incumbent. I believe the numbers for that race are worse than what down-ballot Democrats will see. I also don’t believe that the down-ballot Democrats will match the numbers in the Clinton-Trump race. That race is affected by the unfavorability shown towards both candidates, but Trump is allowing the state to potentially flip. We’ve seen in the polls that the Trump drag hasn’t been as significant on down-ballot races and voters are willing to split their ticket.
All that being said, I believe the state house and state senate races will fall between the presidential numbers and the senate numbers. I think they will trend closer to the presidential numbers mostly because the Barksdale campaign has been so bad. So I’ve weighted the presidential numbers as twice the senate numbers. Here is my projected demographic voting share:
I’ve applied this voting share to the updated registration totals and a projected turnout between 2008 and 2012 levels. Obviously, there are many variables that could change, but I’m sticking with these for now. If you need a refresher on my methodology, then I’d suggest starting back at this post:
Updated Projections
House Districts
Senate Districts
As we’re now quickly approaching election day and many voters have utilized early voting, I decided to sort the districts by the projected current margin. As you can see, there is still a significant percentage of undecided/other polling responses. In this update I have not attempted to allocate the undecided/other numbers to the parties. I will, as promised, go through the districts included on my target list. At that time, I’ll allocate this undecided/other voting share to the candidates based on who is actually on the ballot. Regardless, this view gives us a general idea of the districts to watch on election night.
As I did in the past, I’m currently ignoring whether there are candidates on the ballot. This is a view if every race had a candidate from both parties on the ballot. With that view in mind, if the election ended today and the districts fell as they currently lean, then the house has the potential to split 71D/109R and the senate has the potential to split 20D/43R.
Again, those numbers are without accounting for incumbency, undecided/other voters, candidate strengths/weaknesses, fundraising, etc. I’ll take the time to look through those details before the election for each of the targeted districts to attempt a general projection as we approach election day. As this is the first attempt at this type of projection, I make no promises on my accuracy. I attempt to be upfront on my methodology and the assumptions I’ve made, so take that in mind when using these numbers to predict these races across the state.
Just a quick statement on the gerrymandering present in these races. I’ve discussed before how Georgia Democrats haven’t done themselves any favors by grouping themselves into one significant metro area, but these projections only shed more light on that issue. These projections only give the democrats 71 potential house seats and 20 potential senate seats. But these same projections give the Democrats 43.64% of the statewide vote, the Republicans 43.98%, and undecided/other 12.38% of the vote. This shows the massive ground Democrats have to make up. With essentially half of the vote, the outcome is still a significant disadvantage in the General Assembly. These numbers are why it’s important to take back both chambers by 2020 so the maps can be redrawn.
As we move on to election day, I’ll be working on a series of posts highlighting the individual districts I included on my targeted list. As a reminder here is the targeted list found in Part II of my previous analysis:
For each district on this list, I’ll take a look at the candidates, fundraising, and the demographics of the district. At the end, I will compile the numbers for a final look at the projections prior to election day.
Read the first of the competitive district breakdowns: The Uncontested
— 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.
