Georgia and American History
Abraham Lincoln was elected 16th President of the United States and the first Republican to hold the office on November 6, 1860. By his inauguration in March, seven states had seceded.
On November 6, 1861, one year after Lincoln’s election, Jefferson Davis and Alexander Stephens of Georgia were elected President and Vice President of the Confederate States of America.
President Teddy Roosevelt left for a 17-day trip to Panama on November 6, 1906 to inspect work on the Panama Canal; he was the first President to take an official tour outside the continental United States.
A dam on the campus of Toccoa Falls Bible College burst on November 6, 1977 under pressure from heavy rains, killing 39 students and faculty.
Speaker of the United States House of Representatives Newt Gingrich (R-GA) resigned his office and his Congressional seat on November 6, 1998, effective in January 1999, despite having been reelected three days earlier.
Georgia Politics, Campaigns, and Elections
The DunwoodyTalk blog has as good an explanation as I’ve read for how Holmes E. Pyles came in first in the DeKalb County Commission District 1 race.
A simple word was placed under each candidate’s name on the ballot. Four of the five had ‘Republican’ listed and one had ‘Independent’ listed. Only Holmes E. Pyles was not listed as a Republican.
The runoff of Jester and Pyles will take place next month and will be the only item on the ballot. The Dems won’t be back as the Republican vote is much bigger. Jester will not have three opponents competing for the Republican votes. We all know now that it is possible for a Democrat to make the runoff in District 1, but the chances of a Democrat winning the spot is low. Jester will need a strong turnout for the runoff.
The blog notes that even in the only Republican-majority district in DeKalb, Michelle Nunn took a majority in 20 of 37 precincts. [Disclaimer: I am a consultant for Nancy Jester’s campaign.]
Joel McElhannon, who served as the lead political consultant to the Georgia Republican Party’s Victory 2014 effort has penned “Seven takeaways from the 2014 Elections,” which we were happy to publish. It’s well worth reading the thoughts and conclusions one of Georgia’s top political minds who was instrumental to the best effort I’ve ever seen the GAGOP put forth. Here’s an excerpt:
1. Georgia Republicans Need A Competitiveness Assessment.
Last night was a huge win for Republicans nationally and in the state of Georgia. The GAGOP Victory Program, led by Chairman John Padgett and staffed by countless volunteers and sharp field directors, executed an unprecedented ground game in the Peach State. Over 350,000 doors knocked. Over 1.2 million volunteer phone calls – including 87,000 on Monday alone. Millions of pieces of mail dropped. It provided the rock solid foundation of success for our entire statewide ticket.
But Georgia Republicans should not be lured into complacency by this one night of success. We must also see clearly the political environment and the national wave the swept the country last night.
President Obama’s failed leadership is as popular as Ebola right now.
But he won’t be on the ballot again.
2. It’s Time For Georgia Republicans To Get Real.
Georgia is diversifying. In comparative demographic terms, Georgia is now the state of Virginia (metro Atlanta) dropped down in the middle of Alabama (the rest of our state). Our rural areas may continue to be part of the “old south” but the metro Atlanta region is a vibrant and diverse international community. Bluntly speaking, Georgia Republicans can no longer rely on simply appealing to white voters. We must diversify our approaches and speak to this new Georgia with a bold message about economic opportunity and effective governing.
3. Public Polling In Georgia This Cycle Was A National Embarrassment.
In the recent article “Are Bad Pollsters Copying Good Pollsters” on the highly respected Five Thirty Eight Blog, Harry Enten details how “polling” by non professional polling groups in states where a “Gold Standard” polling program does not exist are wildly inaccurate and tend to copy the results of legitimate pollsters as election day nears. In 2014, Georgia is the new case study for this assessment. As a highly respected political consultant friend told me recently, if these supposed pollsters for media outlets had been employed by campaigns and had been so wrong so frequently, they would have been laughed out of the business.
It’s worth reading in its entirety if you’re interested in the business and process of winning elections, even if I don’t agree with everything he writes.
Polling and Predictions
Speaking of polling, the Atlanta Journal-Constitution is living in a glass house when it criticizes the public polling in Georgia, but they continue anyway.
Some of those predicting runoffs didn’t take into account caveats, like margins of error and undecided voters, that swung the numbers.
Meanwhile, some earlier surveys were simply imprecise. They relied on automated calling and Internet surveys, cheaper methods scorned by more established pollsters.
“We have major polling problems (in Georgia),” said Kerwin Swint, chairman of the political science department at Kennesaw State University.
“No one here knows how to model turnout based on voting patterns, population, and issues.”
Survey after survey suggested that Republicans Gov. Nathan Deal and U.S. Senator-elect David Perdue might not surpass the 50 percent benchmarks needed to avoid long, costly and unpredictable runoffs.
Landmark Communications, based in Alpharetta, surveyed Georgia voters in the final days before the election and placed both Deal and Perdue with leads.
“We identified the Republican surge that took place in the closing days,” Landmark president Mark Rountree said.
“And in the end Georgia had the same surge for Republicans that the rest of the country saw, so the GOP candidates scored a few more percentage points than our, or anyone’s, poll reflected.”
The Atlanta Journal-Constitution commissioned New York-based Abt SRBI Inc. That survey, which used a mix of live calls to land lines and cellphones, took place Oct. 16-23. It showed the governor’s race in a dead heat and Perdue holding a slim lead in the Senate race. In it, the Libertarian candidates had 6 and 5 percent of the vote respectively. Ultimately that support was pegged at just 2 percent Tuesday night.
SRBI founder and chief research officer Mark Schulman said there were signs of a Republican wave in Georgia and elsewhere but the size of it “has befuddled the pollsters.”
Experts say the technique used by pollsters is significant. Live calling to homes and cell phones is considered the gold standard. Most of the public polls are done through automated calls to homes that under federal law cannot be made to cell phones. About 30 percent or more of registered voters only have cell phones so they are excluded.
First of all, to call the AJC’s polling the “gold standard” is laughable. Not only were they not any more accurate than most of the others, they were flat unable to poll the Republican and Democratic Primary elections earlier this year. In May, the Atlanta Journal-Constitution published a major “gold standard” poll that tested all the possible November head-to-head combinations for Governor and Senate, but then they wrote this,
The AJC did not poll the Republican or Democratic primary races because low turnout and primaries not confined to party registrants would have made the polling results, in its view, too unreliable.
Gold standard my tailfeathers. Not only were they unable or unwilling to poll the primary elections, they also didn’t poll the last twelve days – nearly two weeks – of the election. There are strengths to live agent phoning to random-digit phone numbers, but getting in-and-out of the field quickly is not one of them.
The question of whether to use IVR “robopolls” or the much-more expensive live-agent polling is best answered, “yes.” That is, use both. Live agent polling is often better in the early stages of the election for message testing when you’re using a long survey instrument, and as occasional benchmarks to fine-tune your sampling frame and the model that predicts the composition of the electorate. IVR is stronger when you need results fast and often. I often run IVR surveys every night the last two-to-three weeks of an election, with sample sizes of 1000-1500 every night. This allows you to be in the field every night affordably. The continuous nature of this style of tracking allows you to pick up trends earlier and more accurately track how voters are converting from undecided to decided. IVR is also very strong with a homogenous electorate, such as within a Republican Congressional District or a Metro Atlanta county Primary, less so in a more diverse electorate.
The difference between the two forms of polling is like the difference between a Ford that you can buy at the dealership and the cars that carry the blue oval in NASCAR races and on drag strips. They both carry the same name, but the difference in specific use, cost, and convenience will often determine which you use. If you’ve got millions of dollars and want to win a race more than anything else on earth, you buy a racecar. If you want to go to the grocery store and pickup the kids from school, you buy a Taurus. If you’re running a second-tier statewide race and don’t have millions of dollars, you might be able to run a live-agent poll one time – at the beginning or the end – but it won’t be of any use and you’d be better using Robopolling or spending the money on advertising.
Media polls are not designed to provide the level of information that campaigns rely on and no sane campaign strategist will pay attention to make his or her decisions on the basis of what public pollsters say. Media polls are designed to provide inexpensive fodder for “horse race” stories, and while the respective media outlets take their accuracy seriously, it’s simply not the same as strapping on a race car.
Non-professionals following public polls closely also may have unrealistic expectations when it comes to polls – polls taken weeks out do not by themselves predict the results of elections.
When I predicted last Friday that Nathan Deal and David Perdue would win without runoffs, it wasn’t simply because I checked the most recent polls. I looked at the RealClearPolitics average and saw that Deal was in the exact same position – 48.0% – that he was in 2010 when he walked away with a victory over Roy Barnes. I considered the strength of the GAGOP voter contact program that at the time had made more than 1.5 million direct voter contacts. I considered what appeared to be a trend nationally of Democratic candidates cratering and undecideds breaking for the Republican party. Finally, I applied “Kentucky windage,” or my estimate of which way the wind was blowing based on my own personal experiences.
Professional strategists running multimillion dollar campaigns will have all these tools, plus their own internal polling, probably that of their respective state party and national organizations, and other analytics, like the results of Voter Indentification calls.
Mark Rountree of Landmark Communications, who did the polling for WSB-TV this cycle, also responded to the AJC’s article, via Facebook:
An AJC article is out tonight saying that polls were off in Georgia. Actually, no, and this premise is not correct.
Here are the actual election results. Our poll results are on our website at LandmarkCommunications.net (#1 and #2 below copied from AJC article)
#1. GOVERNOR’S RACE POLLING:
Election Results: Nathan Deal-R 53 percent; Jason Carter-D 45 percent
Landmark Communications Poll: Deal 51 percent; Carter 45 percent
SRBI Inc-AJC: Deal 43 percent: Carter 42 percent
Survey USA: Deal 47 percent; Carter 42 percent
#2. U.S. SENATE RACE:
Election Results: David Perdue-R 53 percent; Michelle Nunn-D 45 percent
Landmark Communications Poll: Perdue 50 percent; Nunn 46 percent
SRBI Inc-AJC: Perdue 45 percent; Nunn 41 percent
Survey USA: Perdue 47 percent; Nunn 44 percent
• Landmark correctly nailed in Georgia the GOP surge that surprised many other pollsters across the country.
• Landmark nailed the Democratic candidates’ numbers essentially on the head (actual was 45% for both, we had them with 45% and 46% respectively).
• Landmark quite accurately nailed the Libertarian numbers (2% & 3%).
• Landmark also reported the GOP candidate numbers very close to the mark — it’s pretty hard to get much closer than what we released in our final poll.
• Landmark also had undecideds lower than anyone and ran with the call.
Remember also, I wrote earlier this week and again today, that the RealClearPolitics average showed Gov. Deal at 48.0 just before election day, the exact same as he was at that time in 2010. Deal won 52.8% Tuesday night and in 2010 he took 52.9% against Democrat Roy Barnes. Consistency of results and repeatability are also important criteria for judging polling, and the aggregate of public polling was both consistent and repeated its performance.
Remember also that a single poll shows a snapshot of a moment in time for an electorate in flux and under the influence of millions of dollars of advertising. You can’t make a good prediction from one poll – looking at polling holistically, not only did the public polls show consistently both Deal and Perdue ahead, they also showed both Republican candidates on upward trajectories as undecided converted in favor of the GOP. If you got the wrong answer from this year’s polling, you weren’t looking at the whole situation.
Exit Polls
Georgia was the subject of National Election Pool exit polling this year for the first time since 2008. We’ll be diving into both the exit polls and the Secretary of State’s data on voter turnout over the coming days, weeks, and months, but here are a few snapshots from the early analysis, here from the New York Times.

These graphics show that the Republican candidate for U.S. Senate, David Perdue, appears to have doubled the GOP’s share among African-American voters and increased it among both men and women. Interesting. I suspect Governor Deal carried more of the votes of African-Americans. We’ll see.