Skip to main content
ABC News
Why The Georgia Special Election Matters

The “takes” you’ll read about the special election in Georgia’s 6th Congressional District on Tuesday night are probably going to be dumb. A close outcome in either direction would be consistent with what we know about the political environment. So if either Democrat Jon Ossoff or Republican Karen Handel wins narrowly, it will be portrayed as a more important predictive signal than it really is. A blowout result would be a bigger deal. But even then, Georgia 6 is a slightly unusual district, and the election would be one data point among many. Georgia isn’t even the only special election on Tuesday; South Carolina’s 5th Congressional District is holding one also.

Here’s the thing, though: Sometimes dumb things matter if everyone agrees that they matter. Congressional Republicans could use a signal of any kind right now to coordinate their strategy around two vexing issues: first, their health care bill, and second, their behavior toward President Trump and the investigations surrounding him. Whatever direction Republicans take on these questions, they will find some degree of strength in numbers. Republicans would probably be less afraid of publicly rebuking Trump, for instance — and becoming the subject of a @realDonaldTrump tweetstorm or Trump-backed primary challenge — if other GOPers were doing the same.

The Georgia 6 outcome might trigger some herd behavior among Republicans, therefore, changing the political environment in the weeks and months ahead. A loss for Handel would probably be interpreted by the GOP as a sign that the status quo wasn’t working. If even a few members of Congress began taking the exit ramp on Trump and the American Health Care Act, a number of others might follow. A win, conversely, would have a morale-boosting effect; Republicans would probably tell themselves that they could preserve their congressional majorities by turning out their base, even if some swing voters had abandoned them.

Georgia 6 is a tough district to diagnose because its politics in presidential elections shifted a lot from 2012 to 2016. In 2012, the district went for Mitt Romney by 23 percentage points in an election that then-President Barack Obama won by 4 points nationally. That made it 27 points more Republican than the country as a whole. In 2016, by contrast, it chose Trump over Hillary Clinton by only 1.5 points in an election where Clinton won the popular vote by about 2 points nationally. Therefore, it was only 3 to 4 points more Republican than the national average.

If one uses the 2016 presidential election as a benchmark, this is a race that Democrats should be winning. They currently lead in the generic congressional ballot by about 7 percentage points, which ought to be enough for Ossoff to overcome the district’s modest GOP lean from 2016. By contrast, if one takes 2012 as the benchmark in the district, then even coming within single digits of Handel would represent a massive overperformance for Ossoff.

FiveThirtyEight’s usual procedure — what has produced the best predictions in past elections — is to combine the past two presidential cycles in a 3-1 ratio. That is to say, we’d take three parts from 2016, when the district was only slightly red-leaning, and one part from 2012, when it overwhelmingly backed Romney. By that formula, Georgia 6 is 9.5 points more Republican than the country as a whole. According to the generic ballot, the current political environment isn’t quite Democratic enough to overcome a 9.5-point deficit, so you’d expect Handel to win, although narrowly. (The table here describes a number of methods for projecting the Georgia 6 result, which I’ll explain further below. The one I just mentioned — taking a blend of recent presidential results and adjusting it based on the generic ballot — is the third in the table.)

2016 presidential result + generic ballot Ossoff +3.3
2012 presidential result + generic ballot Handel +20.3
Blend* of past presidential results + generic ballot Handel +2.6
2016 congressional result** + generic ballot Handel +7.8
2016 presidential result + special election results so far Ossoff +11.9
2012 presidential result + special election results so far Handel +11.7
Blend of past presidential results + special election results so far Ossoff +6.0
2016 congressional result + special election results so far Ossoff +0.8
Aggregate party margin in first round Handel +2.1
Aggregate party margin in first round + shift in generic ballot since first round Ossoff +1.1
Aggregate party margin in first round + top-two margin in first round Ossoff +4.8
Aggregate party margin in first round + top-two margin in first round + blend of past presidential results Handel +0.4
Aggregate party margin in first round + top-two margin in first round + blend of past presidential results + generic ballot Ossoff +3.3
Average of recent polls*** Ossoff +2.4
It’s hard to know how Georgia 6 “should” go

* The blend of past presidential results uses a 3:1 ratio of results from the 2016 and 2012 presidential elections
** The 2016 congressional result is adjusted for former Rep. Tom Price’s incumbency advantage
*** Recent polls include those from Trafalgar Group, Opinion Savvy, SurveyUSA, Landmark Communications and the Atlanta Journal-Constitution

Given that this is an election to Congress, one could also look at past congressional results for guidance. In 2016, the Republican incumbent Tom Price, who vacated the seat to become Trump’s secretary of Health and Human Services, won the district by 23 percentage points. This was somewhat less than his 32-point margin in 2014 and his 29-point margin in 2012, but it nevertheless suggested that the district was still quite red. Republican representatives such as Price had a significant incumbency advantage in 2016, as incumbents almost always do — even in supposedly anti-establishment environments. But even if you adjust for that advantage, the Republican lean of the district is about 15 points, still too much for Ossoff to overcome given the generic ballot. In fact, coming within several points of Handel would count as a good result for him given Price’s performance.

One could argue, however, that the standard for Democratic success should not be based on the generic ballot, but instead their results in special elections so far this year. By our method of benchmarking special elections, these have been very good for Democrats and are consistent with a political climate in which Democrats would win the national House vote by about 15 points.

That makes for a big difference. With a 7-point win in the House popular vote, as the generic ballot shows, Democrats would be only about even money to take over the House next year, given the way the vote is distributed among districts. With a 15-point win, they’d be all but assured of doing so — in fact, they’d probably have a massive wave election on the scale of 1994 or 2010, with lots of Republicans in supposedly safe districts being caught in the undertow. If Democrats are to keep pace with their special election results so far, then Ossoff probably should be winning the race, not just coming close — and Georgia 6 should be the election where Democrats go from “moral victories” to actual wins.

Another set of benchmarking methods involves extrapolating from the first-round vote in Georgia on April 18. In that election, multiple candidates from both parties were on the ballot. Ossoff got by far the most votes with 48 percent, with Handel finishing in a distant second at 20 percent to claim the other runoff spot. However, the Republican candidates combined got about 51 percent of the vote, compared to 49 percent for Ossoff and other Democrats together. Methods that factor in these results generally suggest a close race, but perhaps with Ossoff as a slight favorite. One of these methods, for instance, takes into account that while Republicans narrowly won the aggregate vote on April 18, the political climate has become somewhat more Democratic since then, perhaps just enough for Ossoff to win by a point or two.

Finally, there’s the polling, which shows Ossoff ahead by a not-very-safe margin of about 2 percentage points. You’d rather be 2 points ahead than 2 points behind, however.

So there are a lot of rather different, but nevertheless entirely reasonable, ways to interpret what might constitute a good or bad result for the parties on Tuesday. If either Handel or Ossoff wins by more than about 5 percentage points — which is entirely possible given the historic (in)accuracy of special election polls — you can dispense with some of the subtlety in interpreting the results, especially if the South Carolina outcome tells a similar story.1 Otherwise, Tuesday’s results probably ought to be interpreted with a fair amount of caution — and they probably won’t be.

As I said, however, the vote comes at a critical time for Republicans — and extracting any signal at all from Georgia might be enough to influence their behavior. Republicans really are in a pickle on health care. The AHCA is so unpopular that they’d have been better off politically letting it die back in March, at least in my view. But I don’t have a vote in Congress and Republicans do, and they’ve tallied the costs and benefits differently, given that the bill has already passed the House and is very much alive in the Senate. The central political argument Republicans have advanced on behalf of the bill is that failing to pass it would constitute a broken promise to repeal Obamacare, demotivating the GOP base. That argument will lose credibility if a Democrat wins in a traditionally Republican district despite what looks as though it will be high turnout.

There are also some tentative signs of congressional Republicans breaking with Trump, at least when it comes to matters related to Russia. Two weeks ago, former FBI Director James Comey testified before the Senate Intelligence Committee. Directors of other intelligence agencies and Attorney General Jeff Sessions have also recently testified before the committee. Then last week, the Senate overwhelmingly (by a 97-2 margin) approved a package of sanctions against Russia despite the White House’s objections to them. Those are serious, tangible steps for a Congress that had often moved in lockstep with Trump until the last few weeks.

As is the case with health care, Republicans don’t have an obviously correct strategy for handling Trump. Pulling the rug out from under him could create a vicious cycle in which Trump’s approval rating continues to decline and even some fairly partisan Republican voters begin to disapprove of his performance. But polls suggest that Trump is already quite unpopular with swing voters (and that his base has already shrunk), which might suggest that it’s smart for Republicans in competitive seats to rebuke Trump before matters get even worse.

In either case, the narrative that emerges from the Georgia 6 runoff will lack nuance and will oversimplify complex evidence. While special elections overall are a reasonably useful indicator in forecasting upcoming midterms, their power comes in numbers. A half-dozen special elections taken together are a useful sign; any one of them is less so. But we’re at a moment when Republicans have a lot of decisions to make now, and the story they tell themselves about the political environment matters as much as the reality of it. The narrative will probably be dumb, but it might matter all the same.


  1. By our formula, South Carolina’s 5th Congressional District is 19 percentage points more Republican than the country overall, so we’d suggest that Republicans should be worried if Democrat Archie Parnell comes within single digits of Republican Ralph Norman.

Nate Silver is the founder and editor in chief of FiveThirtyEight.