Election 2014: What (in the world) happened in Virginia?

[Update: Republican candidate Ed Gillespie conceded the race to Sen. Mark Warner (D) on Friday.]

As I noted in a post in the early morning hours of Nov. 5, one of the biggest shocks of the 2014 midterms came in Virginia where Sen. Mark Warner (D), who had been leading in the polls with a very healthy margin all the way up until Election Day, found himself, at the end of the night, locked in a tougher than expected battle with former Republican National Committee Chairman Ed Gillespie, who was running his first actual campaign for office in a bid that nobody outside of his campaign seemed to believe had any real chance of succeeding or even coming close to making it a race against a man who was, by all measures, the most popular statewide political figure in the Commonwealth. All of that changed as the result rolled in on Tuesday night, though. While Gillespie did about as well as expected in the traditionally Republican parts of the state in the south, central, and southwest, the expectation was that the early lead he was showing wouldn’t last very long once results started pouring in from Northern Virginia and elsewhere, where Democrats have dominated these last several years. Warner did, indeed, eventually take the lead, but it was far narrower than anyone expected, and by the end of the night, the result was close enough that nobody was willing to declare a winner.

As things stand right now, the numbers from the Virginia State Board of Elections show Warner with 1,071,049 votes (49.12%) to Gillespie’s 1,054,50 votes (48.36%) for a difference between the two of 16,540 votes. As occurred last year in the race for governor, Libertarian Party nominee Robert Sarvis garnered more votes (52,984 votes) than the gap between the two major parties although his percentage of the vote (2.43%) was below where he ended up last year when he received 146,084 votes, a 6.54% share of the total votes in the 2013 gubernatorial election.  As to the race itself, it’s unlikely that Gillepsie will concede the race until the official vote certification process is completed next week, but given how that process has typically gone, it is likely that the gap between him and Warner will grow slightly, rather than shrink to any large degree. This is significant because, while Virginia law does allow for a recount if the difference between the top two candidates is less than 1%, reality dictates that it is highly unlikely that any recount would find enough uncounted votes, miscounted votes, counting errors, reporting errors, or misreported votes to make up a difference as large as the one between Warner and Gillespie. If anything, a recount would likely end up finding more votes for Warner and thus end up increasing the gap. This is what happened last year when the race for attorney general between Republican Mark Obenshain and Democrat Mark Herring started out with a 165 vote lead for Herring, one of the closest elections in the modern history of the Commonwealth of Virginia. By the time the recount was over, that gap had grown to 810 votes in Herring’s favor. This wasn’t unusual, since it has historically been the case in recounts that a re-canvas of the votes ends up following the trend that existed on Election Day in favoring the person who was in the lead at the end of the day. Most likely, that’s what would happen in the case of a Gillespie-Warner recount. This is likely the reason we are already seeing calls from many Virginia Republicans for Gillespie to let the certification go forward, but to decline to seek what would likely be a fruitless recount that is unlikely to change the actual outcome of the race. Instead, they are suggesting that he concede graciously, at that point and, if he wants to, move forward to 2017 when he would arguably be the frontrunner for the Republican nomination for governor.

Whatever Gillespie does, though, it’s clear that the results are something of a political shock. Most everyone, including me, saw Warner as among the safest Democratic senators running this year, and yet he has come withing a hairs breadth of losing his first race for reelection. The Virginia senator had been briefly courted to run for his old job as governor before the 2013 elections, and has been on the short list of potential vice-presidential running mates for the Democrats in 2016, the last item in no small part due to his popularity in a state that will be crucial to Electoral College victory for any candidate in 2016. Much of that veneer seems to be stripped off now, of course, and even if Warner wins, he will have done so in a chastening manner that will likely trim back much of the national enthusiasm for him going forward.

There are many theories about what happened in Virginia this year, but, at the very least, it’s clear that the polling was very, very wrong yet again, and the people who analyze these things for a living are starting to wonder why. Nate Silver offered some of his own theories yesterday, and pointed to turnout as part of the key:

Turnout in Virginia this year is projected to 37 percent of the voting-eligible population – in line with the figure nationally in a year where turnout is down across the board. But this is low in comparison to other states to have held competitive Senate races, where turnout averaged about 44 percent instead. Virginia’s 2014 turnout was also well below its figure in 2006 (44 percent), when the state hosted a competitive Senate race between James Webb and incumbent George Allen. It’s even lower than 2010 (39 percent) when it had neither a Senate nor gubernatorial race.

(…)

So … turnout it explains it all? It’s a big part of the story. As The Washington Post reports, the turnout drop-off was higher in Democratic-leaning counties. And pollsters may have had trouble modeling the Senate race in Virginia: There was no precedent from 2010, and they would have overestimated it if turnout had been pegged to 2006.

But a few things are still hard to reconcile. If turnout were the whole story, you’d expect substantial differences between the demographics in exit polls and pre-election surveys. Instead, they closely match one another.

The Christopher Newport University poll projected black turnout at 16 percent, for example, and the Roanoke College poll had it at 19 percent — compared to 19 percent in the exit poll.¹  Partisan and age demographics were also similar.

And as I mentioned, Warner’s favorability rating was 56 percent even among voters on the exit poll. Candidates with those sorts of numbers don’t lose – or come so close to doing so – very often.

So, what explains the result? I don’t know, but I have a theory – one that ties the race together with the ones in New York and Michigan. It’s that some voters who would otherwise be inclined to vote for Warner went for Gillespie because they didn’t think Warner needed their vote.

That Warner and Bloomberg were quite popular among voters in the exit polls contributes to the theory. Some voters may have been perfectly willing to vote for them if they thought they’d needed to. But because they didn’t see the races as competitive, they felt free to use their vote in other ways. They might have voted for Gillespie as an expression with their overall dissatisfaction of the direction of the country, for instance, or Thompson as a protest against Bloomberg’s change to term-limit laws.

There may be a danger zone when a candidate leads by about 10 percentage points in the polls, as Warner and Blanchard did. That’s a big enough lead that the news media may not portray the race as competitive – Blanchard’s re-election was thought to be quite inevitable, for example, as was Bloomberg’s. Another example is Barack Obama’s lead of about 8 percentage points over Hillary Clinton in polls of the New Hampshire Democratic presidential primary in 2008. The media seemed to regard another win for Obama as a sure thing after his victory in the Iowa caucuses.

But a 10-point lead isn’t completely safe. If the polls were a bit wrong to begin with, and turnout is a bit lower than expected, and a few voters change their mind in the ballot booth, the leading candidate can lose.

Nate Cohn is even more adamant that turnout, specifically low turnout among Democrats, was the reason Warner nearly lost and, in any case, has ended up with a result that has significantly blunted his political career:

Some weak Democratic performances in the midterm elections Tuesday are hard to explain. Virginia is not.

Senator Mark Warner, the Democrat who unexpectedly finds himself aheadby just 16,000 votes in a race he was expected to dominate, received the weakest turnout of any Democratic candidate in a competitive Senate race.

Just 2.2 million votes were counted in Virginia (with some absentee and provisional ballots still remaining), or about 57 percent of the 3.9 million votes cast in 2012. No other competitive Senate race had turnout beneath 64 percent of 2012 levels.

The largest declines in turnout were in Democratic-leaning areas. Turnout in Republican-leaning areas in Virginia was at levels fairly typical for a competitive midterm election.

There was barely a race in Virginia. Mr. Warner was thought to have a significant edge, and the state was not a top priority for Democratic turnout efforts. The Republican, Ed Gillespie, didn’t even have the money to air television advertisements throughout October.

Cohn provides this chart showing just how far off Virginia turnout was compared to other states based on preliminary data:

So does this mean turnout explains it all?

Well, as Silver says you can’t pin the explanation solely on one factor, but we know for certain that the outcome of an election can turn significantly on who comes out to vote. If it really was the case that voters inclined to vote for Senator Warner stayed home because they figured he would win, then that would go a long way toward explaining how it came to be that a candidate who was leading by double digits has ended up within less than a percentage point of an opponent who was a virtual unknown when the campaign began. If that’s what happened, then the blame for what could have been an historic and embarrassing loss lies squarely at the feet of the Warner campaign and a Democratic Party that didn’t have a good enough Get Out The Vote effort to make sure that the people who supported them came out to vote. After all, half the job of a campaign is to make sure the people you know support you actually show up to vote, and if that doesn’t happen, then it isn’t going to matter how flawlessly the rest of your campaign may have been run.

It strikes me, though, that turnout may only be part of the explanation for what happened in Virginia. As I noted, this is the second election cycle in a row where a race that was supposed to have been in the bag for a Democratic candidate ended up being a nail biter. Like Governor McAuliffe, Senator Warner seems likely to win his race in the end, but it will be by a much closer margin than anticipated, and much closer than the polls predicted. That suggests to me that many of the assumptions that have been made about politics in Virginia in the wake of President Obama’s wins here in 2008 and 2012 may have been overstated. It’s true that the President did something in those two campaigns that no Democrat since John F. Kennedy in 1960 and Lyndon Johnson in 1964 had been able to do when he won the Commonwealth twice in a presidential election. However, that may mean less than some political observers have assumed. Virginia is still a purple state, of course, and the influence of voters in areas like Northern Virginia are likely to benefit Democrats well into the future. However, the assumption that Virginia is a purple state turning blue that many seem to have made appears to be incorrect. Instead, Virginia is likely to be much more competitive going forward than Obama’s two victories may have suggested. Bob McDonnell’s win in 2009 showed that Republicans could still win statewidec notwithstanding Obama’s historic win the year before. In 2010, Republican victories at the congressional level showed that to an even greater extent, and the GOP’s continued success at the state legislative level reinforces that idea. Now, we have two elections in a row where Republicans did far better than polling had said that they would and, indeed, there is plenty of reason to believe that Republicans would have had a much better 2013 if they’d chosen better candidates for governor and lieutenant governor. This tells me that the state will be more competitive in 2016 than some are assuming, and that those who may have declared the Republican Party of Virginia dead and buried will need to reassess their conclusions pretty quickly.

Doug Mataconis appears on the Outside the Beltway blog at http://www.outsidethebeltway.com/.