How Terry McAuliffe mapped his Virginia win
By: Alexander Burns
November 6, 2013 06:28 PM EST
Terry McAuliffe’s gubernatorial campaign was never built for a landslide victory.
Not that the legendary Democratic operative-turned-candidate would have objected to a bigger win – but McAuliffe’s 2.5-point, 55,000-vote edge over Republican Ken Cuccinelli was the narrow margin his team planned for starting in early 2013, when McAuliffe’s advisers mapped out a strategy for winning in the difficult environment of an off-year Virginia election.
In detailed interviews with POLITICO the day after the election, members of McAuliffe’s team said that despite ample public polling that predicted a more leisurely victory and an Election Night perception that anger over Obamacare had narrowed the margin, it was always clear to them that the best-case outcome was a close win. And for McAuliffe, that always meant based capturing the ideological center and turning out a larger-than-usual share of the Democratic base. Those dual goals were no small challenge in a bipolar state that voted for a Republican governor by 18 points in 2009, a year after handing its Electoral College votes to President Barack Obama.
Recognizing from the start that his path to victory was slim, McAuliffe’s campaign invested early and heavily in establishing powerful tactical advantages over Cuccinelli, including sophisticated modeling of the Virginia electorate, experimentally-vetted messaging and a vast turnout operation that sent more than 13,000 volunteers into the field in the last four days of the election.
And even then, heading into Election Day, the campaign’s internal modeling predicted an electorate that voted for Mitt Romney over Obama by 2 percentage points.
It also showed, however, an electorate in which McAuliffe boosted African-American turnout on the margins and ran ahead of Obama’s 2012 performance with white voters in key areas such as Norfolk and Virginia Beach. Clashing with the perception that McAuliffe was hobbled by his profile as a political glad-hander, the modeling also showed him slightly outperforming a generic Democrat. All that was enough to overcome the Republican tilt of the voting population, if only by a little.
McAuliffe communications director Brennan Bilberry said that despite Democrats’ winning streak in the state’s federal elections, the fundamentals of the state leave only a “much, much narrower” path to take off-year gubernatorial races. The Democrat had to both disqualify Cuccinelli as an ideological firebrand and make sure his own party’s voters didn’t stay home.
“You have to do turnout and you have to win the middle,” Bilberry said. “[Democrats] have to do both.”
McAuliffe’s efforts to map a path to victory began early in 2013. There was little room for error for the former Democratic National Committee chairman: the last two Democratic governors of Virginia, Mark Warner and Tim Kaine, each won their elections with a modest 52 percent of the vote.
In February, McAuliffe took a large-sample poll of 10,000 Virginia voters. That huge tranche of data allowed the campaign to create a detailed model of the off-year electorate. While media attention has largely focused on whether McAuliffe would be able to turn out apathetic Democrats who participated in the 2012 presidential race, his campaign also had another objective in mind: identifying all the occasional voters who participated in the 2009 Virginia elections and ensuring that they showed up at the polls in a second off-cycle year.
That model – updated repeatedly over the course of 2013 – did more than just give McAuliffe a clear-eyed view of the state in which he was competing. The information included in that model, which also encompassed data on voter participation collected in previous campaigns, guided granular decisions about the McAuliffe ground game.
The campaign placed field offices based on where likely volunteers and likely get-out-the-vote targets were located. It used its modeling to create “heat maps” of the entire state, showing in florid color where the most intense field efforts should be concentrated.
Those field efforts – aimed at turning Virginia’s traditionally conservative-leaning off-year electorate into one McAuliffe could win – began in the spring, more than half a year before Election Day.
In April, McAuliffe volunteers were already knocking on doors and asking voters to sign postcards pledging to vote in November. They collected nearly 80,000 of them over the following months and mailed them back to voters in the middle of last week, in a tactic proven by prior Democratic campaigns to increase turnout among targeted voters.
By October, McAuliffe’s team had recruited just over 200 volunteer “team leaders” – supporters who committed more than 20 hours a week to the campaign – and fielded over 13,000 additional volunteers to get out the vote in the home stretch. It was a huge commitment of manpower in a race where voter participation on the margins would be the difference between success and failure.
All the while, the Democratic operation was mounting a large-scale persuasion effort aimed at dislodging wavering voters from Cuccinelli and encouraging apathetic, undecided voters to commit to McAuliffe.
Like the turnout machine, McAuliffe’s persuasion campaign was guided by experimentally-honed techniques: Much as his campaign built a turnout model that it could apply to households across Virginia, it also built a persuasion model. McAuliffe’s team tested a range of contrast messages in phone calls to targeted voters, and then polled the same set of households – plus a control group that received no contrast messaging – to test which voters were movable, and on what basis.
The information collected from those tests guided messaging across the Democrat’s campaign, including campaign mail, digital media and even some TV advertising, which McAuliffe placed on programs with relatively few viewers overall but high proportions of persuadable viewers.
Many of the conclusions from both the turnout and persuasion models were intuitive, said McAuliffe adviser Michael Halle, who managed the Virginia coordinated campaign. It was little surprise to anyone in McAuliffe’s camp that college-educated men and independent-leaning women were important persuasion targets, or that suburban and exurban counties such as Henrico and Chesterfield represented opportunities for McAuliffe to pick up ground even in a conservative electorate.
Nor was it shocking, McAuliffe aides said, to learn that both Democrats and persuadable swing voters responded positively to an upbeat message about working across the aisle in Richmond to strengthen education and job creation for the middle class. Or that young people reacted with horror to Cuccinelli’s views on gay rights, or that Hispanic voters recoiled from Cuccinelli’s past support for legislation allowing businesses to fire employees for speaking Spanish in the workplace.
Where the Democrats’ extensive modeling made a difference, Halle said, was in allowing them to “apply that research to the specific households” and laser-target voter contact methods and messages.
“Everything that we saw in our internal numbers is what happened last night,” said Halle, called the modeling was “as accurate if not more accurate than polling.”
The McAuliffe campaign offered up numbers from several targeted areas to prove the point. Prior to the election, the McAuliffe modeling predicted that without extensive campaign efforts, McAuliffe would win 53.6 percent in the Norfolk media market, 50.1 percent in the Richmond market and 55.3 percent in the Washington D.C. media market.
The model predicted that with a heavy-duty campaign operation, including field, mail and digital – McAuliffe would slightly increase his vote share to 55.1 percent in Norfolk, 52 percent in Richmond and 56.6 percent in Washington.
The final score closely mirrored those projections, as McAuliffe took 55.4 percent in Norfolk, 51.7 percent in Richmond and 56.7 percent in the D.C. market.
That’s only small movement in each of those areas, but even small movement is meaningful in a race decided by 55,000 votes.
McAuliffe digital consultant Andrew Bleeker said the precision-targeting of voters using a persuasion model – a tool that was too expensive for previous statewide campaigns – was critical in allowing Democrats to deliver their message even in GOP-leaning areas of Virginia.
“Four years ago, we would have said ‘Let’s just worry about Northern Virginia and not run a progressive message anywhere else in the state,’” Bleeker said. “But we have persuadable voters all over the state.”
He cited the campaign’s ability to communicate narrowly with young voters on “the crazy stuff Ken Cuccinelli said, particularly on LGBT issues,” with African Americans on voting rights and with Latino voters in ads featuring Kaine delivering a message in Spanish.
If national Democrats were moving Wednesday to draw sweeping conclusions from the Virginia race, calling it a repudiation of the tea party in a critical swing state, the McAuliffe camp objected more bluntly to the emerging national GOP argument that the race was so close due to public outrage over the Affordable Care Act.
Bilberry called that view “entirely inaccurate.” In an email, McAuliffe pollster Geoff Garin offered a deep eye-roll at the notion that McAuliffe would have won in a romp if not for the law known as “Obamacare.”
“With the exception of a bump up during the shutdown, Terry’s lead was always in the 2-4 range,” Garin said. “This was always bound to be a close race, given the structure of the electorate. Our analytics team, which conducted modeling calls on an ongoing basis, consistently modeled this as a 2-point race. The idea that it was close because of Obamacare is just plain wrong.”