by Sean Spicer
Katie called me.
“Sean, this is huge. We need you to get back now.”
“Katie, if I get off this plane, I’m not getting to St. Louis. This is the last available flight before Sunday.”
She didn’t like it, but I stayed on the plane, stewing the whole way to St. Louis, unable to talk to anyone or do anything. And, because of a lack of Wi-Fi, I was unaware of how it was playing out. The instant the plane hit the runway in St. Louis, I turned on my phone and saw that I had missed dozens of calls. For the rest of the day I was on the phone, nonstop, till sleep overtook me.
At first, we thought we were dealing with a dump of opposition research by the Democrats, but as events unfolded, it appeared that the Access Hollywood recording had been leaked by someone from NBC to the Washington Post. TMZ reported the following:
NBC execs had a plan to time the release of the Donald Trump audio to have maximum impact on both the 2nd presidential debate and the general election . . . top network execs knew about the video long before they publicly said they did, but wanted to hold it because it was too early in the election. The sources say many NBC execs have open disdain for Trump and their plan was to roll out the tape 48 hours before the debate so it would dominate the news cycle leading up to the face-off.1
Every campaign has a pivotal crisis that marks either the final decline of the candidate or the moment when the candidate transcends his or her own shortcomings. For example, Bill Clinton won the presidency when he and Hillary Clinton went on 60 Minutes to explain away the taped conversation of shared, intimate recollections between the then governor and his mistress, Gennifer Flowers. Bill Clinton admitted to CBS’s Steve Kroft that he had caused “pain in my marriage.”
Everyone at the RNC and in the Trump campaign knew that this video was our pivotal crisis. Trump had three options at that point. One was to curl up and die. The second was to do an apology tour, like the Clintons. The third was to barrel through it and come out with all guns blazing. At this moment, the media couldn’t comprehend how Donald Trump could even dare to show his face in St. Louis, much less present a spirit of proud defiance. Most political operatives would opt for the apology route.
Didn’t Donald Trump know that it was all over?
But Trump read the national mood better than his critics. On Saturday, as I took calls from Republican committee members, donors, reporters, pundits, and friends, I was surprised by how many women who contacted me did not consider Trump’s comments a big deal. One prominent Republican woman told me, “You all talk like this; we know it.” (Actually, we men don’t all talk like this, but I held my tongue.) Clearly there were a lot of detractors, but many people I spoke with didn’t think Trump’s vulgar remarks from eleven years ago were relevant or important.
As he often does, Trump moved forward when any other candidate would have been left for dead.
Trump succeeded by dismissing his remarks as “locker-room” talk—which many people apparently thought they were—and by pivoting against Hillary in a way that no other Republican candidate would have dared: why should I be crucified for locker room talk when you protected your husband after he faced a credible allegation of rape?
Steve Bannon supported the candidate’s response by bringing a surprise to the debate. He invited Juanita Broaddrick, Kathleen Willey, Kathy Shelton, and Paula Jones—women who had accused Bill Clinton of rape, groping, and sexual harassment—as the campaign’s guests. Steve tried to keep their arrival secret (though, inevitably, rumors started leaking) lest someone in the campaign or the debate organization overrule him.
Then Steve did something no one saw coming. Prior to the start of the debate, he gathered the women and held a press conference in which they appeared with Donald Trump.
If anyone was unnerved by this turn of events after the release of the Access Hollywood tape, it wasn’t Trump. It was the Clintons. Hillary had to debate with these women staring her down from the audience. With this panel of women, Donald Trump had not only leveled the playing field, he had put Hillary on the defensive. Many people had to ask themselves: if what those women said is true, what kind of a woman would stand by a man who did such things? In the debate, Trump did not act chagrined or pull his punches. He did not even shake hands with Hillary. He was aggressive and pointed in his criticism. Hillary later said that Trump had made “her skin crawl” during the debate. The pivotal moment had arrived . . . and the pivot turned in Trump’s direction.
The media didn’t see it. David A. Graham, writing in The Atlantic under the headline “Donald Trump’s Disastrous Debate” declared, “With Republicans abandoning Trump in droves just 29 days before the election, Trump seemed content to drive all of them off . . .”2 While the Access Hollywood video had taken a temporary toll, the debate had provided an even bigger boost. Trump’s underdog campaign was in fact regaining its momentum.
In the third and final debate, in Las Vegas, Donald Trump employed another nontraditional tactic. The format required one candidate to stay seated while the other stood and took the floor. Donald Trump refused to sit still. He paced around while Hillary spoke, always staying in the frame behind her and towering over her. It was domineering and a little strange, but it worked. Everything she said was under the scowling face of Donald J. Trump. It was not something I would have thought of or advised. But Trump showed, once again, that he is a master at emotional imagery. The political playbook for debates had been thrown out the window and burned.
Throughout the primary and general debates, Donald Trump was like an inflatable ball in a swimming pool. Hillary Clinton and the media tried hard to keep him submerged, but he always popped back up to the surface. Nothing could sink him. And that fact caused quite a few media heads to explode.
“Relax,” The Nation assured its left-wing readers in a choice bit of graveyard whistling, “Donald Trump Can’t Win.”
“Trump won’t win,” a writer in the UK’s The Guardian declared, “In fact, the US could be on the brink of a liberal renaissance.”
Even some of our side’s own political gurus were not projecting a win. Karl Rove, no stranger to presidential politics, said on Fox News, “I don’t see it happening . . . If he plays an inside straight, he could get it, but I doubt he’s going to be able to play it.”3
In making such predictions, the media and even some Republican loyalists didn’t see the power of Trump’s populism and message; they also misread “outside the Capital Beltway” voters. The smugness and rhetoric of the naysayers created an echo chamber that helped amplify the over-confidence of the Clinton campaign, the worst case of choosing-the-Oval-Office-drapes-before-the-election since Thomas Dewey. Clinton headquarters in Brooklyn assumed the election was in the bag.
But then came two more October surprises.
On October 28, 2016, three of us from the RNC—Sean Cairncross, Katie Walsh, and I—were asked to attend a briefing held by the Department of Homeland Security. We met with senior officials in a conference room at their headquarters in Washington, D.C., near American University. They informed us that they were aware of and monitoring Russia involvement in our upcoming elections. The message they wanted us to convey publicly, however, was that there was no way to infiltrate or manipulate the outcome of a national election because we have a disparate voting system spread across thousands of counties, cities, towns, and precincts. Seeming to believe that Clinton would win, they implored us to publicly express confidence in the integrity of the voting process, system, and outcome.
On the same day as this briefing, FBI Director James Comey sent a new letter to Capitol Hill reopening the Clinton email scandal by revealing that more classified material had possibly been exported to an improper source, this time by Clinton aide Huma Abedin onto a computer owned by her husband and former Democrat Congressman Anthony Weiner.
Even though two days before the election Comey popped up yet again, bizarrely, to announce, “Based on our review, we have not changed our conclusions that we expressed in
July,”4 most Americans felt they had reason to doubt Hillary Clinton’s honesty. If Trump was frank to a fault, openly imprecise in his language, and brashly indulgent of “truthful hyperbole” on his own behalf, many found Hillary to be skirting the truth in ways that were much more serious. For example, the Clinton staff’s efforts to hide her pneumonia, revealed by her fainting episode after a 9/11 commemoration in New York, had underscored the predicate that she lacked candor. Had Hillary Clinton gone to the hospital right away, it would have been better for her. At every turn, Hillary and her overly protective staff acted as if they had something to hide. Every email and Clinton Foundation scandal, every too-clever-by-half statement by the Clintons and their team, came together to define her in that moment as the soul of duplicity.
In politics, there comes an instant when people have had enough of someone. It happened to Lyndon B. Johnson. It happened to Richard Nixon. This was Hillary Clinton’s moment.
Of course, Donald Trump had huge negatives of his own. He largely got a pass because he spoke about a reality that the Clintons were not talking about—the economy—even though it was the Clinton administration years earlier that had coined the phrase, “It’s the economy, stupid.” And Trump wasn’t just talking about corporate earnings and investments, he was talking about jobs and wages in a way that resonated with disenfranchised, unemployed Americans, all while Hillary and the Democrats played identity politics. Whatever you believe about transgendered bathrooms, they don’t put meals on the table for an unemployed welder in Michigan. Bill Clinton had begun his career with a powerful connection to working Americans. Now Hillary was calling these Americans “deplorables.”
Her last campaign stop was in Raleigh, North Carolina, where her campaign had said she would win. The event featured a high-energy, get-out-the-vote speech by Lady Gaga. That was Hillary Clinton’s idea of populism.
Lady Gaga would play really well in many states, but she was the wrong messenger in the wrong state at the wrong time.
Hillary Clinton would go on to lose North Carolina and its fifteen electoral votes by more than 3 percent.
It was in the wee hours on Wednesday, November 9, 2016, that Donald Trump was declared the winner of the presidential election, with 306 electoral votes.
Since then, Hillary has thought a lot about why she lost—and how it could have happened. It was Comey’s press conference that did it. It was Anthony Weiner. It was voter suppression or Russian interference. It was anyone and everyone . . . except the candidate.
It’s easy to be a Monday morning quarterback in presidential elections, but in the end, the buck stops with the candidate. It’s a matter of having the right message, the right campaign operation, and the right judgment. Hillary had failed on all three counts.
At the RNC, we had learned our lesson during the Romney campaign about the importance of using voter data effectively. We began what would ultimately be a $175 million investment on a system of high-tech, integrated data as part of our Growth and Opportunity Project.
Chris Carr, our political director at the time, had made an especially brilliant move to turn our data into actionable knowledge. After carefully studying what the Obama campaign had done in 2008 and 2012, Chris spearheaded a major transformation in how the Republican Party would track and get out voters. He created a system that assigned a sentiment score (using a scale of one to one hundred) on key issues for every voter in the country. Our data operation did this by vacuuming up data from magazine and digital subscriptions, voting history, membership rolls of groups ranging from the NRA to Planned Parenthood, donations, social media actions, and combining them all to give us a reliable profile of every voter.
We compiled these consumer data points on all 190 million American voters: Republicans, Democrats, and Independents across all fifty states, in each congressional district and in every single voting precinct.
We knew from our data how likely someone is to vote, whether that person is likely to vote early or absentee, and his or her views on education, gun control, national defense, and many other areas. Is he registered with one party but independent-minded? Is she willing to vote across party lines?
This strategy allowed us to follow and communicate with specific voters on a weekly basis. The regularity of our communications gave us the ability to follow and analyze trends and identify messages that persuaded undecided voters to become committed voters.
The media’s pessimism about the prospects of the Trump campaign bled over into stories claiming that the RNC’s internal models indicated Trump was destined to lose. Our models showed no such thing. In fact, our models didn’t project a win for either candidate. But our models did highlight what we needed to do to win the election.
Winning a campaign is like building a house. If your house is going to cost $100,000 to build, and you have only $90,000 in the bank, you wouldn’t throw up your hands and quit. You figure out how to scrape together another $10,000 or how to cut corners. You know what you need to acquire or change.
In a similar way, our predictive modeling, which was being spearheaded by long-time GOP strategist Bill Skelly, told us where we were short and which voters we needed to win over to win.
After a campaign targets its valued, likely voters, it must map where they are concentrated. Thanks to Chris Carr, we had a map that was broken up into “turfs”—areas with high concentrations of gettable voters. The Trump-rich “turfs” were where we put a pin in the map and placed a field office. Chris had derived these strategies from concepts that the Obama campaign had used effectively in 2008 and 2012, but he modernized them to benefit our entire party.
In 2004, the Bush campaign made two revolutionary changes to political campaigns. First, they created a seventy-two-hour task force that reached voters in the final three days before an election. The goal was to motivate potential voters to get out and cast their vote. The task force effectively activated a system that micro-targeted voters. Simply put, micro-targeting means you look at the traits at a small group of voters and apply that to a larger group. Obama’s 2008 and 2012 campaigns took that to a new level: they recognized that people were voting earlier and earlier, and they utilized data to target voters on a personal level. The issue for the Democrats was that the Obama campaign had built a great model for that particular candidate, and their data was not necessarily relevant for another candidate like Hillary. Frankly, it’s simple marketing.
The RNC had learned from the Obama campaign, but we were going to apply those principles to a national party and benefit our presidential campaign as well as candidates up and down the ballot.
On election day, we needed to maximize that data with a well-organized effort composed of volunteers and paid staff who, in large part, had been working in communities we had targeted for months. They had been making calls to persuade “our” voters to vote before election day, during their lunch hour or before or after going to work. More and more voters are casting ballots earlier and earlier in the election season—not waiting for election day—sometimes as early as September. Our goal was to encourage as many people as possible to vote before election day, and we followed that with a strong second push on election day to get any remaining voters to go vote. Katie Walsh had worked with Chris and Bill to oversee that effort. Here, as in so many cases, the RNC did essential tasks that most national campaigns did for themselves. I don’t say that as a criticism of the Trump campaign but to highlight the fact that those who say there was a fundamental rift between the RNC and Trump are wrong; the RNC’s efforts were perfectly integrated with what the Trump campaign needed to do, providing the organization, data, and ground game.
While Bill Skelly and Chris Carr are two names most Americans will never know, I am convinced that beyond Donald Trump himself, they are the two people most responsible for the victory of the forty-fifth president of the United States.
Here’s how they did it.
The RNC data revealed many blue-collar workers in Pennsylvania, Ohio, Michigan,
and Wisconsin were likely Trump voters. They had a history of voting Republican. We just had to bring them home. We knew how many votes we needed in key states to make the difference, so we focused our resources on where we could pick up enough outstanding voters to push us over the top.
It was a tall order, and it was getting taller. Since February, our voter scores showed that a group of likely GOP voters had moved about 15 percent away from the Republican ticket.
Chris and Bill met nightly with Steve Bannon to plan our response to this deterioration. The group expanded to include Marc Short from Mike Pence’s team; Brad Parscale, who led the campaign’s digital effort; and others from the campaign like Jared Kushner, Kellyanne, Donald Trump Jr., and Eric Trump. We felt, in our guts, that the people who were slipping away were still our voters. Chris and Bill greenlit focus groups that allowed us to drive every decision about where to go, what to say, and what ads to run. What we found were voters who would never cast a ballot for Hillary Clinton. They were not what we called “a persuasion group.” They were already persuaded. They were what we called a “turnout group.” They would never vote for Clinton, but they were uncomfortable voting for Trump, even as solid Republicans. If we could get them into a booth or get them face to face with a ballot, where they had a strict, binary choice, Carr and Skelly bet they would vote for Trump.
For that reason, Chris and Bill had taken to calling these voters “downshifters.” After the Access Hollywood video, our model showed Trump dropping to 180 electoral votes, almost overnight. But even at the lowest point, we saw that our comprehensive portrait of individual voters and predictive analysis could guide Brad Parscale and his digital team, and Chris and his staff in each battleground state, to target the right ad to the right voter at the right time, millions of times.