“Frankly, We Did Win This Election”: The Inside Story of How Trump Lost

by Michael C. Bender

  “I can’t tell you what his involvement was with Bridgegate—I can’t vouch for him,” Christie said.

  Corey relayed Christie’s warning and, two days after the Iowa caucuses, Trump shelved Jared’s plan.

  “Let’s not take the risk,” Trump told Jared during a campaign stop in Arkansas.

  Trump demolished all his competitors in New Hampshire, including Christie, who finished a distant seventh place, dropped out of the race, and, two weeks later, endorsed Trump.

  It wasn’t for six more months—after Corey was fired in June—that Jared found the space to bring Stepien into Trump World as the campaign’s national field director.

  Stepien was put in charge of state directors for the 2016 Trump campaign. While Brad had merged fairly seamlessly with the Republican National Committee’s digital team, Stepien clashed with his counterparts at the RNC from an adjacent office in Trump Tower. Stepien complained to colleagues that Brad was ceding too much to the party. Officials from the regional political teams for both the Trump campaign and the RNC complained they were working at cross-purposes and on the verge of a civil war.

  Three weeks out from Election Day, RNC chief of staff Katie Walsh and Chris Carr, the RNC’s political director, approached Stepien about setting up a joint call to show a united front to the troops. Stepien refused to commit, and Walsh escalated the issue to Bannon, the campaign’s chief executive.

  “I don’t know what the fuck this guy’s problem is,” Walsh told Bannon.

  Bannon, like Brad, viewed the RNC as a crucial tool, from fundraising to political infrastructure to, most important, its data operation.

  “This is ridiculous,” Bannon told Walsh, and ordered Stepien to fall in line.

  But Stepien continued to keep Walsh and Carr at arm’s length. Emails went unanswered. He refused to schedule a meeting and complained that his team was already overburdened with conference calls.

  “Maybe we can get through this week, and I’ll introduce the idea this weekend to them,” Stepien told Walsh.

  Walsh unleashed a flurry of expletives that caught the attention of others in the office. “I have multiple email exchanges with Steve telling you to do joint calls!” she shouted. “I’ve printed them out, and they’re getting leaked the day after the election unless you get on the fucking phone.”

  The state director call was held the next day.

  After his promotion to campaign manager in the summer of 2020, Stepien moved into Brad’s office. Stepien’s elevation made a certain amount of sense. He had been deputy campaign manager. Trump was familiar with him. And he and Jared were close.10 But it was never clear what problem he solved for the president’s reelection bid.

  It had been obvious for months to many in Trump World that the reelection team’s vulnerabilities were disparate messaging and de minimis strategizing. The messaging issue always came with high risk and high reward in Trump World. And yield on that bet had been terrible for most of the summer. In the week Stepien was hired, Trump surprised his team by threatening to ban TikTok, the popular Chinese video app; equated Portland, Oregon, with Afghanistan as he complained about bursts of violence at Black Lives Matter protests; and boasted about passing a cognitive test—a simple exam meant to detect dementia—in an attempt to attack Biden.11

  The issue of strategy, on its face, seemed solvable. Even Brad, who had the closest view of the campaign’s strengths and weaknesses, had begged Jared to bring in an experienced strategist who would give him air cover to finish building out the campaign. The decision to install Stepien was its own acknowledgment that such a person no longer existed in Trump World—or wouldn’t take the job.

  Stepien was a veteran political operative, which was no small thing in Trump World. The personnel departments in the White House and campaign often seemed to more closely resemble a public bus stop: Just standing in the right spot at the right time was mostly all it took to catch a ride. Stepien had paid his dues running and winning campaigns. But Christie’s 2013 campaign had cost $11.4 million, just a bit more than double what Trump had spent on a single Super Bowl commercial earlier in 2020. And Stepien had never been the big-picture strategist with the sweeping vision that guided tactical decisions. That had always been Mike DuHaime’s role. Even for Christie’s two races, Stepien had been campaign manager but DuHaime was the top strategist, consulting from the outside.

  Stepien and DuHaime had barely spoken since Bridgegate, but the night Trump offered the job to Stepien, he phoned DuHaime and asked his old friend for advice. The skate guard and the Zamboni driver from Chimney Rock Ice Rink rehashed some of the good old days, what worked and what didn’t. They talked strategy and tactics. They compared notes on motivating staff and maximizing the talent on the team. The two men continued the conversation over the remainder of the campaign.

  The thing Stepien wanted most was structure and logic. He had an Election Day countdown clock prominently displayed in campaign headquarters. He scheduled weekly senior staff meetings and daily check-ins with a tighter circle of deputies. He drew up a new organizational chart, and immediately implemented a strict chain of command around himself, telling his team that he wouldn’t speak to more than ten people per day. Otherwise, he would lose focus and become less efficient. He joked that he suffered from obsessive compulsive disorder, but there was a kernel of truth in his self-deprecation. Stepien kept his desk highly organized. He wrote everything down, and he tracked all the details. His own checkbook was perfectly balanced. It had made him anxious to watch staffers flow in and out of Brad’s office all day long.

  Stepien’s first move was to elevate Justin Clark, his comrade for the past four years in Trump World, to deputy campaign manager. A forty-five-year-old attorney, Clark was a father of four and grew up in a middle-class home in West Hartford, Connecticut. He was as affable as Stepien was curmudgeonly. Clark drank beer, ate barbecue, and played linebacker on the same high school football team as Brett McGurk (defensive line), who had held national security positions in three consecutive administrations.

  Clark was undoubtedly the only member of the Trump campaign who was also a veteran of Al Gore’s 2000 presidential bid. Clark had helped with accounting issues on the former Democratic vice president’s campaign before converting to the lonely life of Connecticut Republicanism and working on a series of unsuccessful statewide campaigns, including the failed 2012 U.S. Senate bid from future Trump Cabinet member Linda McMahon. He’d worked closely with Stepien since the two met on the 2016 campaign, including two years together in the White House.

  Stepien also promoted Nick Trainer, his thirty-year-old protégé with a bald head and bushy red beard, as well as a keen ability to keep many of his colleagues at the campaign irritated. The biggest reason for the griping was his unenviable position as Stepien’s enforcer.

  Trainer denied requests for more staff or more money. He clapped back at state directors who talked to the president without clearing it first through headquarters. But he also often executed with an air of self-importance that drove some staffers crazy. Some colleagues privately referred to him solely by the awkward title of battleground strategy director as if it was his name: “Mr. Director of Battleground Strategy.”

  The traditional strategist job under Stepien mostly fell to Jason Miller, just as it had for roughly two months before Brad’s demotion. Stepien and Miller had both worked for Trump in 2016, and they knew each other from Giuliani’s presidential bid. Miller, a rotund forty-five-year-old with a black goatee, was a methodical communications operative whose career seemed perpetually on the brink due to a proclivity for misbehavior and hedonism.

  Trump had expected Miller, who was married, to join him in the White House as communications director in 2017 until his affair with a campaign adviser spilled into public view—and remained there for much of the next four years. A fellow Trump campaign adviser, A. J. Delgado, became pregnant with Miller’s son while Miller’s wife was also pregnant with their second daughter. Miller reconci
led with his wife, but Delgado repeatedly sued him over failure to pay child support and accused him of trying to secretly drug her with an abortion pill slipped into a smoothie. When Miller sued for $100 million claiming Delgado’s accusations had ruined his reputation, he ended up acknowledging in a deposition for the case that he’d been a frequent customer of massage parlors in New York, lap dances in Tampa, and escorts in Washington. The judge ruled against him. The feud between Miller and Delgado then turned to issues of child support, again playing out in public.Trump constantly needled Miller about it and complained to others that he just needed to pay Delgado and make the problem go away.12

  Miller declined the White House job, landed a $500,000-per-year job as a managing director of Teneo, but then appeared to lose the gig with the global strategic communications firm when he fired off a series of obscene tweets at House Judiciary chairman Jerrold Nadler, a New York Democrat. In the aftermath, Miller announced that he and the company had agreed to sign a formal “separation agreement and general release.” But behind closed doors, Miller and Teneo quietly signed a new deal for the same amount of money that he’d be paid as a consultant, according to the Guardian. By the summer of 2020, Miller had firmly reestablished himself inside the campaign and proved to be something of a Trump whisperer. He awoke before dawn, organized talking points, prepared a political briefing, and delivered both to the president every morning. By July, Miller was largely credited both inside the White House and at the campaign for Trump’s good days. He parlayed that goodwill into a top post at the convention, which he oversaw from the production trailer parked outside the convention center, and a more influential role overseeing the content of the campaign’s television advertising.

  Within forty-eight hours of Stepien’s promotion to campaign manager, he sought to get firmer control of the budget. He canceled more than $10 million for billboards and to fly banner ads over beaches on Labor Day. Stepien’s team found at least one no-show contract worth $15,000 per month and terminated it. He reduced staffing at events, limited the number of aides who could fly on Air Force One, and banned the use of charter flights to fly the team to rallies. To better understand the campaign’s TV advertising, which was one of the most significant costs, Stepien scheduled a conference call on his second day on the job with Jason Miller and the political advertising team, which included the campaign’s pollster, RNC staff, an outside ad buying agency, a data analytics firm, and staff from American Made Media Consultants, which was the in-house ad-buying company the campaign had created to avoid vendor commissions and conceal spending details.

  “Okay,” Miller said to start the call. “How are we targeting TV?”

  The seemingly simple question triggered a series of assumptions, arguments, and sidebar conversations that exposed layers of tensions inside the campaign. Staff for American Made Media Consultants, which had been set up by Brad, Lara, and others, were suspicious of why their former boss had been demoted. They closed ranks quickly and made it virtually impossible for Miller to see inside the operation. Miller had previous political ad-buying experience, but that had predated the availability of some of the analytical tools Brad had been using, and much of the ad team had dismissed Miller as a dinosaur.

  Miller’s question about TV targeting was one any new department head would have asked. But he also viewed analytics as hocus-pocus. The campaign had spent tens of millions of dollars on law-and-order campaign ads that summer, and Miller had been told that the spots tested well. But the spots had failed to improve Trump’s poll numbers. When Miller’s team looked into the testing, they determined that the sample sizes were much too small. In some cases, Miller told Stepien that advertising decisions had been based on responses of fewer than ten undecided voters out of sample sizes of around 200 respondents.

  The ad testing was being done by Deep Root Analytics, a political data company the RNC had brought in to test ads and monitor the effectiveness of political messages across television markets and online audiences. Brent Seaborn, a Republican operative and one of the company’s founders, defended the work product on the call. He told the group that the RNC had used Fabrizio’s polling data to build a universe of voter targets, which he then had used to test the spots. The campaign’s pollster reacted like he’d been backed into a corner. Fabrizio questioned the wisdom of intertwining the two data sets, saying his polling data was based on survey questions that differed from what the RNC asked in their surveys.

  The call left the campaign with more questions than answers. Miller was concerned about the results coming out of the analytics company, but Stepien was also suspicious of the RNC data being fed into the algorithms in the first place.

  “I think this data is shit,” Stepien told Miller after the call.

  “Then what the fuck are we paying attention to it for?” Miller said.

  Fabrizio suggested a secret poll of the RNC’s universe of persuadable voters—the group of Americans the party had determined were open to supporting Trump. Stepien agreed. The next week, the results showed just 15 percent of respondents said they were undecided. It seemed to confirm Stepien’s hunch.

  “Their data is shit,” Fabrizio told Stepien.

  An alarmed Stepien called another meeting. The goal was to assess any breakdowns occurring in Brad’s data operation. It quickly became clear to Stepien that he’d inherited a mess. But Brad caught wind that his advertising plan was under pressure and started to fight back. He complained to Jared and Trump that the data program was being undermined over internal grudges, and misunderstandings about how it had been built. The struggle over data reflected the vastly different personalities, management styles, and political approaches between the two men. From a certain angle they had complimentary skills, and, to their credit, they had discussed a deputy campaign manager role for Bill earlier in 2019. The conversation didn’t go well, and Bill declined.

  The bigger problem was that Trump had asked Brad to build a political campaign based on marketing, advertising, and branding, and then hired Bill to run it like an accounting firm. The problem wasn’t as much that one was wrong and one was right. It was that they had been forced on each other, and neither liked it.

  To help him straighten out the data program, Stepien brought in old pal Bill Skelly, a fellow New Jersey native whom he’d met during the 2004 cycle when both worked at the RNC. Skelly would be teamed up with Matt Oczkowski, who Brad fought to bring into the campaign. Before he was demoted, Brad had been on the verge of hiring Oczkowski to oversee the data program and start predictive modeling of Election Day scenarios. Even Brad’s allies shook their heads in disbelief that the campaign hadn’t put those critical pieces in place months earlier.

  Yet Oczkowski was controversial, too. He had worked with the Trump campaign in 2016 when he was head of product at Cambridge Analytica, a data company that shut down in 2018 after allegations about its misuse of Facebook data. Trump campaign officials denied they used Cambridge’s Facebook data in 2016, or the psychographic targeting the company had said it could perform based on personality types. Instead, Brad said he relied on Cambridge to help find persuadable voters and track polling. But the company’s modeling had been off in a big way—it had predicted an almost certain defeat for Trump on Election Day.

  Fabrizio’s view of Oczkowski’s operation was similar to Miller’s perception of Deep Root—a slightly more sophisticated form of witchcraft. For Fabrizio, Oczkowski might as well have been trying to dunk voter data in water to see what floated. Oczkowski, in return, didn’t really believe in Fabrizio’s polling.

  It also became clear to Stepien that Trump World had competing plans for how to collect 270 electoral votes, which effectively meant that there was no plan. The RNC had its path to 270, which differed from Fabrizio’s plan, which conflicted with Oczkowski’s. Fabrizio’s polling showed Trump had issues that needed to be fixed with white independent males. But that assessment confused the analytics team—with no party registration in places like Michigan, th
at meant nearly every male in Michigan was a white independent male. Fabrizio’s universe included more than two million people. Oczkowski’s universe of persuadable voters was closer to 100,000 people. He’d also identified white independent males—just not all of them. Meanwhile, the RNC was focused on a group of Republicans they referred to as “disengagers”—conservatives who had moved away from Trump.

  The competing opinions meant that the RNC thought the campaign was responsible for persuading disengagers, while the campaign thought that the RNC was turning them out. Instead, no one from Trump World was consistently talking to one of the most important subsets of the electorate—a major gap in the voter universes of which almost no one had taken ownership.

  While Brad had stood up various independent pieces of the campaign, he hadn’t been able to pull them all together. Ideally, the data modeling team and pollsters would work hand-in-glove to build a universe of persuadable voters to target, test the messages that those voters would find most persuasive, and continue to refine the targets and sharpen the pitch. But Fabrizio and Oczkowski looked at the same data and drew different conclusions. Fabrizio thought he’d proven his point when he tested Oczkowski’s universe of persuadable voters and found that—similar to the RNC’s universe—nearly 80 percent said they were definitely going to vote for either Trump or Biden. But Oczkowski viewed the results as validation of what they all knew already—Trump was losing, and the campaign needed to persuade voters who were planning to vote for Biden to switch to Trump.