Behind Nate Silver’s war with The New York Times

Nate Silver made his name as The New York Times’ data guru, creating the methodology that predicted Barack Obama’s reelection.

Now, Silver’s method seems to be to stir up trouble for the Times.

The 41-year-old editor of the data-driven news site FiveThirtyEight recently called his former employer arrogant, engaged in an extended Twitter debate with his successor (who is also named Nate), and helped ignite outrage online over the paper's front-page headline on the president's response to two mass shootings, a wound that caused such pain it obliged Times publisher A.G. Sulzberger and executive editor Dean Baquet to hold an all-staff meeting this week to clear the air.

Silver’s persistent criticism of the Times, stretching back to the last presidential election, has long struck some inside the newsroom as less about methodology and more about personal grievances with the paper that was unable to meet his demands to expand FiveThirtyEight and now publishes The Upshot, which features Nate Cohn’s coverage of elections, polling and demographics.

In discussions with POLITICO, Times staffers questioned Silver’s motivation for repeatedly criticizing the paper, though they did not want to comment publicly, declining to pick a fight with Silver and his 3 million-plus Twitter followers. However, frustrations have spilled out publicly at times, with reporters suggesting on Twitter that Silver’s assessments of the paper’s journalism veered beyond respectful disagreement.

White House reporter Maggie Haberman once derided his “gratuitous jabs at a former employer,” while political and investigative reporter Nick Confessore characterized a tweet criticizing the paper’s 2016 coverage as “a cheap shot masquerading as something else.”

Silver, however, has not been deterred. In recent weeks, he has taken issue with reporter Peter Baker’s analysis of former special counsel Robert Mueller’s congressional testimony as “not the blockbuster Democrats had sought,” by grousing, “Why not just report the news and not reach for a weird like 5th-order conditional/counterfactual narrative.” He also poured some gasoline on the embers of a dispute over Times deputy Washington editor Jonathan Weisman’s racially divisive tweets. And he “sparked a social media furor,” as the Columbia Journalism Review noted, by being the first to call attention to the Times' now-infamous headline on President Donald Trump’s response to shootings in El Paso, Texas, and Dayton, Ohio.

“Not sure 'TRUMP URGES UNITY VS. RACISM' is how I would have framed the story,” Silver wrote in a tweet that was seized on by 2020 Democratic presidential candidates and others critical of the headline. Editors changed the headline, which Baquet told staffers Monday was “a fucking mess."

“I do think a lot of NYTs problems (and to be clear, there are *many* things they do very well along with some *serious* problems) are born out of arrogance; thinking they're the most important voice in the room,” Silver wrote after the headline blowup. “Nothing engenders that sort of arrogance like a lack of competition.”

When it comes to data analysis, at least, Silver offers competition of his own via FiveThirtyEight, a former Times site now under the auspices of ABC. This summer, a Silver-generated dispute over how the Times analyzes polling data has morphed into the “Battle of the Nates,” as the wonkish Twitter beef has been dubbed. In one recent skirmish, Silver told the younger Cohn that he hoped he’d “grow out of” a phase of calling things “simultaneously a prediction and not a prediction.”

Cohn and Silver declined requests for interviews. But in emails to POLITICO, Silver praised Cohn’s work, while dismissing suggestions his criticism of the Times is motivated by a rivalry or personal grudge.

“Media criticism is a core part of what I do,” Silver said. “I founded FiveThirtyEight 11 years ago because I had a lot of critiques of how the American press covers elections.

“Given the extremely prominent and influential place that the Times holds in coverage of American politics, any media critiques are naturally going to invoke the Times frequently,” he continued. “I think they'd do well to take those criticisms to heart instead of constantly accusing their critics of having ulterior motives or acting in bad faith.”

Nonetheless, a New York Times spokesperson told POLITICO: "We don't always agree with Nate but he sometimes offers smart criticism of our work and we value that, it makes us better.”

The Times’ history with Silver dates to 2010, when it took over FiveThirtyEight under a three-year agreement. His posts were a traffic-driver for the Times’ site and — like Cohn — he used polling data to raise provocative questions about elections, such as his November 2011 New York Times Magazine cover story asking, “Is Obama Toast?” Obama wasn’t, and Silver went on to correctly call all 50 states ahead of the 2012 election.

As his star rose in the media world, Silver, who was based in New York, developed a strained relationship with the Times’ political reporters in Washington. Former Times public editor Margaret Sullivan noted in 2013 how Silver didn’t “really fit into the Times culture” and Silver has accused the paper’s political reporters of being “incredibly hostile and incredibly unhelpful.”

Silver had fans in Times management, but the paper couldn’t meet his demands to stay.

“When Nate Silver was negotiating a new, richer contract, his lawyer told me that his client was ‘the prettiest girl at the prom,’” former Times executive editor Jill Abramson wrote in her 2018 book, “Merchants of Truth.” “I told him, perhaps echoing the past, ‘The Times is always the prettiest girl.’”

Abramson told POLITICO that Silver sought funding for about 20 “stat mavens in different areas, including sports and weather,” which “would have required a huge investment” from the paper. In July 2013, Silver decamped to ESPN, which bought the FiveThirtyEight site and provided Silver with resources to build a data-driven newsroom. ABC News — which, like ESPN, is owned by Disney — acquired FiveThirtyEight last year.

The Times rebounded after Silver’s departure by hiring Cohn, another young data wiz, in November 2013 and launching The Upshot five months later.

Both Silver and Cohn came under scrutiny during the 2016 presidential race for underestimating Donald Trump in the Republican primary. On Election Day, The Upshot put Hillary Clinton’s chances of winning the Electoral College at 85 percent, while FiveThirtyEight was less bullish than most election models, at 71 percent.

After the 2016 shocker, Silver published an 11-part opus: “The Real Story Of 2016: What reporters — and lots of data geeks, too — missed about the election, and what they’re still getting wrong.” And the Times, Silver wrote, “is a good place to look for where coverage went wrong.”

While the Times was Silver’s primary target, he suggested the paper’s political coverage was emblematic of broader problems in 2016, such as journalist groupthink, access-driven reporting, and a misplaced notion of Clinton’s inevitability. In the Trump era, Silver has accused the Times of normalizing neo-Nazism and being beholden to White House access.

On Tuesday, Silver suggested the Times is “too self-conscious about trying to prove” that it’s not part of the resistance to Trump that it will “likely worsen their journalism” and critics such as the president “will accuse them all of it all the same.”

And then there’s the running dispute with Cohn, who wrote on July 19 that Trump’s “Electoral College edge could grow in 2020.” Despite Trump’s low national polls, Cohn pointed to the importance of a potential “tipping-point state,” such as Wisconsin, where the president’s favorability is higher than average.

Silver argued that Cohn’s analysis of Trump’s Electoral College path is premature based on the polling currently available. “I'm pretty skeptical that we can say much right now about whether it will be larger, smaller or nonexistent in 2020,” he tweeted.

Cohn acknowledged the map can change before the 2020 election, but responded: “The piece is plainly describing the president's current standing, and goes to unusual lengths to explain that it can change.”

“I think it's kind of BS to lean really heavily into a particular takeaway in top 80 percent of the article, not to mention the headline/lead/social promotion/etc., and then to introduce the caveats in (literally!!) the 42nd paragraph,” Silver responded. The spat resumed nearly a week later, with Silver accusing Cohn of cherry-picking data to support his hypothesis about an election 16 months away.

Rachel Bitecofer, assistant director of the Wason Center for Public Policy at Christopher Newport University, who has analyzed the 2020 race using her own election model, told POLITICO that both Silver and Cohn had legitimate critiques.

But, she added, there seemed “to be a personal hostility that was coming through that was more than about methodology.” She said it looked like Silver was criticizing Cohn “for doing the same type of piece that Nate Silver has written many times.”

In an email to POLITICO, Silver praised Cohn for being “exceptionally good at what he does” and said that he doesn’t think “there are actually many philosophical differences” between them. He said “a lot of things that boil down to what you might call matters of taste,” such as “how certain things are described or when to apply a certain method to a certain problem.

“Practitioners often feel passionately about matters like these,” he added, “but I don't know that they're terribly newsworthy or important in the grand scheme of things.”

Yet Silver didn’t let the dispute go. Days later, he offered a mocking subtweet in response to new polls: “bUt I WaS ToLd ThAt TrUmP LoOkEd StRoNg In ThE TiPPiNg PoInT StAtEs aNd HaS a BiGgEr eLeCtoRaL CoLLeGe AdVaNtAgE tHaN iN 2o16.”

New York magazine’s Jonathan Chait, one of the many journalists following the recurring Twitter spat, responded: “At this point, the best advice I can give you, as a fan of both, is to settle this with a straight-out fistfight.”