Nate Silver Defends FiveThirtyEight’s Missed Call on Donald Trump Nomination

'I'm glad we didn't predict something as crazy as a Trump nomination,' data journalist says

Nate Silver Variety Entertainment and Technology
Joe Russo/Variety/REX/Shutterstock

Nate Silver, founder and head stat-master of ESPN’s data-driven FiveThirtyEight, explained how the site failed to predict Donald Trump would win the Republican Party presidential nomination — saying it resulted from a reliance on using history as a guide.

“I’m glad we didn’t predict something as crazy as a Trump nomination,” said Silver, speaking at the Variety Entertainment & Technology Summit in New York Thursday. He labeled the candidate a “demagogue” who has risen to power by fostering cultural resentment against Muslims, Mexicans and sometimes women and black people, as he previously wrote in a May 4 post admitting FiveThirtyEight got it wrong on Trump.

There’s been some schadenfreude over FiveThirtyEight’s missed calls this election cycle. Famously, Silver correctly called the results of the 2008 presidential election in 49 of 50 states — and all 50 states in 2012 — by crunching reams of data. Silver dismissed critics who have singled out FiveThirtyEight for incorrect predictions of late: “The consumer understands our brand. I don’t care if media columnists understand our brand.”

FiveThirtyEight, according to Silver, had always said Trump could win the nomination “but we gave it a low probability.” The site had the odds of his primary win at 50:1, then 5:1 before he became the odds-on favorite recently. “I don’t think it’s any kind of existential crisis,” he said. “We live in an uncertain world.”

One of Silver’s main takeaways from the real-estate mogul/reality TV star’s ascendancy is that the conventional wisdom that parties generally get to anoint a candidate was wrong. The Republican Party’s nomination process was a “clusterf—,” he said. (Or, he added, one could call it a “Dumpster fire.”)

“When you have something this crazy happen in American politics, it’s worth reflecting on our assumptions about politics that are no longer true,” said Silver. Trump has deviated from the Republican line on numerous issues, and “we thought based on history that candidates like Trump that go against their party platform are not very electable historically.”

At this point, Silver puts the odds of Hillary Clinton beating Trump in the general election at 75%. But he’s not ruling out a Trump victory. “Sometimes I hear commentary, ‘It’s impossible for Trump to win.’ That’s B.S.,” he said.

Silver argues that FiveThirtyEight’s brand of journalism provides more transparency than traditional news organizations, because the site provides odds. (He noted that FiveThirtyEight was the first to predict that the NBA’s Golden State Warriors would win a record-breaking 73 games this season.) “We’re more honest. We give people a number” for the probability of an event happening, he said.

In the past year, FiveThirtyEight’s audience has more than tripled, growing from 3.6 million unique visitors in March 2015 to 11.5 million this past March. Clearly, that’s been fueled by interest in the “nutty” election season, Silver conceded, but he pointed out the site covers more than just politics, with stories spanning sports, culture, social justice and entertainment.

ESPN hired Silver after it acquired FiveThirtyEight.com from the New York Times in 2013. In the first few months, “the traffic frankly was not very good” on the site, he said, “but ESPN made a long-term commitment, and as a result we’ve been able to improve the growth.” FiveThirtyEight now has about 30 journalists on staff, and has embarked on several video projects, including a series about whether Howard Dean’s 2004 election-rally scream was what actually killed his chances (the data showed his campaign was already on the decline).

Silver, who was interviewed by Variety Co-Editor-in-Chief Andrew Wallenstein, said FiveThirtyEight is not a huge fan of highly distributed publishing a la BuzzFeed and said the site can succeed without scaling to billions of views. “We’re emphatically medium-size,” Silver said. “Being the second-best BuzzFeed, when your content is not as fun or interesting as BuzzFeed? That’s not where I’d want to be.”

Pictured above: Variety Co-Editor-in-Chief Andrew Wallenstein (left) and FiveThirtyEight’s Nate Silver