Welcome!
2016-11-10 17:52:16
Mediator: News Outlets Wonder Where the Predictions Went Wrong

The country’s major news organizations, as surprised as anybody by Donald J. Trump’s ascension to the presidency, faced a question from their audiences on Wednesday that was laced with a sense of betrayal and anger: How did you get it so wrong?

The question came in letters. (“To editors and writers of The NYT,” one reader wrote, “you were so wrong for so long. You misled your readers and were blinded by your own journalistic bigotry.”) It came in Facebook posts. (“You were in a bubble and weren’t paying attention to your fellow Americans,” the filmmaker Michael Moore wrote in a post shared more than 100,000 times.) Most ominously, it came in the form of canceled subscriptions, something that will surely be monitored.

After projecting a relatively easy victory for Hillary Clinton with all the certainty of a calculus solution, news outlets like The New York Times, The Huffington Post and the major networks scrambled to provide candid answers.

With a new administration about to take shape in Washington, news executives tried to take stock of their mistakes and fix them on the fly, lest the fast-moving story leave them behind again.

Data journalism in particular, including websites like The Times’s Upshot and The Huffington Post’s Pollster, was under fire after guiding audiences — often through visually appealing speedometer-type graphics that forecast the probability of winning — to the conclusion that Mrs. Clinton would prevail in electoral votes. On Tuesday, the meters went haywire. The Times’s Upshot graphic, for instance, moved from a high probability favoring Mrs. Clinton to a huge advantage for Mr. Trump, with readers left to guess what the dancing needle really signified.

Though several national polls came close to capturing the final result in the popular vote — which Mrs. Clinton appears likely to win when the counting is done — all the number-crunching of state polls pointed to resounding success for Mrs. Clinton in the Electoral College.

The news media’s self-reflection on Wednesday brought to mind the awkward position Fox News found itself in four years ago, when it was criticized for creating an insular information bubble that led some viewers to believe Mitt Romney would defeat President Obama.

That raised questions about Fox News’s objectivity. Now it is mainstream news outlets, which Fox News so often portrays as liberally biased, that are facing a wave of skepticism. In some cases, the questions come from people at Fox News.

“A lot of media outlets made a decision sometime after the convention that Donald Trump was beyond the pale and they no longer had to observe the normal rules of journalism and objectivity,” Chris Wallace, a Fox News anchor, said in an interview on Tuesday. “I thought The New York Times was one of the worst offenders,” but, he added, “we were all guilty — myself included — of kind of writing him off.”

The weeks of soul-searching that are bound to follow were well underway on Wednesday.

“Whenever the news media gets surprised by a big story, there follows a round of questioning,” Dean Baquet, the executive editor of The Times, wrote in a note to his staff. “What could we have done better? How did we and other news organizations underestimate the support for such an unusual, even divisive candidate?”

Mr. Baquet praised his political team and other Times journalists for “agility and creativity,” citing articles about Mr. Trump’s taxes and Mrs. Clinton’s record in Libya.

But in an interview in his office, he said, “If I have a mea culpa for journalists and journalism, it’s that we’ve got to do a much better job of being on the road, out in the country, talking to different kinds of people than the people we talk to — especially if you happen to be a New York-based news organization — and remind ourselves that New York is not the real world.”

The NBC anchor Tom Brokaw echoed that sentiment, lamenting that for all its efforts at advancing diversity, the news media was still “pretty confined” to “the Eastern Seaboard.”

The sense of failure was not universal, with some news executives saying that journalists by and large did their job in presenting the news and leaving it to voters to decide.

“You can’t say that this campaign was undercovered or that this result is because of some failure to report on these candidates,” said David Rhodes, the president of CBS News. “I think it’s presumptuous of anybody — media or anybody else — to suggest that the reason for the result is some information failure.”

That said, there was a widespread feeling that the coastal and Beltway sensibilities of many mainstream reporters left them unable to relate to the sense of anger and resentment of the core Trump voter, which led them to miss Mr. Trump’s groundswell of support, some of which appeared to have been lost by polling, too.

And it left them blind to the fact that the political rules to which they adhered no longer seemed to apply.

Experience told them that Mr. Trump’s misstatements, flaws and gaffes would prove disqualifying, which at times led them to present their journalism with a knowingness that only served to convince a large subset of voters that reporters, at best, didn’t get them.

At NBC News on Wednesday, sleep-deprived executives discussed how to redouble efforts to capture issues close to voters, at a time when institutions like the news media are not trusted by many Americans.

“We cannot go in there with a sense that we think we know better,” said Phil Griffin, the president of MSNBC. “I think that’s often what happens in journalism: Some go in with a sense, from the experience they’ve had, that their reporting can anticipate things. We’ve got to go in with a total open mind.”

Journalists are also operating in a new media world — of Twitter, Facebook and the alt-right players who operate on the platforms — in which they are no longer the center of gravity.

“I think the larger point, and the more profound dilemma, is this new media landscape,” said David Remnick, the editor of The New Yorker.

Data journalism outlets like The Upshot have provided news organizations with compelling content offering a different presentation of journalism that shows readers the great breadth of data that can illuminate public opinion. It also has the benefit of inducing them to click early and often.

On The Times’s home page, a standing Upshot feature regularly told readers that Mrs. Clinton was the odds-on favorite to win the Electoral College — at times on the order of more than 90 percent — when there were stand-alone head-to-head national polls that showed a harder-fought race and, in some cases, even a lead for Mr. Trump.

Amanda Cox, the editor of The Upshot, said the model clearly indicated that Mrs. Clinton was the favorite based on polling data from several important states, including Wisconsin and Michigan.

Ms. Cox said the mistake lay elsewhere; there was a “failure of communication,” she said, in telling readers what the numbers actually meant — that probabilities are just that, probabilities, not guarantees.

The Times did try — for example, it translated the probability of Mrs. Clinton losing into the probability of missing certain field goals in football — but Ms. Cox said there was a way to make the meaning of the numbers clearer. “I think you could argue that there is room to improve how we communicate probability and uncertainty,” she said.

At Nate Silver’s data website, FiveThirtyEight, he had predicted that Mrs. Clinton would win with more than 300 electoral votes, and gave her a 71 percent chance of winning with the site’s polls-only prediction model. He also emphasized that Mr. Trump’s chances were real, with analysis pieces predicting that Trump could win the presidency while losing the popular vote.

“We explained over and over again how much uncertainty there was in this year’s race,” said David Firestone, the managing editor of FiveThirtyEight. “We were telling people up to the very end, ‘This is not a sure thing.’”

Mr. Firestone added that low turnout among Democrats was probably a result of voters’ apathy toward Mrs. Clinton’s candidacy, and not of certainty that she would win.

While he acknowledged that some of the modeling was out of step with reality, he disputed the notion that Mr. Trump’s surprise victory spelled doom for data-driven journalism.

“There’s a huge demand among readers for this stuff, and I don’t think that it’s going to go away because Donald Trump won by a few electoral votes,” he said.