What it means to get the election wrong

Yesterday morning, the New York Times asked, in a headline, “Can Biden still win?” The story made clear that he could, but as FiveThirtyEight’s Nate Silver pointed out, the question seemed backward, since President Trump appeared to be facing the thinner path to victory. That the Times framed the question around Biden, Silver wrote, felt rather like an “artifact of the mood right now”—Biden had not, as many liberals had hoped or expected, crushed Trump across the map, leading to a feeling, in many quarters, that Trump was on the front foot.

Astead W. Herndon, a politics reporter at the Times, framed the situation better in a tweet: “If you’re a Democrat whose goal was America repudiates Donald Trump and his politics I understand disappointment,” he wrote, but “if you’re a Democrat whose goal is Joe Biden becomes the next president, it’s not that bleak.” As yesterday progressed and more votes were counted, the key states of Wisconsin and Michigan were called for Biden, who now looks to be in a much stronger electoral position than Trump. It should be noted here that the perception of one candidate gaining on, or pulling away from, another in a dynamic way is a function not of any actual voting, but of vote counting, the timing of which is political. It’s likely that Wisconsin, Michigan, and Pennsylvania, which remains too close to call, could have declared their results earlier if Republican officials in those states hadn’t stood in the way; now, of course, the Republican president is baselessly wielding the delayed count in these states as evidence of fraud. News anchors haven’t noted this discrepancy enough.

Related: What the polls show, and the press missed, again

Still, we don’t yet know how the final uncounted votes will break, nor the tricks Trump might try to subvert the outcome if he loses. And whatever happens now, Herndon’s first point stands: “America” clearly has not repudiated either Trump or Trumpism. Since Biden’s path to victory has started to look clearer, the national conversation has moved on from pure suspense to also consider the ramifications of Trump’s better-than-expected showing. Many liberals are shocked and horrified that millions more Americans voted for Trump the president in 2020 than voted for Trump the long-shot outsider in 2016. And many members of the media are wondering—again—how they failed to see this coming. “It’s 2016 all over again,” Kyle Pope, CJR’s editor and publisher, concluded yesterday. “Major media institutions made it all but impossible to envision that, despite the wealth of reporting on the president’s lies and his racism and his circus, nearly half the country remains beholden to the man and his beliefs.” We understand Trump, but not his electorate.

One immediate subject of recrimination was the polls, many of which, Pope and many others noted, got the election badly wrong, up and down the ballot. On The Daily, Alex Burns, of the Times, said that reporters should have listened more carefully to what the Trump and Biden campaigns were saying about their internal numbers. Other observers went further. The Atlantic’s David A. Graham wrote that the polling industry and media number crunchers now face “serious existential questions”; the authors of Politico’s Playbook newsletter called for the industry to be “blown up.” As Pope also notes, some newsrooms seem to have used polling as a substitute for the more time-consuming, expensive work of sending reporters out to properly observe the country. And local news organizations—a crucial check on skewed numbers and myopic national narrative-making—are increasingly hollowed out, where they still exist at all.

The election is a sharp reminder: we need to talk about polls way less, however right or wrong they prove to be. Major outlets still organize campaign coverage around minute, repetitive parsing of the horse race, leaning way too hard on polls and probability forecasts, and imputing to them a predictive value that they don’t claim to offer. We need, too, to urgently revitalize local news and improve newsroom diversity, basic steps that will help elevate perspectives that are routinely missing from our work.

Sign up for CJR's daily email

It also strikes me as important that we recognize some caveats here. Neither less polling nor more reporting is a guaranteed corrective: as Laura Wagner, of Vice, put it, “data journalists who make models and reporters who talk to people can both be good and bad!” It’s way too soon to know exactly how and why the polls were wrong, and the coming postmortem will be instructive as to whether the industry can be salvaged, even if salvaging it won’t alleviate the media’s basic need to kick its polling addiction. It’ll take time, too, to work out precisely what we missed in the national mood. In the meantime, we shouldn’t jump to clichéd conclusions or lose perspective. Biden has already won more votes than any presidential candidate ever, including in a number of states where Democrats have recently struggled, and the arcane mechanics of the Electoral College shouldn’t obscure that fact. Many more votes were suppressed. And for all that Republicans might try to paint themselves as the party of the working class, exit polls suggest that Trump won among voters with family incomes higher than $100,000, whereas Biden won among those who earn less.

We should be careful, in our introspection, to distinguish between what it means to be factually wrong and what it means to be morally wrong. We clearly missed an ongoing groundswell of pro-Trump sentiment. There’s no question, though, that a significant proportion of that sentiment is noxious—a grim mix of reality-denying magical thinking and hate that Trump has stoked, day in and day out, first on the campaign trail and then in office. Trump and his allies in right-wing media—itself an entrenched power structure backed by big corporate money—have poisoned millions of Americans not just against the mainstream press as it currently exists, with all its flaws and blind spots, but against the basic principles undergirding what journalism is. If anything, learning that this worldview holds more sway than we thought makes it more urgent that we tell the truth; as I’ve chronicled in this newsletter, the most consequential recent failures of the reality-based media have involved not holding Trump to account well enough. Going forward, we shouldn’t confuse necessary humility with ambivalence about the values of the free press.

Yesterday, my CJR colleague Lauren Harris tweeted an apt observation that Masha Gessen, of The New Yorker, shared in a recent interview with Anand Giridharadas, of The Ink. “Everything is not nuanced,” Gessen said. “Everything is not subtle. Some things are clear; some things are morally abhorrent and have no justification, and then other things are things that call for empathy. Some things are morally abhorrent but require strategic empathy because we still need to understand where they came from. That’s where stories come in.”

Below, more on the election:


Other notable stories:

  • Yesterday, as we all watched the election unfold, the US reported more than a hundred thousand confirmed cases of covid-19—the first time any country has passed that daily milestone. More than fifty thousand people are now hospitalized with covid, the highest figure in three months. “As winter nears, the country’s third surge of infection is dangerously accelerating in almost every region of the country,” Robinson Meyer and Alexis C. Madrigal, of The Atlantic’s covid Tracking Project, write. “The next president will take power in a country where 100,000 cases forms a new baseline.… The United States is not a healthy country.”
  • For the Times, Amelia Nierenberg reports that, with covid cases surging on college campuses, we’re relying on student journalists to hold their administrators and classmates accountable. Even prior to the pandemic, college papers played “an increasingly vital part in their communities,” Nierenberg writes. “The crisis in local journalism, which has forced more than 1,800 US newspapers to close or merge since 2004, has left some of them as the sole remaining daily paper in college towns.”
  • Gannett says that it now has more than one million paying digital-only subscribers. The company reported third-quarter results showing a net loss of $31 million and a 20 percent year-over-year decline in revenue; the results were better than those posted in the second quarter, but Gannett’s ad revenue is way down, and the company has made sharp cuts during the pandemic. USA Today’s Nathan Bomey has more.
  • For Newslaundry, an independent news site in India, Anmol Somanchi and Supriti David analyzed ad spending by the Indian government since Prime Minister Narendra Modi took power in 2014. Modi’s government has spent much more than its predecessor on ads, and has placed more of them on TV than in print media. Generally, governments can use ads “as a stick to beat journalists into compliance,” Somanchi and David write.
  • And, following a series of terrorist attacks in France, Emmanuel Macron, the country’s president, wrote to the Financial Times pushing back on an opinion piece in the paper that he felt accused him “of stigmatising French Muslims for electoral purposes and of fostering a climate of fear and suspicion towards them.” The article, which was published online on Monday, has been taken down for review after readers pointed out errors.

ICYMI: Surprise! Election night was a confusing mess.

Has America ever needed a media watchdog more than now? Help us by joining CJR today.

Jon Allsop is a freelance journalist. He writes CJR’s newsletter The Media Today. Find him on Twitter @Jon_Allsop.