Let’s put aside for a moment who is right and who was wrong. Certainly there is plenty to be said in defense of the Nates: They are working with the data they are given, and if we just calmed down and didn’t get so emotional about everything all the time, we might get a grip on the difference between a probability and a promise.
“I’m a control freak, so I always need to know all of the information that exists so I can feel like I am in control of a situation,” says Jon Favreau, the former Obama speechwriter and current liberal podcaster.
But does having an insatiable poll habit actually make him feel more or less in control?
So maybe it’s not them. Maybe it’s us, and our addiction to election needles and red-and-blue maps of counties that we only just learned about, like, two Magic Wall segments ago. Maybe we’re too scrambled by hope to see a favorable forecast as anything but a foregone conclusion — a Nate accompli if you will — and too prone to disappointment to see an unlikely occurrence as anything but a system failure.
Either way, it’s not clear the Nates are good for anyone’s health.
“To a large extent, it’s our own fault,” says Terry Sullivan, a Republican operative. “But they are the street dealer of the drug — polling — that so many people find themselves addicted to.”
Neither Nate immediately responded for a request for comment. Though to be fair, they’re both pretty busy at the moment.
Let’s back up a million Magic Wall segments ago to Tuesday, when followers of both Nates went into the evening secretly feeling like prophets: They knew Democrats had a great shot at winning the Senate. They knew, per FiveThirtyEight, that Joe Biden had an 89 percent chance of winning the White House. They knew, as the ballot counts rolled in in real time, the odds in not one but three key swing states at any given moment, thanks to the “election needles” on the Times’s website.
“This is way smarter than the 2016 needle,” Cohn crowed on Twitter. (Never mind that thousands of people nearly overdosed on a similar needle in 2016.)
“People get led astray on a night like last night by vacillating space needles,” says Jesse Ferguson, a former Hillary Clinton spokesman.
Ferguson was there at the Javits Center in New York as the original needle dived right into a dramatic Donald Trump upset. He says he didn’t once look at FiveThirtyEight or the Times’s Upshot page this year.
Not that he judges people for feeling the need to do so. It’s only human.
“We pretend we know what’s going to happen because often the fear of the unknown is emotionally worse than deciding you know something even if you might be wrong,” he says.
Timeout for some relationship therapy: Let’s look at some of the many ways that we and the Nates and the surrounding Nate Industrial Complex did wrong by one another.
In a recent poll, The Washington Post estimated that Biden was leading in Wisconsin by 17 points. Biden appears to have won by less than one point. “That is not just wrong, that is holy s--- wrong,” said Frank Luntz, a Republican pollster. “That’s malpractice wrong.”
“We are taking time after the election to review all our polling, including why our October Wisconsin poll overestimated Biden’s support, and will use the findings to strengthen future polling,” Molly Gannon Conway, communications manager at The Post, wrote in an emailed statement. (She added that the balance of The Post’s polling was accurate and pointed to a Florida poll indicating a margin favoring Trump within one point of where he stands at the moment.)
If you went into the night pretty sure that Sens. Susan Collins of Maine, Joni Ernst of Iowa and Thom Tillis of North Carolina would be sent packing, you were not alone. But you were also not correct. All of them appear to have outperformed the polls. By a lot.
There are still plenty of votes coming in across the country that could cause these misses to creep closer to the margin of error. (“It is premature to make sweeping judgments on the polls’ overall performance before all the ballots are counted,” read a news release sent Thursday by the American Association for Public Opinion Research.) Still, even if Biden becomes president, polls in various states including Florida, Texas and Ohio made him look a lot stronger than he proved to be. And so while the election returns revealed a country divided, it made Democrats and Republicans more unified than ever in their bipartisan contempt for the polls.
“These polls are not to be trusted,” Van Jones, a Democratic commentator, said on CNN. “There is something wrong with the polling industry.”
“VERY STRANGE,” President Trump tweeted Wednesday morning. “The ‘pollsters’ got it completely & historically wrong!”
It has, pollsters will tell you, gotten harder and harder to poll in general. Fewer people have landlines, and who in their right mind picks up a call from an unknown number these days, unless they are curious about international scam centers. What are your views of President Trump’s handling of the economy? And what is your Social Security number?
But the past two election cycles have made it clear that there is a specific difficulty when it comes to reaching Trump supporters, who appear to have been undercounted in polls all across the country. “They’ve come to see pollsters as people who disrespect them,” Luntz says.
Luntz says he has tried to counter that problem by including new language in his surveys: “People like you have complained for years that nobody listens to you and nobody cares what you think,” he says he asks. “Here’s your chance to be heard.”
So where are the Nates in all this? They may not have been responsible for the bad polling — they assess the quality of polls and aggregate their wisdom — but that doesn’t necessarily absolve them of contributing to our collective freakout. New York Times Nate gave regular updates on his pseudo-sentient laboratory creation. “Biden is still a very narrow favored in the view of the needle,” he tweeted Wednesday, interpreting the oracle’s murmurings about Georgia. And as results trickled in throughout the night, commentary trickled out on Twitter from FiveThirtyEight Nate.
“I have nonpublic version of our paths-to-270 interactive where I can enter in provisional ratings instead of calling states outright,” Nate Silver tweeted at 5 a.m. “If you treat NV, AZ and WI as ‘likely Biden,’ NC as ‘likely Trump,’ and GA, PA and MI as ‘toss-up,’ then Biden is 83% to win eventually.”
“I am now a single issue voter,” tweeted journalist Heidi Moore, “and that issue is ‘stop nate silver from sharing every statistical hot take.’ ”
Ruth Bernstein, a Democratic pollster in California, says polling has become “too much of the news.” So much of the coverage she says is about the horse race instead of the issues, and more than that, the predictive polling may even have an effect on the voting itself. “If the Nates say that Trump only has a 10 percent chance of winning, maybe a voter in Michigan decides he doesn’t want to go out in a pandemic and decides to sit this one out,” she said.
Okay, but where are we, the bulb-eyed Nate watchers, in this feedback loop? Not so innocent ourselves. Perhaps we hate the Nates because we hate the polls, and we hate the polls because we hate ourselves. Maybe we’re looking for something the Nates can’t give us.
“We are having to confront what we know about America,” says Connie Schultz, a Pulitzer Prize-winning journalist and the wife of Sen. Sherrod Brown (D-Ohio). “We learned that it was not going to be this massive repudiation of all this horrible stuff that has been coming out of Donald Trump’s administration, and in fact, many people are quite fine with it, far more than the polling told us.”
So maybe we should just call it quits — for ourselves and for everyone else, says Jess McIntosh, a former Clinton 2016 staffer. “The less time we all spend playing pundit,” she says, “the healthier our democracy will be.”
In the end, the Nates will certainly find a way to tell us they were right all along. Biden does look as if he’s going to have a good shot at the White House, and in the end, that might be enough for voters to memory-hole the toxic parts of the relationship and come running back in two or four years.
“If you’re coming after FiveThirtyEight,” Silver said in a FiveThirtyEight podcast Wednesday, “Then the answer is: ‘F--- you. We did a good job.’ ”
He had defenders.
“For the record, national polls were at +8 for Biden, down from +11 earlier in Oct., so the final national vote margin may not be too much lower than this final 538 estimate,” a Louisiana professor tweeted Wednesday.
The professor’s name was Nathan Kalmoe. But as far as we’re concerned, you can call him Nate.