President Trump is an unusual politician. A real estate heir and reality television star, he had no experience in public office before running for president. He ignores democratic norms, by, for example, announcing foreign policy via Twitter, regularly attacking the press, and denying the legitimacy of election results. He says things that “no other politician could get away with” like that he prefers veterans who weren’t captured or that he can grab women by the. . . .

For all these reasons, and more, the media has struggled to understand and analyze Donald Trump’s popularity.

In 2015 and 2016, commentators consistently underrated Trump’s chances of winning the nomination and the presidency. As Tom McCarthy reported in the Guardian in 2015, many were certain that his initial strength in the primary polls was just a function of his celebrity name recognition, and that he would fade over time.

After Trump did win the Republican primary, pundits believed he would be doomed in the general election. Joe Wiener in the left-leaning Nation, for example, argued that Trump’s campaign was incompetent, and had poor fundraising numbers. Republicans seemed to agree that Trump’s campaign was hopeless, especially after a tape emerged in which he boasted about sexually harassing women. “Trump is underperforming so comprehensively a… it would take video evidence of a smiling Hillary drowning a litter of puppies [for her to lose]” said one Republican insider quoted in Politico.

President-elect Donald Trump shakes hands with Vice President-elect Mike Pence as he gives his acceptance speech during his election night rally, Wednesday, Nov. 9, 2016, in New York. (AP Photo/John Locher)

The puppy drowning video never came, but Donald Trump’s victory did. Then came the media scrambling to save face and explain this major upset to a dumbfounded national audience who couldn’t understand why the world the media had constructed for them didn’t seem true anymore. As humans have done for thousands of years, they responded by creating a mythology to explain something they got wrong, something they couldn’t understand—denouncing the former conventional wisdom of politics, highlighting Trump’s unwavering appeal to an unprecedentedly devoted base, and emphasizing that this President is special.

Some claimed that polling Trump was useless, perhaps because voters lied about whether they’d vote for him.  As a commentator for The Independent put it, “[t]he moment the real votes started to come in, our untested assumptions were suddenly exposed. Hispanics won’t vote for Trump. Well, no, it turns out that Hispanics won’t tell pollsters they’ll vote for Trump.”

Others, such as Van Jones, interviewed Trump voters to try to understand his appeal. Liberals even started reading and recommending some book called “Hillbilly Elegy.”

Over the course of his presidency, a hush-money scandal killed by FoxNews during the election eventually broke, and evidence began to mount that Trump may have obstructed investigations into Russian election interference. When special counsel Robert Mueller released his long-anticipated report on his office’s investigation into Russian interference in the presidential election, it laid out ten separate instances of conduct which may have constituted obstruction of justice.

Special counsel Robert Mueller speaks at the Department of Justice Wednesday, May 29, 2019, in Washington, about the Russia investigation. (AP Photo/Carolyn Kaster)

Frustrating Trump’s critics, Mueller did not reach any legal conclusions, citing the DOJ custom that a sitting president cannot be indicted. He then resigned his post and passed his report to Congress. House Democrats are still fighting over the proper course of action going forward.

Despite these and other scandals, numerous reporters have written pieces insisting that Trump’s base will continue to support him no matter what. As one commentator put it, “Trump’s working-class supporters continue to fascinate and flummox political experts in part because they are stubbornly, heartbreakingly hopeful that he will eventually keep his promises to them.”

This is a fiction that flows straight from the mythology built around Donald Trump’s presidency. The claim that candidate Trump couldn’t win was wrong. But the argument that President Trump can’t be polled, or that nothing he does can erode his popularity is just as wrong. We need to reject the myth of Donald Trump and look at the facts.

The truth is that polling gave Donald Trump a decent chance of winning the 2016 presidential election. FiveThirtyEight’s Nate Silver spent the week before the election making a lot of noise about people underrating Trump’s chances of winning. The problem wasn’t that the polls were off. The problem was that pundits were so sure that Trump couldn’t win, and critics were so averse to the idea that he could win, that they ignored any facts that were at all flattering to Trump’s chances.

In fact, polls since 2016 have accurately predicted subsequent results. Trump’s low approval in 2018 led pollsters to predict a big Democratic win in the midterm elections. Sure enough, Democrats scored a commanding victory, taking back the House and keeping the losses in the Senate to a minimum despite a very hostile map.

New York Democratic Congressional candidate Alexandria Ocasio-Cortez speaks to supporters, Tuesday, Nov. 6, 2018 in Queens the Queens borough of New York, after defeating Republican challenger Anthony Pappas in the race for the 14th Congressional district of New York. Cortez was one of many Democratic victors in 2018 that took the House back from Republicans. (AP Photo/Stephen Groves)

Nor is it true that nothing Trump does affects his polling.  During the government shutdown at the beginning of 2019, Trump’s ratings fell quickly. According to the FiveThirtyEight poll tracker, he dropped from 42.5% approval in mid-December to a low of 39.3% before he surrendered to Congress and reopened the government. Trump’s approval also rose after his first year in office. In 2017, he polled mostly under 40%; since then he’s mostly been over 40%. Clearly Trump’s numbers are movable.

It’s true though, that the numbers haven’t moved that much. Trump’s approval has stayed fairly steady—which is to say he’s been consistently unpopular. His approval has hovered between the high 30s and the low 40s for most of his term, according to FiveThirtyEight. His disapproval rating has varied between 52% and 55%.

Donald Trump’s approval and disapproval ratings over time (via https://projects.fivethirtyeight.com/trump-approval-ratings/)

Those are historically weak numbers. In fact, Trump’s disapproval has been the worst of modern era presidents for virtually every day of his presidency (though Jimmy Carter beat him out recently).

While Trump’s polling has been poor, it hasn’t been catastrophic. That’s probably because there hasn’t been a recession during his term. Recessions pushed Reagan into the mid-30s for approval during his first term; Carter fell into the 20s during his recession.  George W. Bush’s failures in Iraq, and Nixon’s Watergate disaster, dropped both presidents into the 20s during their second terms.

Donald Trump’s approval lags behind every modern president (via https://projects.fivethirtyeight.com/trump-approval-ratings/)

In fact, the economy is doing quite well. Real wages are finally increasing at a rate we haven’t seen since before the Great Recession. The unemployment rate is under 4%. Given the state of the economy, President Trump should be polling higher. But he could certainly be doing worse.

Trump, then, has benefited from a strong economy and a loyal base. Yet he has still yet to break 50% popularity on any day of his presidency. If his polling stays where it is, he will have difficulty winning re-election. But there’s a long way to go before November 2020, and his chances would improve dramatically if he could pick up only four or five points.

Trump has been viewed as variously impossible or inevitable, but he’s neither. At the moment, he’s a very unpopular president. While he nonetheless has a lot of room to rise or to fall, his tendency to do either does not defy prediction or elude relevant facts.