President Trump waves as he departs after speaking in the Rose Garden of the White House on June 14 in Washington. (Alex Brandon/AP)

Reality tends to be persistent. There are ways to coax it and shape it, to massage understanding of what happened after the fact. But, philosophy seminar classes notwithstanding, if a thing is, it’s generally harder to deny that fact than to accept it.

It is a reality that President Trump's political position is shaky as his reelection approaches.

To hear him tell it, there’s a populist groundswell of support for his presidency and himself, the sort of thing that prompts him to offer that “the people” may demand he serve beyond eight years if reelected. Some people might make such a demand, sure, and many of those people are the people who are regularly in Trump’s presence: at his rallies or in his White House. But the idea that there will be a massive push for overturning the 22nd Amendment so that Trump can remain in office seems at odds with his having both lost the popular vote in his only election to date and being approved of by only 40 percent of the country according to Gallup polling.

Polls serve as something of a thermometer for the reality of politics. There are mistakes, sure, but generally they serve as an accurate tool for gauging where the public stands. Since Trump took office, reputable polls have been consistent: Trump isn’t very popular and views of his presidency haven’t changed much.

Trump himself used to celebrate polls, lest we forget. During the Republican primary in 2015 and 2016, he surged to the front of polling and would celebrate new numbers at the beginning of his rallies. He’d literally pull poll numbers out of his pocket and describe his front-runner status to cheering crowds. As those same pollsters then declared that he was trailing Hillary Clinton in the general election — and then accurately pegged his popular-vote loss — polls became a nuisance or a threat. As recently as Monday morning, polls were “fake” if they showed him trailing possible Democratic nominees in the 2020 election — a response, it seems, to a poll released over the weekend showing him trailing former vice president Joe Biden and Sen. Bernie Sanders (I-Vt.) and in ties with a number of other possible Democratic candidates.

That poll was from Fox News, not generally an outlet that Trump paints with his “fake news” brush. This comes after a period in which Trump and his team denied the existence of polls conducted by their own campaign showing Trump trailing Biden — denials that were undercut significantly when ABC News obtained the actual poll numbers. So the denial shifted: Those polls were old and there were new ones that showed Trump doing quite well.

We haven’t seen those new polls in any detail at all. We’re left to trust the Trump campaign when it says that they exist, an obviously fraught request. Since the campaign itself is essentially the one source of information that Trump can’t deride as fake, it’s moving to cut off pollsters that it worries might have leaked the internal numbers — or which might leak future details.

Because, again, it's in a battle to offset the reality of its position. It's certainly the case that general-election polling taken more than 16 months in advance doesn't have a great track record of predicting outcomes. It's similarly true that, as in 2016, late changes in the election can dramatically reshape the outcome. But it is hard to say that Trump's reelection prospects are particularly rosy given the data we've seen.

Arguments to the contrary generally rely on the strength of the economy as an indicator of Trump's strong prospects. It's certainly true that presidents seeking reelection in the past have won when the economy has been doing as well as it currently is. It's also true, of course, that presidents at 40 percent approval have had an uphill climb when they next faced voters.

As we’ve noted before, we’ve already seen a test of these two conflicting indicators: Last November. Gallup data focused on the economy suggested that Republicans would lose fewer then 10 seats in the 2018 midterm elections. Data focused on Trump’s approval suggested they’d lose 39.

They lost 41.

What’s more, we’ve seen repeated polling that suggests Trump isn’t getting as much credit for the economy from voters as he’s giving himself. That Fox News poll released over the weekend contains similar warning signs.

Overall, a bit over a third of respondents said either that all Americans or Americans like them were the beneficiaries of Trump's economic policies. More than half said that the beneficiaries were either people who earned more money than they did — or no one.

Among independents, nearly 6 in 10 said either wealthier people or no one benefited, while about 2 in 10 said everyone or people like them did. Even among rural whites, a stronghold for Trump, responses were mixed.


(Philip Bump/The Washington Post)

Or consider another breakdown: Whites by education and gender. White men and women without degrees voted heavily for Trump in 2016 according to exit polls, but more than half of white women without degrees see the benefits of Trump's economic decisions helping the wealthy or no one at all.


(Philip Bump/The Washington Post)

The question becomes this: Where does Trump gain votes on the economy if even among voters who supported him three years ago there's skepticism about the effectiveness of what he's done?

Trump presents the economy as a key success of his presidency and has claimed, at times, that bad reporting has muted his approval rating despite that success. The reality? So far, it’s only his existing base that overwhelmingly agrees.

What's particularly baffling about Trump's presentation of the polls as inaccurate is that he could, alternatively, simply say that they were wrong. After all, he is the most robust existing demonstration of a failure of polling to predict an outcome that exists in American politics. While national 2016 polling was accurate, several state polls projected Clinton victories that didn't materialize — leading observers to assume that Clinton would win the electoral vote along with the popular tally.

So why not simply say, “The polls have been wrong before”? Why not note that he’s embarrassed the pundit class once already?

Because the frustration in the reality of Trump’s unpopularity extends beyond his political future. It is also personally frustrating to Trump that most Americans view his presidency and himself negatively. Trump’s electoral college win was quickly described as a “landslide”; his base of support quickly substituted in for the American people at large. Trump’s furiously reshaping the political world so that he’s a popular success who’s coasting to reelection, perhaps in part because his base’s enthusiasm seems to be tied to that perception.

But it’s not the reality that, to his frustration, insists on continuing to exist.