Donald Trump’s historic unpopularity has yet to harden into conventional wisdom, even as he’s trailed previous presidents in job approval at nearly every point in his administration. Instead, mainstream pundits and reporters focus on his relatively high marks with Republican voters, convinced that this gives insight into his political standing. The idea isn’t completely unreasonable; without Republican unity, Trump would have lost the election. The moment GOP voters break with the president is the moment he’s undeniably in political trouble.
But this is also a truism: the moment any president loses his base is the moment he is trouble. A better measure, especially for the winner of a close election like Donald Trump, is performance with more marginal supporters: people who weren’t typical Republican voters, who didn’t like Trump himself, but who chose him over the Democratic alternative for contingent and provisional reasons. They are voters who liked Trump’s message on jobs and trade, and ignored his crude and inflammatory rhetoric—voters who wanted a generic change in the White House, and thought Trump might deliver.
We know that these voters were disproportionately white, and less likely to hold a college degree. And we know they were concentrated in the industrial Midwest, where Trump made dramatic gains over previous Republican presidential nominees. The latest national survey from Marist and NBC News polls several of those states, giving us a snapshot of the president’s popularity with voters he has to hold to win a second term. The picture isn’t great.
In Michigan, where Trump won by 11,000 votes, 54 percent of registered voters disapprove of his performance, compared to 36 percent who approve. In Minnesota, where he narrowly lost, 51 percent currently disapprove, compared to 38 percent who approve. And in Wisconsin, a state he won by 23,000 votes, 52 percent disapprove, compared to 36 who approve.
By wide margins, voters in these states want a Democratic Congress. They want that Congress to act as a “check and balance” on Trump. Despite the growing economy, few give him credit for economic improvement, and looking ahead to 2020, nearly two-thirds in each state say it’s time to “give a new person a chance” in the White House.
Trump’s unpopularity, nationally, and Democrats’ growing advantage on the congressional generic ballot haven’t made a substantial difference in how the president is covered. Some of this reflects the way his most visible supporters—white, blue-collar workers—are used to symbolize an authentic “Americanness” which then colors perceptions of Trump himself. If he wins these most American of Americans, then perhaps he still channels the popular will in some imperceptible way. Likewise, some of this reflects shell shock from the 2016 election: Even those aware Trump could win didn’t believe he would, prompting a crisis of epistemology when he did. If we were wrong then—ignorant of his appeal to millions of Americans—perhaps we are wrong now.
But the lesson of 2016 isn’t to ignore the polls; it’s to pay greater attention to what they’re actually saying. On the eve of the election, they were increasingly favorable to Trump. As Nate Silver of FiveThirtyEight wrote at the time, “the possibility of an Electoral College-popular vote split keeps widening in our forecast.”
What are polls saying now? That Trump, and his party, are in serious political danger.
If you think Slate’s work matters, become a Slate Plus member. You’ll get exclusive members-only content and a suite of great benefits—and you’ll help secure Slate’s future.