There are two major theories about President Trump’s standing heading into his re-election campaign. Over the last few months, they have found backing from two very different kinds of polls.
One theory holds that Mr. Trump is fundamentally like any other president. This would be good news for his chances in 2020: Many presidents have gone on to win after having approval ratings like Mr. Trump’s today, and many presidents have won after a midterm drubbing. The state of the economy could be pivotal; if it stayed strong, he would have a real chance to win. His approval rating could rise, like that of prior presidents, once voters began to assess his presidency in comparison with the alternative. His relative advantage in the Electoral College could put him over the top.
Another theory holds that President Trump and this polarized era are unique. In this view, Mr. Trump is deeply unpopular, and opinions of his presidency are entrenched. The economy cannot save him from defeat. After all, his approval ratings are poor despite low unemployment; nothing short of fundamental changes in his conduct could improve them. In this view, the 2018 midterm election, when Democrats won the national House vote by almost nine percentage points, would be a harbinger of the general.
Of course, the truth could be somewhere in between (and probably is). But these two theories have different consequences for how to think about the campaign in 2020.
In the more traditional view, Mr. Trump would have a solid chance following the same playbook as George W. Bush and Barack Obama, at least if the economy stayed strong enough. He would attack his opponent and use wedge issues (like immigration) to both mobilize his base and lure back some voters who are dissatisfied with his performance. Democrats, meanwhile, would want someone who could keep voters focused on why they’re dissatisfied with the incumbent’s conduct, defuse wedge issues, and appeal to the sort of Obama-Trump voters who have soured on the president.
But if Mr. Trump and this era are unique, then 2020 might hinge on turnout: In a polarized environment with few persuadable voters left, little else would matter. Democrats might not need to worry about whom they nominate, as long as they can energize irregular voters. In the extreme, you could argue that the president has basically already lost re-election: Voters have made up their minds, and too many dislike him for him to win.
These theories are impossible to test before the election. But the main evidence for the polarization theory comes from the polls. They have always shown the president’s approval ratings well under water, with around half of voters saying they strongly disapprove of his performance. If that ever stopped being true, it would essentially disprove the theory.
That’s why movement in the polls has been particularly interesting lately. Over the last few months, online polls and live-interview polls have split in a way that would either support or undermine each of these theories, depending on which set of polls you believe.
The online polls support the polarization story: They show that the president’s approval rating has been astonishingly steady throughout his presidency, including over the last month.
But the telephone polls have been more variable, particularly over the last few months. In June and July, live-interview polls seemed to show the president’s approval rating matching the highest level of his term. If that’s accurate, it undermines the polarization theory and suggests the president still has the ability to broaden his appeal.
The president’s peak came in early July, just after the first Democratic debate, which could mean that the Democrats helped his cause, perhaps as a result of voters assessing his performance through the lens of the alternative. Democrats might have been particularly helpful to Mr. Trump if they repelled some persuadable voters by focusing on busing, ending private health insurance, extending health insurance to undocumented immigrants, or decriminalizing unauthorized entry into the United States.
Since then, the president’s ratings have dropped by several points, and there is no shortage of potential explanations: his verbal attacks on four congresswomen; mass shootings in El Paso and Dayton; growing concern about the economy and the trade war. A decline in the president’s ratings wouldn’t be good for his chances, of course, but it nonetheless suggests that public opinion is at least somewhat responsive to events, the economy and his conduct. If his ratings can go down, they can also go up.
It’s hard to know which set of polls is right. Live-interview polls have long been considered the gold standard in survey research. But they’re increasingly rare, which makes it harder to be sure of where they stand at any given time. They also aren’t typically weighted by a measure of partisanship, like party registration, so they’re more susceptible to shift with changes in which groups are likeliest to respond to polls.
But the online polls aren’t fully mature. There is plenty of reason to think that it remains harder to design high-quality online surveys than telephone polls. Many remain unproven.
Polling averages struggle to bridge the divide. An average might simply give more weight to whichever class of poll has published most often, which would generally give more weight to the online polls.
A more sophisticated approach is to adjust for the past tendencies of a pollster, called a house effect. This would help if the online polls were consistently better for Mr. Trump than the live-interview polls, as they have been for most of his term. But it is of little use if some polls turn out to be more variable than others. It could be outright counterproductive if house effects turned out to be inconsistent, as was the case when Mr. Trump began to gain ground in the live-interview surveys. For the first time, the online polls were worse for Mr. Trump than live-interview polls, yet a house effect adjustment would have assumed the opposite.
Either way, the president’s approval ratings are well under 50 percent, lower than they were heading into the 2018 midterm elections, and probably lower than they were a month ago. And the chance of a recession seems to have increased. Whatever your preferred theory of today’s political climate, it’s not an optimal position for a president heading into re-election.