Biden should assume the polls are right, not wrong

For months, Democratic strategists, the Biden campaign and even the president himself have failed to make sense of the numbers. The hope was that, as the economy improved, so, too, would Mr. Biden’s political fortunes. This year, the stock market is up over 10 percent, reaching record highs. Inflation is easing. By most metrics, the U.S. economy is the strongest in the developed world. Yet Mr. Biden’s disapproval rating has continued to rise from 53 percent last fall to over 56 percent today.

Polls suggest that several of Mr. Biden’s core constituencies — young people, Black people and Hispanics — are increasingly Trump-curious. In a Wall Street Journal poll of swing states, 30 percent of Black men said they would definitely or probably vote for Mr. Trump. For the disaffected, Mr. Trump offers the promise of radical change. In the Times poll, these “tear-it-down” voters — some 15 percent of registered voters — prefer Mr. Trump by 32 percentage points. For anti-system voters, what could be better than a candidate who promises to destroy that system?

In response to these findings, Mr. Biden has attacked not just individual polls but polling writ large. “The polling data has been wrong all along,” he said in an interview with CNN. “How many — you guys do a poll at CNN. How many folks you have to call to get one response?” Speaking to donors on May 10, Mr. Biden insisted that “while the press doesn’t want to write about it, the momentum is clearly in our favor.” That is not an attitude on which to build a presidential campaign.

To be sure, counterintuitive polling results such as these deserve scrutiny. Some of the data on Black or young voters suggest a historic swing among these demographics away from the Democratic Party, an astonishing finding. Pollsters could be struggling to reach representative samples of these groups. Perhaps certain types of voters are more likely to pick up their phones or fill out online surveys, a bias that could be uniquely distorting this election cycle.

Polls are just snapshots of public opinion. Some respondents might be registering disapproval of the president now, but will shift when the choice between Mr. Biden and Mr. Trump comes into focus. Most Americans aren’t paying close attention to the presidential race — yet. In the 2016 election, much of the movement in the polls happened in the fall. Polls show Mr. Biden faring poorly among low-information voters — those least likely to be following the news.

But it would be a mistake for the president to take comfort in these caveats. Despite some past misses, poll aggregates such as FiveThirtyEight’s still provide the most reliable information about the state of public opinion. Even in the surprising 2016 election, Nate Silver’s model, based in large part on polling figures, gave Mr. Trump a roughly 30 percent chance of winning — a substantial shot that should not have been discounted.

Yet Hillary Clinton campaign officials were overconfident, having already persuaded themselves that there was no way Americans would vote for a candidate as unpredictable, unqualified and outright dangerous as Mr. Trump. What was clear to partisans was, apparently, unclear to a great many Americans.

Similarly, Mr. Trump’s conviction in the New York hush money case may further the illusion of his unelectability. Would Americans really vote for a felon to be their president? Some initial post-trial polling points to a slight bump for Biden. But slight bumps might not be enough.

If it’s the economy, stupid, then the status quo might actually be good. But many Americans are saying otherwise. Voting is an emotional, personal act, not necessarily one determined by a careful assessment of the relevant macroeconomic indicators. Rejecting the polls — relying instead on anecdotes, instincts and vibes — is political malpractice and, in this highly consequential election, dangerous for the country. Rather than assuming the polls are wrong, Mr. Biden should assume they are right — and act accordingly.