Shannon Stapleton / Reuters

A theory for how to win the 2020 presidential election has quickly become conventional wisdom among Democratic campaign strategists and many prominent pundits. It goes like this: The country has become so polarized that swing voters barely exist anymore. Elections are now decided by which side better manages to mobilize its base. So Democrats need to stop worrying about winning over moderates—and confidently move to the left.

Proponents of this “progressive mobilization theory” can point to a few important pieces of evidence. Plenty of liberal policies, for instance, are popular with voters across the political spectrum. As I’ve argued in the past, this makes it possible for Democratic presidential candidates to develop an ambitious agenda on issues from health care to gun control without jeopardizing their chances of ousting Donald Trump.

But the theory is nonetheless wrong in three crucial respects. A significant number of swing voters do exist, and in close-run elections, they matter. While Democrats do need to mobilize their base to win in 2020, it is far from obvious that moving to the left will help them do so. And to cure America of Trumpism, they need to persuade voters who aren’t already consistent progressives to turn their back on his brand of politics.

While many of the most politically engaged Americans really have sorted into mutually hostile tribes, the country as a whole remains far less politically partisan. According to recent polls, 42 percent of Americans call themselves independent and 35 percent consider themselves moderate.

The outcomes of the past two elections revealed the importance of such voters. According to an analysis by Geoffrey Skelley, for example, 11 percent of those who voted for Trump in 2016 voted for Barack Obama in 2012. Since all these voters subtracted a vote from the Democratic column and added one to the Republican column, they made a real difference in the outcome. If Hillary Clinton had been able to retain as little as one-fifth of this group, she would likely have won a comfortable majority in the Electoral College.

Swing voters also help explain how Democrats were able to stage such a spectacular comeback in the midterm elections. In 2018, the chaos and extremism of Trump’s first years in office drove a lot of moderate voters into the Democratic camp. As Yair Ghitza argued in a comprehensive postmortem, Democrats were able to take the House of Representatives because they won back a lot of voters in suburban swing districts across the country. Democrats also improved their share of the vote in rural areas, something that could make a crucial difference in key states such as Florida next year. Ghitza’s conclusion was unambiguous: “A large portion of gains came from people who voted in both elections, switching from supporting Trump in 2016 to supporting Democrats in 2018.”

None of this means that these voters—let’s call them “nonpartisans,” because they do not see themselves as either Democrat or Republican—are a cohesive political group. As Lee Drutman has recently shown, only a small percentage of so-called moderates or independents are centrists who consistently hold views more conservative than those of the Democratic Party and more progressive than those of the Republican Party. Instead, nonpartisans hold a variety of political views, and tend to be ideologically fluid.

Perhaps for that reason, many of these nonpartisans are easily turned off by what they see as extremism. Alexander Agadjanian found recently that when “independents who could ultimately tilt things in Mr. Trump’s favor” are presented with newspaper articles that emphasize Democratic support for abolishing private health insurance or decriminalizing unauthorized border crossings, they become “six percentage points less likely to vote Democratic.”

Oddly, many of the analysts who emphasize the incoherence of moderate and independent voters also tend to assume that the voters Democrats are trying to “mobilize”—the Democrats who often stay home—have highly coherent, and consistently progressive, views about public policy. Do they?

Advocates for the progressive mobilization theory make much of the fact that turnout in traditionally Democratic areas was lower in 2016 than in 2012 and 2008. But there is little reason to believe that Clinton would have improved voter turnout by moving to the left. As Vox’s Matthew Yglesias rightly pointed out, Americans who voted for both Obama and Clinton have more consistently progressive preferences on issues from immigration to the minimum wage than those who voted for Obama in 2012 but decided to stay home in 2016. It is, then, at best unclear why anyone should assume that a more progressive candidate would mobilize more voters in 2020.

The mobilization theory is right in at least one respect: To beat Trump in 2020, Democrats can’t just rely on winning over white voters in the suburbs; they also need to do a good job of turning out African American voters in states such as Pennsylvania and Ohio, and Latino voters in states such as Florida and Colorado. But again, it is far from obvious that a sharp turn to the left is the best way to mobilize these voters. After all, black and Latino Democrats are actually more likely to be moderate—or even conservative—than white Democrats.

Given Trump’s support for the far right, his cruelty toward minority communities, and his blatant abuse of presidential power, removing him from office in 2020 is of the utmost importance. But winning one presidential election is not enough to free America from Trumpism. Anybody who aspires to discredit the president’s toxic brand of politics must therefore care about the manner of his defeat.

If Trump holds a lot of independents and comes close to winning reelection, he is likely to retain enormous influence in the Republican Party. This would make it far more likely that a candidate in his mold would win the Republican nomination in 2024. The Trumpist nightmare would then haunt American politics for years or decades to come.

The best way to avoid this horror scenario is for Democrats to achieve a resounding victory next year. In 2020, they have a chance to reshape American politics by winning a mandate for real change. But they will be able to do so only by acknowledging that swing voters do exist—and by doing their best to win their trust.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.