For the great majority of Americans who have firmly settled on Kamala Harris or Donald Trump, the idea that anyone could still be undecided in that choice is almost incomprehensible. But the incredulity may be rooted in confusion about who most undecided voters really are.
When most people think about a voter still trying to make up their mind, they probably imagine a person who is highly likely to vote but uncertain whether to support Harris, Trump, or a third-party candidate. Both political parties, however, are more focused on a different—and much larger—group of undecideds: potential voters who are highly likely to support Harris or Trump, but unsure if they will vote at all.
Campaigns typically describe the first group of reliable but conflicted voters as persuadable; they frequently describe the second group as irregular voters. Persuadable voters get the most attention from the media, but campaigns recognize that irregular voters can loom much larger in the outcome—especially in presidential elections when more of them ultimately participate.
“There are a gajillion more of those [irregular] people than the Harris/Trump ‘I don’t know; I’m still thinking about it’” kind of voter, Anat Shenker-Osorio, a communications consultant for Democrats and progressive groups, told me. “There are more humans who are non-habitual voters than there are voters who swing back and forth. That’s just math.”
The first group of undecided Americans—the persuadable voters still vacillating between Harris and Trump—are always the subject of intense media focus. Pollsters use an assortment of questions to gauge how many people fit that description. The NBC News national poll released Sunday, for instance, found that almost exactly one-sixth of voters either declared themselves undecided in the race or said that there was at least a chance they would switch from the candidate they’re now supporting. The most recent national Pew Research Center survey likewise found that the same proportion of Harris and Trump backers said that they either were merely “leaning” toward their candidate or could change their mind. The latest New York Times/Siena College national poll put the shares of undecided voters and persuadable voters at almost exactly the same level.
All of these results suggest that the pool of likely voters not firmly bound to either Harris or Trump is more than large enough to tip the election. The problem is that most strategists in both parties consider those numbers an illusion: They do not believe that roughly one-sixth of likely voters are ambivalent enough about one candidate that they could still switch to the other before November.
“There is an immaterial number of ‘certain to vote’ people who are undecided,” says the longtime GOP pollster Bill McInturff, whose firm has conducted the NBC poll along with a Democratic partner for decades. This is a view widely shared among strategists in both parties.
Mike Podhorzer, a former AFL-CIO political director who has built a large audience among Democrats and progressive groups for his detailed analyses of voting behavior, says that traditional polling questions significantly overstate the number of voters truly up for grabs between the parties. “There are people who will say that they are undecided in a survey,” Podhorzer told me, “and it’s just not true.” Podhorzer says that in polls he’s commissioned over the years, he always asks voters whether they have mostly voted for one major party or the other in the past.
“The effect of turning the question from making a statement about how you identify yourself to reporting on your previous behavior was kind of jaw-dropping,” he told me. “Almost all” of the people who said they were undecided at any given time turned out “to actually be on one side or the other. It was just how they were asked.”
Jim McLaughlin, a pollster for Trump’s campaign, notes that as the electorate has grown more polarized since 2000, winning presidential candidates of both parties have shifted strategy. “You look at Obama’s election,” McLaughlin told me. “It was a turnout election. The same thing with George W. Bush. You’ve got to keep that base motivated, so you are messaging toward that—and what they are voting for and against matters.” This dynamic has only hardened in the age of Trump. “No question, there are not a lot of ‘persuadables’ at this point,” McLaughlin said.
Among the operatives and strategists that I spoke with in both parties, the best estimate is that just 4 to 7 percent of voters in the battleground states are such persuadables—people highly likely to vote but genuinely uncertain about whom they will support.
These include people like Fred, a white project manager from Minneapolis, and Ronmel, a Hispanic securities analyst from Dallas, who participated in a focus group of undecided voters convened in late August, after the Democratic National Convention, by Sarah Longwell, a political consultant and the executive director of the anti-Trump Republican Accountability Project. (Longwell’s focus groups reveal only the first names of participants.) Although both men had supported Biden in 2020, neither was ready to commit to Harris. “I think the issue with Kamala for me is that she does not have or has conveyed the gravitas for the role,” Fred said. Ronmel expressed frustration over inflation under Biden: Even though “you’re making a good living, you still feel like you’re living paycheck to paycheck,” he said.
When Longwell’s firm contacted the two men again last week, after the Harris-Trump debate, Fred had made his choice: “Kamala eliminated all my doubts about gravitas: She is 100 percent ready to be president on day 1.” Fred wrote in a text. “Trump, on the other hand, exacerbated every concern I had.”
But Ronmel was still conflicted. “They don’t seem to have any clear economic project,” he texted, “only promises that we know are not going to be fulfilled.”
The remaining persuadable voters, strategists and pollsters told me, are mostly people like Ronmel who believe that Trump’s presidency generated better results than Biden’s has, particularly on the economy, but who remain hesitant about entrusting Trump again with the presidency. (They cite various doubts—about his character and his views on issues beside the economy, such as abortion rights.)
These persuadable voters wavering between the two candidates split mostly into two camps. The largest group may be the traditionally Republican-leaning voters (including many who identify as independents) uneasy about Trump. These voters are the remnants of the suburban, largely college-educated constituency that favored Nikki Haley during the GOP primaries.
Based on the focus groups she has conducted with a wide array of voters, Longwell said that the persuadable voters “who are left are [mostly] two-time Trump voters who don’t want to vote for him again but are really struggling to get to [Harris].” After listening carefully to their answers and watching their body language, she told me that she expects most of these voters to support Harris eventually, because they are now so resistant to Trump. But she also believed that some of them are “leave-it-blank types” and won’t vote for either candidate.
The other big group of potentially persuadable voters, according to the NBC, Pew, and New York Times/Siena polls, are younger and minority voters who dislike Trump but are disappointed by their economic experience under Biden—and are uncertain whether Harris offers a sufficient change in approach. In the recent Pew survey, Hispanics who currently support Trump were much more likely than white voters to indicate that they might change their mind; for Harris-leaners, both Hispanic and Black voters were more likely to say they might reconsider. For both candidates, more younger than older voters indicated that they might switch.
In the end, however, neither party expects too many of the voters who are telling pollsters today that they might switch to the other candidate to actually do so. The bigger prize for the two campaigns is the irregular voters who are, as Longwell put it, deciding “whether they are going to get off the couch” to vote at all.
How many of these irregular voters are available for the campaign to pursue? Even in the 2020 election, which produced the highest turnout rate since 1900, about one-third of eligible voters didn’t vote. That’s about 80 million people. About two-fifths of both eligible people of color and white people without a college degree didn’t vote last time; neither did nearly half of young people.
Those patterns frame the 2024 mobilization challenge for each party. Catalist, a Democratic voter-targeting firm, shared with me data rarely disclosed in public, based on its modeling, that attempt to quantify the number of infrequent voters in each of the swing states who lean strongly toward Harris or Trump. That research shows, first, that across the battleground states white people without a college degree routinely account for 70 percent or more of the Trump-leaning nonvoters; and, second, that people of color make up a big majority of Harris’s potential targets across the Sun Belt battlegrounds, as well as in Michigan. In the three big Rust Belt battlegrounds—Michigan, Pennsylvania, and Wisconsin—working-class white women without a college degree, Catalist’s projections show, also make up a significant share of the voters who lean Democratic but don’t vote regularly.
The infrequent voters on both parties’ target list have some common characteristics, other strategists say. “Part of what you are seeing in this electorate is: a) a lot of anger; but b) discouragement,” Page Gardner, a Democratic expert on voter turnout, told me. “People are discouraged about their lives and feel … I’m trying really hard and I’m not getting anywhere.” Against that backdrop, she said, the challenge for Democrats is “giving them some sort of agency to feel like My vote matters, because a lot of people feel that no one is paying attention to them.”
As a lead organizer for the Sunrise Movement, a liberal group focused on mobilizing young people to support action on climate change, Paul Campion knows the challenge of engaging irregular voters for Harris. Sunrise is trying to reach young voters of color in battleground states through a combination of phone-banking, door-knocking, and text-messaging.
Like other campaigners seeking to organize young and non-white voters, Campion told me that “the biggest issue is not people choosing between Trump and Harris, but choosing between not voting … or voting for Harris-Walz.” Campion sees a fundamental conflict between Harris’s attempts to reassure centrist swing voters, by emphasizing moderate positions on energy from fossil fuels and on the war in Gaza, and her need to activate more progressive young voters uncertain whether to vote at all. “Young people want to hear Harris articulate over and over again more forcefully how she will fight for them and listen to their demands,” Campion told me.
For years, Podhorzer, the former AFL-CIO official, has been among the Democrats who have argued most ardently that expanding the electorate—rather than focusing on the smaller number of genuine swing voters—can be the key to the party’s success. This, he argues, is especially true when competing against Trump, who has proved so effective at activating his own constituency of infrequent voters. Podhorzer has calculated (using data from Catalist) that about 91 million separate individuals have turned out at least once in the four national elections since 2016 to vote against Trump or Republican candidates, while about 83 million have come out to vote for Trump or the GOP.
Although Democrats have improved their performance in recent years among the most reliable voters—largely because the party has gained ground among college-educated white people, who vote more regularly than any other major group—Podhorzer has calculated that people who voted in all four national elections since 2016 still narrowly favored the GOP in the battleground states. In those crucial Electoral College states, however, Democrats have posted commanding advantages among the infrequent voters who entered the electorate only after Trump’s victory in 2016. That group is disproportionately younger, Black, and Latino. This surge of new voters has been crucial in creating what Podhorzer and other Democratic strategists such as the Hopium Chronicles author Simon Rosenberg call the “anti-MAGA majority” that mostly frustrated GOP expectations in the elections of 2018, 2020, and 2022.
Shenker-Osorio said that replacing Biden with Harris has engaged more of these less reliable voters resistant to Trump. “When we were in the place of an exact rematch between the same two people that we had in 2020, the election was boring for a lot of people,” she told me. “And now it’s Okay, we at least cast somebody different in this season of the reality show, so that’s good.” But Shenker-Osorio added, the level of concern among these inconsistent voters about the potential downsides of another Trump presidency still has not reached the level Democrats need. “The task is to raise the salience of the election itself … and its pivotal role as a crossroads between two extraordinarily different futures,” she told me. “That is just something we have to hammer home and lift up.”
The thin sliver of reliable but persuadable voters still undecided between Harris and Trump matter in the crucial states, Podhorzer said, “because everything matters” there. But he predicted that whichever party turns out more of the irregular voters in its favor will win those states. That’s the bitter irony of modern U.S. politics: In a country divided so ardently and irrevocably between the two parties, the people who aren’t sure they care enough to participate at all are the ones who could tip the balance.
