Polarization narratives tell us that a tsunami like the 2026 midterms wasn’t supposed to be possible. Everyone has their side, the argument goes, and they don’t cross over. Democrats can’t win seats in places that President Trump won by more than 5 points because political gravity takes over. Parties are playing only for the 10-15 percent of voters who might swing in any given election.
The midterms just shredded those narratives on two dimensions: Large blocs of voters have dealigned from the parties, putting their votes up for grabs every cycle, and polarization doesn’t account for swings in who shows up to vote.
The first point has received considerable attention. Per Gallup’s surveys, nearly half of all Americans say they are independent—45 percent—up from about 35 percent in 2000 and 33 percent in 1990. Among Hispanic Americans, the figure has grown to 55 percent, up from 39 percent in 1999. A 56 percent majority of Gen Z-ers identify as independent, compared to 47 percent of millennials in 2012, when they were the same age as Gen Z is now, and 40 percent of Gen X-ers in 1992. Millennials have climbed to 54 percent independent now, while Gen X is steady at 42 percent.
These trends have been less evident in the electorate because, historically, many independents didn’t vote. Talk of a polarized “electorate” has always referred only to those who reliably participate. Those voters typically have their side and rarely stray from it. As a pollster, I appreciate their steadiness, since everyone else is about as predictable as a cat with the zoomies.
What polarization arguments never really grappled with was the fact that while these stable partisans are a large part of the electorate that votes, they are well under half of the potential electorate. When only 49 percent of eligible voters show up, that means we have another half completely unaccounted for. These people are not polarized at all; they are disengaged.
We know roughly who falls into the nonvoter group, with the caveat that the composition of “nonvoters” also changes with each election. My own polling data collected in late September showed that those who were not registered to vote or were registered but not likely to vote were disproportionately Black or Hispanic, young, and less educated compared with those who said they were certain to vote. Among the not-likely voters, 60 percent were independent or not sure of their partisanship, compared to 31 percent in the certain-to-vote group.
What distinguishes nonvoters from voters is engagement, not attitudes. In my survey, the less-likely voters were even more focused on the cost of living as the country’s biggest problem, and they disliked President Trump just as much as certain voters. The less-likely voters are simply tuned out: Nonvoters answer “not sure” more often than voters on every question throughout the survey.
Only 24 percent of less-likely voters say they intentionally look for news and information about politics, compared to 56 percent of those who plan to vote. Nonvoters report seeing political information most frequently on YouTube and social media, while voters report using a wide array of media sources.
Perhaps most telling, 59 percent of nonvoters said the election results would make no difference in their lives or that they weren’t sure how it would impact them over the next two years. Only 19 percent said the outcome might improve their lives, and 22 percent said the outcome might worsen their situation. This is considerably more pessimistic than among voters, 43 percent of whom said the results would improve their lives, and 41 percent of whom said they would make no difference or they were not sure.
Many of these “nonvoters” will simply never vote, but the natural churn in the electorate means some will in any given year. These are the true wild-card swing voters, and they are not even remotely polarized. They also made up half of the potential electorate this year.
Combine that trend with the dealignment among young voters and Hispanic voters who do cast ballots, and you get a very different picture of the electorate than polarization narratives would have you believe. You also get a very different picture of your fellow humans, one that is less focused on politics and who votes for whom, and more focused on the human experience.
Polarization tells us that most people are stuck in their partisan corners and that persuasion isn’t worthwhile. The dealignment narrative tells us that Democrats and Republicans alike have to give people a reason to engage and support them; otherwise, they will just keep renting voters who switch sides—or completely disengage—when it feels like nothing helps improve their lives.





