Swing Voters Aren't Centrists

Back in the aughts, political writer Matt Yglesias coined the “pundit’s fallacy,” which he described as, “that belief that what a politician needs to do to improve his or her political standing is do what the pundit wants substantively.” Put another way, it’s the belief that the political preferences of any given pundit are not unique to that pundit, or not minority preferences, but instead reflect the preferences of either the majority or at least of enough people that a politician can win by making them his own positions. It’s a confusion of perspective: “I see the world this way, and so most others must see it the same way, too.”

The psychological pull of the pundit’s fallacy isn’t unique to pundits, and it’s perfectly understandable. We hold the views we hold because we think they’re correct, even if we have bad reasons for thinking that, and most of us tend to see “lots of people believe them” as further evidence of their truthfulness. Yes, there are extreme heterodox thinkers who thrive on being the epistemic outsider, but few of us are extreme heterodox thinkers. We like to have intellectual compatriots. And, besides, heterodoxy, by itself, doesn’t tell us anything about whether the heterodox belief in question has any truth to it.

Anyway, I was thinking about the pundit fallacy while watching yet another round of calls, from elite opinion columnists and, yes, pundits, for the Democrats to “moderate” or “move to the center” if they want to win in 2026 or 2028 or whenever we have free and fair elections. The argument goes, Trump won because swing voters—those odd folks in the middle who don’t have any explicit or revealed party loyalty, but instead occeslate between Democrats and Republicans election by election—were turned off by Democrats not being middle enough.

The people making this argument are, by and large, those pundits who think of themselves as roughly “centrists.” And a great deal of pundits, especially the big deal Washington insider types, think of themselves as “centrists” because, in the culture they’re a product of, being seen as far from the middle (whatever the middle might be) is unseemly. As I wrote a last summer about the urge to false equivalence common in these circles, “To be thoughtful and wise is to be above the fray. To be above the fray is to be non-partisan. To be non-partisan is to not favor one side over the other.” Moderation, in other words, is a sign of intellectual sophistication, because to be immoderate is to be unmeasured and uncareful in your thinking.

What does this have to do with swing voters? If you listen to the sort of analysis in question, a swing voter is a political centrist. They swing because they are in the middle between those who don’t, and those who don’t are either far enough on the left or far enough on the right that they aren’t willing to switch sides in any election. To appeal to swing voters, then, and so win elections, is to adopt the political philosophy of moderation and centristicy. It is to stake out a position somehow between those of the Republicans and the Democrats. That means, for example, not being a Christian nationalist, but also not talking so much about how trans people ought to have rights of equal participation in society. Because both those “extremes” turn of centrists, and thus turn off swing voters.

How do the pundits know that’s what will turn off swing voters? Because it’s what what turns them off. And, if you’ll recall, those pundits are centrists and swing voters are centrists, and so the pundits can readily use their own preferences as an accurate enough proxy for whatever swing voter you grab from the crowd.

The trouble is, swing voters aren’t centrists. To the extent centrism is a coherent political ideology, swing voters can’t be centrists because swing voters don’t have coherent political ideologies. They tend, first, to be low-information and low-engagement. Both cut against critically examined, consistent views. If you think about the low-information voters in your own life, you’ll likely immediately see what I mean. They tend instead to have grab bags of quite random political beliefs, and like policies that directly contradict each other. They’re less about policy and more about vibes, excitement, and maybe some grievances. Yes, there are swing voters this doesn’t describe. There are actually, genuinely ideologically moderate (whatever that might mean) people who pay a great deal of attention, know quite a lot, and have parked themselves somewhere between the median Republican and the median Democrat in terms of policy. But that’s not most of them, or even very many. Trump won not because people like that decided to vote for him, but because tons of people who don’t tend to turn out for elections with any consistency turned out for this one.

This centrist fallacy is, ultimately, a more intellectually palatable version of the pundit’s fallacy. It’s transparently a pretty bad argument to say flat out, “Politicians can succeed by doing what I want them to do because I’m just assuming everyone else agrees with me.” You need something more than that to persuade anyone to take you seriously. So instead you add an epicycle: “Actually, swing voters are centrists, I’m a centrist, therefore swing voters want what I want.” It’s still not a good argument, but it’s good enough that plenty of people paid to write about politics believe it.