Kamala Harris needs weird voters

Why she wants to go on Rogan

Tomorrow (Wednesday) at 11 a.m., I’ll conduct a live Election Dialogue with Matt Yglesias on the Substack app. Matt’s outstanding newsletter, Slow Boring, has one of the highest audience overlaps with Silver Bulletin of any other Substack newsletter, so I know many of you already know him. But I expect the conversation to be fun, and I hope you’ll be able to join us. I’ll send out a reminder on Twitter and Substack Notes before the conversation.

---

Back during Brat Summer, after she replaced Joe Biden on the ticket, Kamala Harris’s running mate, Tim Walz, adopted a strategy of describing Donald Trump and JD Vance as “weird.”

It may have paid some dividends: Vance remains among the least popular vice presidential picks ever. But this messaging was conspicuously absent in Walz’s surprisingly cordial debate with Vance earlier this month, which instead was a relic from a bygone, less weird political era. That may be because the “weird” message didn’t actually poll very well, according to Pod Save America’s Dan Pfeiffer:

Does calling MAGA Republicans weird actually move voters?

Well, I hate to be the bearer of bad news, we haven’t (yet) found the secret sauce. According to private research done by a top Democratic outfit, calling MAGA Republicans “weird” has minimal impact. The study showed voters various clips of Democrats using the message and then measured how the messages impacted their vote choice. The study found not [sic] significant impact one way or the other.

The lack of a backlash is notable because using a term like “weird” would seem condescending to some voters — something akin to Hillary Clinton’s deplorables comment in 2016. There is no downside to the message, but it doesn’t move the race in our direction. In other words, calling Donald Trump and JD Vance “weird” may be more cathartic than constructive.

Pfeiffer is a reasonably straight shooter, but I wonder if he’s being generous to Democrats in suggesting the “weird” attack was merely neutral rather than having the potential to create a backlash. Because Kamala Harris has a problem: she needs more voters than she currently has. And she needs more of them than Trump. So she needs some of the weird ones.

What do I mean by that? Harris currently has 49.4 percent of the vote in our national polling average:

Meanwhile, Trump has 46.5 percent. That means about 4 percent of the electorate is undecided or says they plan to vote for a third party.¹ Some will actually vote independent, but our model thinks it’s only likely to be 1 or 1.5 percent of the electorate. So Harris needs her fair share of the roughly 3 percent of the remaining truly undecided voters — or she needs to pick off some voters who are only loosely committed to Trump without losing her weakly committed voters to him.

More specifically, Harris has a target of 51 percent of the vote. In our simulations on Monday, Harris projected to get 51.0 percent of the popular vote — and the Electoral College was a toss-up. More than that (even Biden’s 51.3 percent from 2020 would probably do it) and she likely wins; less, and she probably loses. Trump, conversely, has a target line of 48.7 percent of the vote; that reflects the likely slant of the Electoral College toward the GOP.

And those voters may not be so easy to find. Unlike in past elections, Democrats no longer have an edge in party identification; about as many people now identify as Republican as Democrat. (This is also confirmed by party registration data in states that track it by party.) The Democratic Party used to be sort of an island of misfit toys, a coalition of various types of underrepresented minority groups. Now, they’re the party of suburban, college-educated normies. But this is actually a smaller coalition. About 60 percent of the projected November electorate will not have a college degree, according to the recent New York Times/Siena College national poll.

So just who are those undecided voters? I’m going to look again at that NYT/Siena poll. I know we’ve talked about this poll a lot recently, but it has some particular advantages for these purposes. The Times survey has a large sample size (3,385 voters) and goes to great lengths to reach marginal voters that other polls miss (who are more likely to be undecided). Plus, the Times uses a probabilistic rather than a deterministic likely voter model. That’s important because some of these marginal voters will vote, even if we can’t be sure exactly which ones. So here are their demographic crosstabs on voters who say they’re undecided -– or that they’re only “probably” rather than “definitely” committed to their candidate:

Young voters are much more likely to be not fully committed to their candidate choice than older ones, nonwhite voters more likely than white ones, independents more likely than Democrats or Republicans, and noncollege voters more likely than college grads (though I’d have guessed the gap would be bigger).

For what it’s worth, this does seem slightly favorable to Harris, since these are typically Democratic-leaning groups. But I can also easily imagine some Democratic stategists misinterpreting this data, because the priorities of undecided voters don’t at all match those of highly politically engaged voters. Harris hasn’t locked down the youth vote? It must be because of her position on the Israel-Palestine conflict, right?

No. It’s almost certainly not that, or at least not mostly that. In fact, the Middle East ranks as an extremely low priority to most young voters:

This sort of thing can be a problem for Democrats because the political class that works on campaigns or covers them for high-prestige media outlets is almost entirely plucked from college-educated Americans, often from privileged backgrounds. These people have strong and consistently left-of-center political opinions. Undecided and loosely committed voters, by contrast, are often politically moderate or have a checkerboard of opinions. They don’t like the major party brands and don’t trust the mainstream media.

Say what you want about Trump, but he’s always been willing to go after voters wherever they might be found — and however weird they are. He secured the endorsement of Robert F. Kennedy Jr., while Biden and Harris have never seemed to have any plan for attracting Kennedy voters. He went to the South Bronx. In 2019, he attended a massive rally with Indian Prime Minister Narendra Modi. He’s gone on podcasts with a variety of crypto-adjacent, bro-ey influencers. And slowly his favorable ratings have crept up — they’re worse than Harris’s, but considerably better than they were four or eight years ago.

Harris, to her campaign’s credit, has pivoted away from her media-shy strategy of Brat Summer — and importantly, she’s speaking to an eclectic range of audiences, from the Call Her Daddy podcast to an interview set to air tomorrow with Fox News.

She’s also seeking to book an interview with the immensely popular podcast host Joe Rogan.² Why? His audience. They’re young, male, and politically independent, and Rogan overindexes with Hispanics, according to Morning Consult data. You’ll find far more undecided voters here than in the 60 Minutes crowd.

Harris has also sought to be more crypto-friendly than Biden was, to the great annoyance of Democrats in my Twitter feed who find crypto to be cringe. But crypto holders also share some of these characteristics, distrustful of the government and more likely to be Black than the general population.

Look, I’m not going to vouch for every policy position Harris establishes or every interview she books as tactically brilliant. Still, it’s at least directionally speaking the right idea to expand the universe of voters she’s talking with. Because Harris needs a majority of voters and — I think I can say this as a Weird American myself — weird voters are in the majority.

1. Our polling average uses of polls that include third parties and those that don’t. Often, though, third-party voters are a placeholder for undecided.

2. Hey, Joe — I’ll come on your show too.