For the Republic
← Command Center / 🎙 Episode / 2026-02-13 · ~12.5 minutes (~1,880 words)

The Sand Castle

Draft Complete — Pending Host Review

Pitch

2/10
pitch.md

Episode Pitch

Headline

Trump's coalition was never his -- it was borrowed from the national mood, and the bill just came due.

Thesis

Trump didn't win in 2024 by converting America to MAGA ideology. He won because roughly a quarter of the electorate wasn't paying enough attention to know what he'd actually do -- they just knew they were mad about prices under Biden and voted accordingly. New polling shows those same low-information voters have swung 25 points against him, driven overwhelmingly by the one thing they do notice: the cost of groceries. The uncomfortable truth is that Trump's famous "expanded coalition" was never ideological -- it was atmospheric -- and an atmospheric coalition evaporates the moment the weather changes.

Why Today

G. Elliott Morris at Strength In Numbers just published a detailed analysis of January 2026 polling data (with Verasight) showing the staggering collapse of Trump's support among low-knowledge voters -- from +11 in 2024 to -13 now. This comes as Trump's overall net approval hits -13.7, worse than Biden's at the same point in his term. With 2026 midterms nine months away, this is the moment to understand why Trump's numbers are cratering and what kind of voters are driving it -- because the answer reshapes how we think about both Trump's power and Democratic strategy.

The Hook

Open with the single most striking number: In 2024, Americans who consumed no news at all backed Trump over Harris by 20 points -- 60 to 40. Not Fox News viewers. Not Newsmax diehards. People who watched nothing. Today, that same group's support for Trump has collapsed to 43 percent. Pause on that. The people who knew the least liked him the most. And now they don't. That's not a polling blip. That's the sound of a coalition built on sand meeting the tide.

Key Evidence

  • Low-knowledge voters (those who can't identify which party controls Congress) backed Trump by +11 in 2024. They now disapprove of him by -13 -- a 25-point swing. High-knowledge voters moved only modestly, from +2 Harris to -14 Trump. (Morris/Verasight January 2026 poll)
  • The gap is driven overwhelmingly by prices: low-knowledge voters disapprove of Trump on inflation by -40, versus -30 among high-knowledge voters -- a statistically significant 10-point difference. On every other issue (economy, trade, foreign policy, immigration, health care, government funding), the two groups are nearly identical.
  • Low-knowledge voters skew lower-income, younger, less educated -- the people who spend the largest share of their income on groceries and essentials, making them most exposed to tariff-driven price increases. A $100/month grocery increase is background noise at $100k income; it's 5% of after-tax pay at $35k.
  • 19% of low-knowledge adults report losing coverage or facing premium increases since enhanced ACA subsidies expired at end of 2025, versus 11% of high-knowledge adults -- a compounding economic squeeze.
  • Only 58% of low-knowledge voters hold "strong" opinions about Trump (either direction), versus 74% of high-knowledge voters -- confirming that these are soft, movable opinions, not ideological commitments. This tracks with John Zaller's framework: without ideological filters, opinions follow conditions.
  • Trump's overall net approval is -13.7, lower than Biden's at the same point, with strong disapproval breaking 46% for the first time. Presidential approval below 50% has predicted the incumbent party losing seats in every modern midterm except 1998 and 2002.

The "So What?"

The audience should walk away understanding three things. First, Trump's 2024 "realignment" was not an ideological conversion -- it was a vibe shift among people who weren't paying attention, and vibe shifts are inherently reversible. Second, the mechanism of reversal is brutally simple: prices. Not January 6th, not democratic norms, not any argument progressives have been making -- just the grocery bill. That's humbling for people who think persuasion wins elections, but it's also clarifying. Third, this has massive structural implications for 2026. Trump's unique political talent was mobilizing low-propensity, low-information voters -- but those voters are the most elastic. They came because they were angry at Biden; they're leaving because they're angry at Trump. The MAGA movement's greatest electoral asset is also its most fragile. And that asset almost certainly won't transfer to Rubio or Vance in a future cycle. The floor under Republican power is a lot thinner than either party realizes.

Potential Pitfalls

  • Condescension risk. Talking about "low-information voters" can easily slide into sounding like coastal elites sneering at dumb people. We need to be crystal clear: these aren't stupid people. They're busy people. They're people with three jobs and no time to follow the news. The framework should evoke empathy, not contempt. Frame it as a systemic failure -- these voters were failed by the information environment, not deficient in character.
  • Premature victory lap. Polls in February 2026 don't guarantee midterm outcomes in November. A lot can change. We should acknowledge that Trump has defied polling gravity before, and that low-information voters who've drifted away could drift back if economic conditions shift or if Democrats fail to offer a credible alternative.
  • Oversimplifying the "it's just prices" thesis. While prices are the statistically significant differentiator between the groups, the broader anti-Trump shift among all voters suggests other factors (immigration brutality, institutional chaos, DOGE disruptions) are at play too. We shouldn't reduce the entire picture to grocery bills -- but we should argue that for the marginal voters who swung the 2024 election, prices are the dominant variable.
  • Counterargument: "These voters won't show up in midterms anyway." Low-propensity voters tend to skip midterms, which could mean this shift doesn't translate to seats. We should address this directly -- even if they don't vote in 2026, their disapproval shapes the political environment, and the high-knowledge voters are moving against Trump too (just more slowly).

Source Material Summary

  • Primary source: G. Elliott Morris, "The less voters knew, the more they liked Trump in 2024. Not Anymore," Strength In Numbers/Verasight polling analysis (February 2026). This is the analytical backbone of the episode -- original polling data with detailed crosstabs on high- vs. low-knowledge voters, issue-by-issue approval breakdowns, and the Zaller framework for understanding opinion elasticity. Most relevant source.
  • Supporting source: Lawyers, Guns & Money blog post, "Just enough information voters." Reinforces the Morris analysis with additional commentary on how low-information voter mobilization was Trump's unique skill and skepticism that this transfers to other Republican candidates. Useful for the 2026 implications section.
  • Supporting source: The Bulwark, "How Democrats Can Crush the Midterms" by Mona Charen. Provides broader midterm strategy context -- Trump's -13.7 net approval, the tariff backlash polling (60% blame Trump for rising prices, 65% disapprove of tariffs), and specific Democratic messaging opportunities on immigration, trade, and crime. Useful for the "so what" framing around 2026 and for grounding the low-info voter data in the larger political landscape.
  • Supporting source: Axios/Drudge piece on voters saying Biden was better than Trump. Minor supporting evidence for the broader shift in sentiment, but not central to the episode's thesis.