Why Democrats can't win more Trump disapprovers

Democrats are hitting a ceiling. Could crime show us why?

Just because voters are souring on Trump doesn’t mean they’re eager for a return to Democratic governance. (Photo by Lindsey Nicholson/UCG/Universal Images Group via Getty Images)

We’re taking The Argument to San Francisco! On May 13, Kelsey Piper and Jerusalem Demsas are debating a question that feels unavoidable right now: Is AI actually changing how science gets done, or are we in the middle of a very expensive illusion? Jerusalem is bullish; Kelsey is skeptical.

And you won’t just be watching. You’ll get to join in on the argument, too.

Join us May 13 at The Chapel from 7 to 10 p.m. Come argue with us! RSVP here

---

Welcome back to The Argument’s poll series, where we survey Americans on the issues everyone’s fighting about. Our full crosstabs are available below the paywall at the end of this post. Our last surveys have asked about the economy, gender issues, immigration, education and parenting, the lingering politics of COVID-19, immigration, AI, and free speech. The Argument’s full methodology can be read here.

---

Over the last few months, I’ve observed a trend: Although Donald Trump’s approval rating has been declining linearly with time, Democrats have not gained nearly the same amount of ground on the generic ballot.

In fact, they’ve seemingly hit a ceiling for now, despite Trump’s approval careening into the abyss.

The Democratic stagnation is a bit surprising, especially considering how badly Trump’s approval has degraded over the last few months. There is no shortage of voters disenchanted with the president; in fact, our latest national survey of registered voters at The Argument clocked his approval at 39%, with a whopping 60% disapproving.

And yet, Democratic vote share hasn’t moved at all. Today, in our new survey of 1,516 registered voters fielded from April 20 to 23, 2026, Democrats led by “just” six points among all registered voters. That’s the same margin they led by in February, despite Trump’s net approval being seven percentage points worse now. (The full results from our poll are available for paying subscribers at the end of this piece).

Why is it that Democrats are failing to convert so many people who hate Trump, especially given that he is the most unpopular president in a generation?

Here’s an answer: many voters really have sincerely held policy beliefs and vote accordingly. And a good chunk of the ones who dislike Donald Trump still have serious and deeply rooted disagreements with the Democratic Party on policy.

This month’s poll focused on public safety and criminal justice. In the same survey where Trump’s approval sat at -22¹ and Democrats led the generic ballot by six, we asked voters who they trusted more on crime. Republicans led 38% to 33%. And among the 9% of the electorate who disapproved of Trump but also leaned toward voting Republican in 2026,² Republicans led on crime 47% to 1%. That’s a whole lot of potential voters who seem to have no interest in the Democrats handling public safety.³

Democrats still have tangible policy misalignments with many voters who dislike Trump. It’s worth going a bit deeper into what some of these divergences actually are.

A plurality of voters believe that neither society, courts, nor the criminal justice system as a whole deal with criminals harshly enough. That’s a pretty stark departure from the more conciliatory and gentle views on criminal justice that are typically associated with the Democratic Party.⁴

The death penalty is another example, as a large majority of Americans remain in favor of it, especially for serious crimes, whereas the Democratic Party called for abolishing the death penalty in both 2016 and 2020.

And a plurality of voters still believe that people living in the country illegally commit violent crimes at a significantly higher rate than native-born citizens do, despite a large body of evidence contradicting this perception. It is easy to see how this perception significantly hurts Democrats with the electorate on crime, especially given the Democratic Party’s considerably more lenient method of dealing with undocumented immigrants (and given the surge of migrants at the border under Biden’s presidency).

As I’ve written before, Democrats’ immigration problem is largely downstream of their crime problem.

Another place where Democrats may have suffered a longer-term penalty in public perception is shoplifting. Despite the national vibe shift, it’s unlikely that voters have forgotten high-profile progressive prosecutors who demanded leniency for petty theft in places like San Francisco, Los Angeles, Boston, and Chicago, which often meant that shoplifting went unpunished.

When juxtaposed next to borderline-comical videos of people freely ransacking retail stores, this inevitably led to backlash, especially as shoplifting spiked and prosecutions cratered in blue localities. At the time, reform-minded academics justified this by saying that such policies decreased the likelihood of people having future involvement with the criminal justice system.

Whether or not this claim is strictly true, it ignored the public’s frustration with leniency for criminals. In fact, in our survey, 71% of voters actually want prison time for repeat offenders (hello, Jia Tolentino⁵), including 16% who actually wanted jail sentences for first-time offenders.

As we saw in our polling immediately after the Minnesota shootings, voters really hate disorder. And our poll on criminal justice shows that they want strict penalties for causing it.

Have the Democrats fully moved away from their unpopular views? I’m not sure. Let’s take the death penalty as an example. The party’s 2024 platform actually doesn’t mention the death penalty at all, unlike 2016 or 2020. But instead of repudiating its old stances, the party simply … doesn’t mention it. At all. There’s nothing in there, one way or another, about the death penalty.

I’m not here to argue that large numbers of people are basing their votes on the death penalty. Rather, I think this is symptomatic of the Democratic problems at large on things like criminal justice, LGBTQ rights, and immigration. Democrats are not actually repudiating the stances that they originally took; they are instead simply aiming to avoid talking about the issue in hopes that people will just forget.

Perhaps that’s true in the long run, but what this poll shows is that despite declining crime, and despite ongoing efforts by big city mayors to tackle public safety head on, the plurality of voters still trust Republicans on the issue over Democrats. And Democrats haven’t really given them a reason to change their minds.

There are places where the public agrees more with the progressive lens, especially on the structural explanations for crime. For instance, the proposal that gets the most consensus on its potential for crime reduction is actually increasing funding for education and job training.

Political parties don’t only exist to parrot public opinion. In some cases, the voters are actually wrong about what would bring down crime, but most of their expressed opinions in this poll are actually sound, even if their perceptions of crime are off base.

Race for the Senate

There’s one piece of comfort for liberals: Many Trump-disapproving Republican voters may simply refuse to show up in 2026.

Mathematically, this is not as efficient or as powerful as a voter flipping from Republican to Democratic (convincing someone to switch parties nets you two votes on margin, while getting them to stay home nets you just one). But it still helps buffer the Democratic edge among voters who expect to actually turn out this November.

The Democratic lead among voters who said they’re actually likely to vote in 2026 sits at 10 percentage points, which would imply the biggest blue wave since at least the 1980s.

Strong midterms can be deceptive. Many of the voters who sit out 2026 will show up in 2028 (low-propensity voters tend to show up more in presidential years), and if Democrats are still fundamentally mismatched on important issues like crime, that risks a much greater chance of being reexposed.

METHODOLOGY

Our poll was fielded between April 21 and 23, 2026, and surveyed 1,516 registered voters across the nation. The sample was weighted to be representative of the universe of registered voters in the United States by race; age; gender; education; census region; race by gender; age by gender; race by education; modeled presidential partisanship by age, education and gender; and 2024 vote choice. The margin of sampling error was plus or minus 2.5 percentage points. The design effect of the survey, which measures the loss of statistical precision due to weighting and design, was 1.17. Accounting for the design effect, the full margin of error was plus or minus 2.7 percentage points.

This survey was designed and weighted by The Argument. Data collection was performed by Verasight, and voter file data and weighting targets were obtained from Catalist. A full methodology statement is available on the last page of the survey PDF. A detailed explanation on how our surveys work is available here.

Thanks to Josh Kalla, Guy Molyneaux, and Charlotte Swasey for reviewing this month’s polling language. As part of The Argument’s polling advisory committee, these experts have provided comments to ensure the rigor and objectivity of this polling project. They are not responsible for the ultimate language choices made by The Argument, and all errors are our own.

Recommended reading: