Not all Trump voters "voted for this"

Trump’s policy agenda is very unpopular, including with many of his supporters. To win the next election, Democrats need to welcome regretful Trump voters back onto their side

Dear members,

Last week, The New York Times ran a great story about residents of a small Missouri town coming to grips with the consequences of Donald Trump’s mass deportation policy. The story focuses on the community’s reaction to the arrest and imminent deportation of Carol Hui, a 20-year resident of Kennett, MO, who worked at the local diner and is the mother of three children.

One of Carol’s friends, Vanessa Cowart — a Trump voter — is quoted in the story as saying, “I voted for Donald Trump, and so did practically everyone here. But no one voted to deport moms.”

The article is full of quotes from people around the small town (10,000 people — 3x bigger than where I grew up!) saying they do not support what is happening to Carol. That includes people who worked with her, people from her church, and farmers from the county surrounding the town. It seems many of Kennett’s Trump voters do not support deporting people like Carol.

And according to polls, most Americans do not support deporting people like Carol either. Deporting undocumented immigrants who are parents of U.S. citizens is underwater by 36 percentage points, for example, and by a net-18-point margin adults oppose deporting immigrants who have not committed crimes other than unlawful entry/overstaying their visa/etc. The only super popular component of Trump’s immigration policy is deporting undocumented immigrants who have committed other violent crimes.

And as YouGov has shown recently, the vast majority of Trump’s agenda more broadly is also unpopular with the average voter:

Trump is particularly unpopular on health care, executive powers, and foreign policy.

The average Trump policy polled by YouGov is 17 points underwater with the public. That is twice as negative as the president’s overall job approval rating. As an empirical matter, most Americans — including millions of Trump voters — seem unsupportive on the whole of Trump’s policy agenda, despite a plurality of them voting for him.

The other way to say that — and the way Carol’s friends put it — is that millions of Americans voted for Trump, but they didn’t vote for this.

. . .

Yet the quote from Cowart has caused quite a stir online, mostly among progressives. On Blue Sky, the average reaction to the quote above is something like “yes, you did” or “counterpoint: you’re complicit in fascism.” Some of these posts are getting tens of thousands of likes:

But there is a big difference between voting for Trump because you wanted Carol (and people like her) to be deported, or because of something else. That’s especially true when partisanship and negative polarization are so high: Perhaps you didn’t even vote for Trump; you just voted against Harris.

In my opinion, chastising these Trump voters for their choice both misunderstands voter psychology and is very unproductive. In this piece, I’m going to explain why I think that.

Democrats need to wrestle with why Trump actually won, and what policies the average American actually supports, so they can build a large and sustainable coalition to win in 2026 and beyond.

---

The rest of this post is funded by paying members of Strength In Numbers. Paid members get access to additional analysis and political data, such as early access to forecasts and new polling averages, and support the public mission of this website. Become a member today and get premium political analysis — and help spread our independent, data-driven journalism:

---

Voting for Trump is not the same thing as voting for his agenda

Cause and effect are important to understanding the 2024 election.

The biggest issue in the 2024 election was not immigration, deportation, abortion, or even Biden’s age — it was the economy. According to Pew, 82% of voters said the economy was “very important” to their vote, including 93% of Trump voters. Trump had a 10-point lead on his ability to handle the economy.

It is fairly easy, then, to imagine a low-information, low-engagement voter who saw the prices of groceries rise under Biden and defaulted to the out-party, more trusted option. Perhaps they were upset about inflation under Biden and didn’t want to elect someone from his administration, for example. Sure, in this case, the guy you voted for ended up deporting Carol, and you have to deal with that being on your conscience for the rest of your life, but you did not want, and importantly, do not support his current policy agenda.

If you are a Democrat, I know it is tempting to do schadenfreude and voter-bashing toward regretful Trump voters, who you see as enabling authoritarianism. But it is very important right now to make the point that people who voted for Trump did not support all of his policy agenda (the vast majority of which is unpopular) for four reasons.

First, internalizing that voting for someone is not an endorsement of their future policy agenda gives people data to fight the worst excesses of Trump’s agenda, especially on trade and immigration. Data can be a very powerful rhetorical tool in catalyzing opposition to unpopular policies.

In Trump’s first administration, for example, polling on his separation of families crossing the border was almost 50 points underwater. That data and related protests convinced the administration to change course.

The alternative to this is assuming that moderate Trump supporters actually support deporting non-criminal parents, which we know (given the data above) to be false. If you assume that, then there is a much smaller landscape of opinion on which to do battle about politics. You have to believe these people are persuadable to move them.

Second, polls can shape opinion and legitimize beliefs. One reason there is popular demand for polls is that people want to know the opinions of their neighbors. It is natural to wonder what others are thinking, especially those close to us, and especially as partisanship rises. And there is a documented bandwagon effect in polling, too; When people see that opinions they held privately are popular publicly, they become more likely to express them.

This is important because the default assumption today is that all Trump voters support deporting legal residents with children, like Carol. If Democrats want to legitimize Republican opposition to those deportations, to give Trump voters the courage to express doubts about his policies, they may find a lot of utility in anchoring their narratives to the data.

Third, asserting all Trump voters support all of his policies ignores the role the information environment plays in manufacturing opinion and raising issue salience. The Democrats are currently at a steep disadvantage in the game of political informational warfare, and have been arguably for decades. Early in the 2024 campaign, for example, it was common for Trump to exaggerate the economic problems facing America and boast about his own record in office. Television news gave him cover on this, allowing him to conflate his past performance with the likely effects of his future policies.

There are many other examples. Fox News ran frequent segments on crime, too, despite the fact that crime was trending down in 2023 and 2024, which likely contributed to crime being rated highly on issue prioritization in 2024. And the effects of slanted information diets are not hard to find. Take the number of people who think millions of dead people are collecting social security as a prime example.

On immigration, lots of voters cast ballots for Trump because they were told that there were millions of violent criminals here illegally, and that he would deport them all. “Dictator on day one,” etc. But many of them did not have the information about the rest of his deportation plans. The point here is that opinions are partially downstream of information, and we should not assume that the information people are getting is perfect, because it’s self-evidenlty not. Extending people that benefit of the doubt allows us to persuade them, or at least attempt to.

And fourth, once you understand that economic backlash can lead people to enable authoritarianism, you appreciate the stakes of keeping inflation low and campaigning on popular economic messages. One of the bigger strategic errors of Biden’s surrogates, and especially the loud center-left economics statistics truthers on Twitter, was to insist that the economy was great in 2024, so voters had no business punishing Biden for inflation in 2022-2023. This caused them to think Biden’ didn’t care about them.

The evidence from polling suggests people punished Biden for inflation despite the rate of increase coming down in 2024. The right way to look at inflation data wasn’t annual change in the CPI, but that the cumulative increase in prices over Biden’s term was very high, especially compared to the last 50 years.

Right-wing figures throughout history have exploited economic stress to paper over their agenda and then shift policy dramatically to the right. Democrats have to campaign appropriately on economics to prevent that. Messaging about democracy and civil rights did not work as well as Democrats hoped in 2024 in part because voters opposed to Biden on inflation were already conditioned to vote against him.

The most popular figures in the Democratic Party today all use “economic populist” messaging that highlights how the excesses of capitalism hurt everyday people, and how they have become victims of a “rigged” system. Trump deploys similar rhetoric on economics. That may be the key to the party recapturing his voters.

Democrats need regretful Trump voters to win the presidency in 2028 and the Senate in 2026

The counterargument to this piece is that voting is an act of collective, downstream responsibility; When someone votes for a candidate, they are not just endorsing a single issue like the economy, they are supporting the entire platform and are responsible for the consequences that come with it.

I won't deny that; Trump's hardline immigration policies, including the deportation of long-term residents and parents, were not hidden from public view; they were central to his campaign and widely discussed in the media.

But it's also true that many voters do not pay as close attention to politics as Vanessa Cowart's critics on Bluesky. And most voters are also making difficult trade-offs, often prioritizing the issues that most directly affect their lives. If we want to change policy and protect vulnerable communities in the long term, we need to welcome those who are willing to reconsider their choices and join the fight for humane policies.

Accountability is important, but it should be paired with empathy and a path forward.

The reality is that Democrats face a straightforward math problem in the next two elections. In 2024, there were more Trump voters than Harris voters. Democrats have to win some of them back to win the next election.

And there are especially more Trump than Harris voters in North Carolina, the state that elects the 50th Senate seat (and which Democrats would need to win to control the majority in the chamber). According to my analysis of the 2024 election, Democrats need to win the effective popular vote for the Senate by 6 points over the 2024, 2026, and 2028 cycles to win the 50th vote in the Senate. That still leaves them one shy of a majority as long as there is a Republican Vice President (JD Vance is the current tie-breaker).

Whatever the best strategy is for Democrats to win back the voters they need for electoral majorities, it likely involves welcoming into the Big Tent the engaged moderates who voted for Trump because they were pissed about inflation or crime and are now regretful of their choice. As an effort at persuasion, it’s probably best not to berate them on social media.

And it's also a matter of fact that many of them did not, in fact, vote "for" this.

---

Thank you for being a supporting member of Strength In Numbers. Please share this post with your friends and on social media to increase the number of people who see our data-driven, independent journalism.

Be well,

Elliott