Its been a bad stretch of polling for President Donald Trump.

In recent weeks, a string of new polls has found Trump losing ground with key constituencies, especially the young, non-white and low-propensity voters who swung decisively in his direction in 2024. The uptick in support for Trump among those non-traditional Republican voters helped fuel chatter of an enduring “realignment” in the American electorate — but the durability of that realignment is now coming into doubt with those same groups cooling on Trump. Surveying the findings of the most recent New York Times-Siena poll, polling analyst Nate Cohn bluntly declared that “the second Trump coalition has unraveled.”

Is it time to touch up the obituaries for the Trumpian realignment? To find out, I spoke with conservative pollster and strategist Patrick Ruffini, whose 2024 book “Party of the People” was widely credited with predicting the contours of Trump’s electoral realignment.

Ruffini cautioned against prematurely eulogizing the GOP’s new coalition, noting that the erosion of support has so far not extended to the constituencies that have served as the primary drivers of the Trumpian realignment — particularly white working-class voters and working-class Latinos and Asian Americans. But he also acknowledged that the findings of the recent polls should raise alarms for Republicans ahead of 2026 and especially 2028.

His advice to Trump for reversing the trend: a relentless focus on “affordability,” which the White House has so far struggled to muster, and which remains the key issue dragging down the president.

“I think that is undeniable,” he said. “It’s the number one issue among the swing voter electorate.”

This conversation has been edited for length and clarity.

Based on your own polling, do you agree that “the second Trump coalition has unraveled?”

It really depends on how you define the Trump coalition. The coalition that has really reshaped American politics over the last decade has been a coalition that saw voters who are aligned with a more populist view of America come into the Republican Party — in many cases, after voting for Barack Obama twice. Those shifts have proven to be pretty durable, especially among white working-class voters but also among conservative Hispanic voters and conservative Asian American voters.

You have another group of voters who is younger and disconnected from politics — a group that had been really one of the core groups for Barack Obama and the Democrats back in the 2010s. They didn’t always vote, but there was really no hope or prospect for Republicans winning that group or being very competitive with that group. That happens for the first time in 2024, when that specific combination of young, minority, male voters really comes into play in a big way. But that shift right has proven to be a little bit less durable — and maybe a lot less durable — because of the nature of who those voters are. They’re not really connected to one political party, and they’re inherently non-partisan.

So what you’re seeing is less of a shift among people who reliably vote in midterms, and what we are seeing is more of a shift among those infrequent voters. The question then becomes are these voters going to show up in 2026?

How big of a problem is it for Republicans if they don’t? How alarmed should Republicans be by the current trends?

I think they’re right to focus on affordability. You’ve seen that as an intentional effort by the White House, including what seems like embracing some Democratic policy proposals that also are in some ways an end-run around traditional Republican and conservative economics — things like a 10 percent cap on credit card interest.

What’s the evidence that cost of living is the thing that’s primarily eroding Republican support among that group of voters you described?

I think that is undeniable. It’s the number one issue among the swing voter electorate. However you want to define the swing voter electorate in 2024, cost of living was far and away the number one issue among the Biden-to-Trump voters in 2024. It is still the number one issue. And that’s because of demographically who they are. The profile of the voter who swung in ‘24 was not just minority, but young, low-income, who tends to be less college-educated, less married and more exposed to affordability concerns.

So I think that’s obviously their north star right now. The core Democratic voter is concerned about the erosion of norms and democracy. The core Republican voter is concerned about immigration and border security. But this swing vote is very, very much concerned about the cost of living.

Is there any evidence that things like Trump’s immigration crackdown or his foreign policy adventurism are contributing at all to the erosion of support among this group?

I have to laugh at the idea of foreign policy being decisive for a large segment of voters. I think you could probably say that, to the extent that Trump had some non-intervention rhetoric, there might be some backlash among some of the podcast bros, or among the Tucker Carlson universe. But that is practically a non-entity when it comes to the actual electorate and especially this group that is floating between the two political parties. Maybe there’s a dissident faction on the right that is particularly focused on this, but what really matters is this cost-of-living issue, which people don’t view as having been solved by Trump coming into office. The White House would say — and Vance said recently — that it takes a while to turn the Titanic around.

Which is not the most reassuring metaphor, but sure.

Exactly, but nonetheless. I think a lot of these things are very interesting bait for media, but they are not necessarily what is really driving the voters who are disconnected from these narratives.

What about his immigration agenda? Does that seem to be having any specific effect?

I do think there’s probably some aspect of this that might be challenging with Latinos, but I think it’s very easy to fall back into the 2010 pattern of saying Latino voters are inordinately primarily focused on immigration, which has proven incorrect time after time after time. So, yes, I would say the ICE actions are probably a bit negative, but I think Latino voters primarily share the same concerns as other voters in the electorate. They’re primarily focused on cost of living, jobs and health care.

How would Trump’s first year in office have looked different if he had been really laser-focused on consolidating the gains that Republicans saw among these voters in 2024? What would he have done that he didn’t do, and what shouldn’t he have done that he did do?

I would first concede that the focus on affordability needed to be, like, a Day 1 concern. I will also concede how hard it is to move this group that is very, very disaffected from traditional politics and doesn’t trust or believe the promises made by politicians — even one as seemingly authentic as Trump. I go back to 2018. While in some ways you would kill for the economic perceptions that you had in 2018, that didn’t seem to help them much in the midterms.

The other problem with a laser focus on affordability on Day 1 is that I don’t think it clearly aligns with what the policy demanders on the right are actually asking for. If you ask, “What is MAGA economic policy?”, for many, MAGA economic policy is tariffs — and in many ways, tariffs run up against an impulse to do something about affordability. Now, to date, we haven’t really seen that actually play out. We haven’t really seen an increase in the inflation rate, which is good. But there’s an opportunity cost to focusing on certain issues over this focus on affordability.

I think the challenge is that I don’t think either party has a pre-baked agenda that is all about reducing costs. They certainly had a pre-baked agenda around immigration, and they do have a pre-baked agenda around tariffs.

What else has stopped the administration from effectively consolidating this part of the 2024 coalition?

It’s a very hard-to-reach group. In 2024, Trump’s team had the insight to really put him front-and-center in these non-political arenas, whether it was going to UFC matches or appearing on Joe Rogan. I think it’s very easy for any administration to come into office and pivot towards the policy demanders on the right, and I think that we’ve seen a pivot in that direction, at least on the policy. So I would say they should be doing more of that 2024 strategy of actually going into spaces where non-political voters live and talking to them.

Is it possible to turn negative perception around among this group? Or is it a one-way ratchet, where once you’ve lost their support, it’s very hard to get it back?

I don’t think it’s impossible. We are seeing some improvement in the economic perception numbers, but we also saw how hard it is to sustain that. I think the mindset of the average voter is just that they’re in a far different place post-Covid than they were pre-Covid. There’s just been a huge negative bias in the economy since Covid, so I think any thought that, “Oh, it would be easy that Trump gets elected, and that’s going to be the thing that restores optimism” was wrong. I think he’s taken really decisive action, and he has solved a lot of problems, but the big nut to crack is, How do you break people out of this post-Covid economic pessimism?

The more critical case that could be made against Trump’s approach to economic policy is not just that he’s failed to address the cost-of-living crisis, but that he’s actively done things that run contrary to any stated vision of economic populism. The tax cuts are the major one, which included some populist components tacked on, but which was essentially a massively regressive tax cut. Do you think that has contributed to the sour feeling among this cohort at all?

I think we know very clearly when red lines are crossed and when different policies really get voters writ large to sit up and take notice. For instance, it was only when you had SNAP benefits really being cut off that Congress had any impetus to actually solve the shutdown. I don’t think people are quite as tuned in to the distributional effects of tax policy. The White House would say that there were very popular parts of this proposal, like the Trump accounts and no tax on tips, that didn’t get coverage — and our polling has shown that people have barely actually heard about those things compared to some of the Democratic lines of attack.

So I think that the tax policy debate is relatively overrated, because it simply doesn’t matter as much to voters as much as the cultural issues or the general sense that life is not as affordable as it was.

Assuming these trends continue and this cohort of sort of young, low-propensity voters continues to shift away from Trump, what does the picture look like for Republicans in 2026 and 2028?

I would say 2026 is perhaps a false indicator. In the midterms, you’re really talking about an electorate that is going to be much older, much whiter, much more college-educated. I think you really have to have a presidential campaign to test how these voters are going to behave.

And presidential campaigns are also a choice between Republicans and Democrats. I think certainly Republicans would want to make it into a Republican-versus-Democrat choice, because polling is very clear that voters do not trust the Democrats either on these issues. It’s clear that a lot of these voters have actually moved away from the Democratic Party — they just haven’t necessarily moved into the Republican Party.

Thinking big picture, does this erosion of support change or alter your view of the “realignment” in any respect?

I’ve always said that we are headed towards a future where these groups are up for grabs, and whichever party captures them has the advantage. That’s different from the politics of the Obama era, where we were talking about an emerging Democratic majority driven by a generational shift and by the rise of non-white voters in the electorate.

The most recent New York Times poll has Democrats ahead among Latino voters by 16 points, which is certainly different than 2024, when Trump lost them by just single digits, but that is a far cry from where we were in 2016 and 2018. So I think in many respects, that version of it is coming true. But if 2024 was a best-case scenario for the right, and 2026 is a worst-case scenario, we really have to wait till 2028 to see where this all shakes out.