Less than one year into President Donald Trump’s second term, we finally have solid evidence that the coalition that carried him to victory one year ago today is unraveling. The slate of Republican losses — and the magnitude of Democratic wins in New Jersey and Virginia, especially — suggest not just that the Democratic backlash to Trump has finally arrived, but that a key part of Trump’s majority — Latino voters —might not actually be loyal Republican voters after all.
In no place was that more clear last night than in New Jersey, where exit polls suggest that Democratic Gov.-elect Mikie Sherrill won nearly 70 percent of Latino voters, compared to the 31 percent who sided with Republican Jack Ciattarelli. That’s a huge reversal: Trump got within 6 points of winning the state in 2024, largely with the help of Black and Latino voters who swung for him across the state. Now, in the precincts, municipalities, and counties with large Latino populations that swung toward Trump last year, particularly in North Jersey, Latino voters seemed to have turned out at higher-than-expected rates, and mostly returned to the Democratic side.
But that still leaves many more questions: Are these disillusioned Trumpers turning back to the Democrats? Or are they just going to sit out more elections? To get a better sense of what we can — and can’t — discern from Tuesday’s results, I turned to one of the leading pollsters and experts on Latino politics in the US: Carlos Odio, the co-founder of the Democratic-leaning research firm Equis, for some answers. This interview has been edited for length and clarity.
Heading into the election, what was your main question concerning Latino voters?
My question going in was whether this race was going to look more like 2021, the last governor’s race, [with lower but steady Latino support for Democrats] or 2024, when Trump made these huge gains. Ultimately it does look like 2021. And I think what that suggests is, going into 2026, Latino support for Democrats is more likely to be equivalent to what it was in 2021 than what it was in 2024.
Can you unpack that for me? What is the deeper significance of Latino voting patterns looking more like 2021 than 2024?
So if [Tuesday’s results] looked more like 2024, then what you would say is, Well, these Trump gains weren’t a one-off. There is a durable shift that Republicans are holding onto and might even build on. Had that happened, it would’ve suggested that Latinos were not going to be part of any “blue wave,” if such a thing were to even happen.
If the results were more like 2021, however, then the takeaway would’ve been more like, Well, we’re back to a more normal off-year election, something like that middle [level of Latino support for Democrats] that we saw coming out of 2020.
So, from our vantage today, It’s almost like you’re turning back the clock to how the Latino vote looked like before the Biden presidency happened.
Which, it should be noted, most Democrats would still consider not great. Not devastating, like in 2024, but still not great.
That’s exactly right. We are still in the Trump era.
I remember I was working in Florida in the 2018 cycle when Florida Hispanics obviously sat out the national “blue wave.” I saw how Latinos in Florida behaved differently, shall we say. The resistance was mobilizing people, but Latino voters didn’t feel caught up in that wave or in that moment.
So that was a question for me going into this election: How many Latinos were basically going to be like, I just don’t feel mobilized against Donald Trump. I am so frustrated with both parties that I’m either going to sit this out or I’m going to vote on other considerations; but I’m not catching the anti-Trump fever.
How much can we reasonably extrapolate from this election and apply to our expectations for the midterm elections or even the next presidential election?
Can you draw conclusions from last night in New Jersey to what will happen in Texas? I don’t think that’s quite right, but these results are certainly a stronger data point than anything else we’ve had so far this year. When you look at the trend lines back to 2016, it’s true off-years and presidential years tell a different story, so what happened in 2025 isn’t necessarily going to tell you very much about what’s going to happen in 2028. But I think it is a pretty good indicator of what to expect for 2026, or at least it sets a better benchmark than 2024 for what to expect in 2026.
How much of the results were persuasion — Latinos who voted for Trump in 2024 voting for Democrats in 2025 — versus pro-Trump Latinos simply sitting this off-year election out?
We’ve got to tear down the wall implicit in your question that separates these two categories of voters. The truth is, support and turnout point in a similar direction. When people have somebody to support, they’re more likely to vote. When people don’t like their choices, they’re less likely to vote. And so what we’ve been seeing in our polling is 11 percent of Latino Trump voters saying they’d vote for a Democrat in 2026.
That actually ends up being a lot. That’s a meaningful chunk. And at the same time, you can look at Trump’s strengths being among more of an irregular-voting Latino — the kinds of people who didn’t vote in 2022, who didn’t turn out — who might not vote in 2026. But also I’d say that, even among these voters, it’s not so simple. I think there’s another aspect there, which is that low-propensity voters are the swingiest, the most sensitive to the environment. And what we’re seeing right now is that the Latinos who vote irregularly are the most disappointed in Trump.
