Trump’s approval is terrible. Here’s the latest Pew polling on that. Apparently, starting a war for no reason while threatening to start up the draft has have harmed his approval among younger Trump voters. Ditto for ICE roundups and hispanics.
Trump can be less popular than the clap and hemorrhoids, yet we know that a lot of those voters who are upset with him will still vote Republican. The question is who’s “gettable”. There’s a lot of motivated reasoning around this question, because a bunch of money gets funneled into Democratic consultants who will take those Trump disapproval numbers and try to convince us to adopt the Republican positions that donors seem to love.
G. Elliott Morris has a piece out that tries to separate the gettables from the never-gonna-get-its. He starts with what he calls “cross-pressured voters” — people who disapprove of Trump but don’t say they’ll vote for Democrats on the generic ballot. His polling shows a 12.7 point gap between Trump net disapproval and the Democratic generic ballot lead, which is 8.1 points in his polling. Here’s how he segments those voters:
Pool 1 — real holdouts: 3.4% of registered voters. This group did not vote for Trump in 2024 and has no other Republican markers at all. These voters are the most likely to vote Democratic in 2026.
Pool 2 — Trump’s borrowed coalition: 0.8% of RVs. These are Trump disapprovers whose only Republican marker is casting a Trump vote in 2024; otherwise, they are independent and moderate or liberal. They are more Republican-leaning than the average voter, but not reliable members of the Republican coalition.
Pool 3 — Republican-aligned Trump disapprovers: 6.8% of registered voters. These are cross-pressured voters with current Republican markers: they identify with or lean toward the GOP and/or place themselves on the conservative side of the ideology scale. Many, but not all, also voted for Trump in 2024.
Pool 3 is a loss — they’re probably never going to vote for a Democrat. They’ll “come home”, especially in the 2028 general, even if they sit out 2026. When Morris adds in Pool 1 and Pool 2 — the “gettables” — into his generic ballot calculation, he ends up with D+13.3. That’s good!
Morris digs into the largest group of gettables, Pool 1:
The first thing to know is that these voters have not really chosen against Democrats. For the most part, they have just not chosen at all. On the generic ballot, 83% of these voters say they “don’t know” how they’d vote, and just 17% say they’d vote for the Republican in their local district.
It is also, very clearly, a turnout story. Nearly 40% of Pool 1 did not vote in 2024. Another 26.1% voted for a third-party candidate. Only 34.0% voted for Harris, and none voted for Trump (because of the way we defined this group). The non-voter rate in this group is roughly four times that of firm Democratic voters.
And their demographic profile does not look much like the familiar suburban-moderate caricature. Pool 1 is 42.8% with no education beyond high school, 37.3% under $50,000 in household income, 24.3% Hispanic, 18.3% Black, and 26.0% under age 30, compared with just 8.1% over 65. More than three-quarters — 76.8% — are pure independents on the five-point party ID scale, and 57.2% are women.
In short, these voters are working-class, multiracial, heavily independent, and significantly politically disengaged. These are not voters who have carefully weighed the parties and decided they prefer Republicans. They are voters who are barely engaged. As I argued in “The Strategist’s Fallacy”, many voters do not think about politics the way analysts do. Pool 1 is a perfect example of this: they are low-information voters who have not fully sorted themselves onto either ideological camp. The biggest challenge in winning them is engagement, not persuasion.
So what do these voters care about? Prices, jobs, healthcare and social programs, not crime or immigration — look at the blue chart below.
As has been true for what seems like forever, there are a group of voters who are generally sympathetic to Democrats’ issues, they are unengaged, and we need to figure out how to connect with them and get them to vote. If they vote, they’ll probably vote for our candidates. I guess this doesn’t appeal to the large number of consultants and candidates who think that the only way to win is to try to court Republicans who will never vote for Democrats, but judging from the vast number of works of art about unrequited love, human nature just works that way.
