How much does Biden’s support for Israel hurt him among young voters?

Even this summer, a poll with Donald J. Trump as the leader among young voters would have been dazzling.

Now it’s becoming more widely known — and our new New York Times/Siena College national survey released Tuesday morning is no exception.

For the first time, Mr. Trump leads President Biden among young voters in a national Times/Siena survey, 49 percent to 43 percent. It’s enough to give him a slim 46-44 lead among registered voters overall.

Normally it’s not worth dwelling too much on a sub-sample from a single poll, but this basic story about young voters is present in almost every major survey at this point. Our own state polls in the fall showed something similar, with Mr. Biden leading by one point among those 18 to 29. Both numbers are a big shift from Mr. Biden’s 21-point lead in our last poll before the midterms or his 10-point lead in our last national poll in July.

And there is a plausible explanation for the shift in recent months: Israel.

As my colleagues Jonathan Weisman, Ruth Igielnik, and Alyce McFadden report, young voters in the survey had an extremely negative view of Israel’s recent behavior: They overwhelmingly say Israel is not doing enough to prevent civilian casualties in Gaza, and believe that Israel is not interested. in peace, and believe that Israel should end its military campaign even if it means not eliminating Hamas.

You might think that young voters with these progressive or even left-wing views are among the group most likely to stick with Biden. At least for the time being, that is not the case. Young Biden ’20 voters with anti-Israel views are the most likely to report switching to Mr. Trump.

Overall, Trump wins 21 percent of young Biden ’20 voters who sympathize more with Palestinians than Israel, while winning 12 percent of other young Biden ’20 voters. In an even more striking sign of defections among his own supporters, Mr. Biden has only a 64-24 lead among young Biden ’20 voters who say Israel is deliberately killing civilians, compared with an 84-8 lead among the Biden ’20 voters. voters who don’t think Israel deliberately kills civilians.

It’s possible that the kind of young voters who were against Israel were already against Biden before the war. That cannot be ruled out. But it’s still evidence that opposition to the war itself is likely contributing to Biden’s unusual weakness among young voters.

Here are some other findings from the survey:

Biden ahead among likely voters?

Even though he trails among registered voters, Mr. Biden actually leads Mr. Trump in our first measure of the likely 2024 electorate, 47 percent to 45 percent.

If you read this newsletter carefully, this may not have come out of nowhere. Our polls consistently show Biden doing better among highly regular and engaged voters — especially among those who voted in the last midterm elections. In those polls, Republican voters were the ones who voted in 2020, but not in 2022. It helps explain why Democrats continue to do so well in low-turnout special elections even as they struggle in polls of registered voters or adults .

But in this particular poll, the division is not just between midterm and midterm voters. It’s between people who voted in the 2020 general election and those who didn’t. Mr. Biden leads by six points among voters who participated in the 2020 election, while Mr. Trump has an overwhelming 22-point lead among those who did not vote in 2020. Needless to say, non-voters in 2020 are more likely to vote in 2024, which is why we show Mr. Biden ahead among likely voters.

It’s an intriguing pattern, but there is good reason for caution.

First, our previous polls have not shown anything this extreme, including our battlefield polls conducted eight weeks ago. That doesn’t mean it’s wrong, but our 2020 sample of abstainers includes only 296 respondents—a sample too small for serious conclusions.

For another, people who voted in 2020 reported supporting Mr. Biden over Mr. Trump by 10 points in the 2020 election, 51 percent to 41 percent. In reality, Mr. Biden won by 4.5 points.

There’s a good reason why respondents in our survey were less likely to have expressed support for Mr. Trump: we concluded the substantive portion of the survey with a series of questions about Mr. Trump’s upcoming legal battle, including the question of whether he has committed crimes, whether he will be convicted, whether he should go to prison, and so on. Then at the very end of the survey we asked them how they voted in 2020.

It’s possible that these questions about Trump’s legal troubles made his supporters less likely to concede to supporting him in the 2020 election. Registered Republicans with a 2020 record were three times as likely as Democrats to refuse to vote for us. tell who they supported in the last presidential election. But it’s also possible that our sample actually just contains too many Biden ’20 voters relative to non-voters, producing a lopsided shift in his direction among likely voters.

The underlying data still looks largely normal.

Every time I see something that looks like a crazy result — like Mr. Trump taking the lead among young voters or a nearly 30-point gap between voters and non-voters in 2020 — I think I’m going to look deeper into the data and start to see the signs that something is wrong.

I haven’t seen it yet.

In fact, this survey has a more Democratic sample of young people based on party registration than in the past, but a much more Trump-friendly result.

It’s a similar story for 2020 nonvoters. They may support Trump by a wide margin, but 27 percent are registered as Democrats, compared to 17 percent as Republicans. Mr. Trump nevertheless leads among them as Mr. Biden has only a 49-34 lead among those registered Democrats who did not turn out in the 2020 election. He has an 83-8 lead among registered Democrats who did vote.

A lead of just 49-34 for Mr. Biden among Democratic nonvoters sounds pretty far-fetched, but it is at least easy to imagine why these types of Democrats are less likely to support Mr. Biden. If you’re a Democrat who didn’t vote in 2020, you probably don’t oppose Mr. Trump as strongly and passionately as those who did show up. Non-voters also tend to be young, non-white, less educated and low-income — all groups Biden has struggled with. They also tend to be less partisan and less ideological, and therefore potentially less loyal to the party.

But for now it’s just one relatively small data point. And curiously, it’s a data point that we may never get the chance to validate. After all, non-voters don’t vote. In all likelihood, people with a good voting record will play a big role in the election, and that is good news for Mr. Biden at least in this poll.