Trump’s conviction may be hurting him — but it’s early

There’s a scene in the movie “Airplane!” where, having learned the breaking news of an impending airline disaster, a cadre of reporters run into a bank of phone booths so fast, it falls over. That’s what I imagine the nation’s pollsters looked like this weekend when they rushed to ask Americans how they were reacting to former President Donald Trump’s conviction on 34 felony charges last week.

Four pollsters have already conducted national polls of the 2024 presidential election entirely since Trump was found guilty, and a handful more have asked about the verdict but not the election explicitly. They show that Americans are taking the verdict seriously, and it may be giving President Joe Biden a small boost in support. However, it’s still too early to draw any definitive conclusions.

First off, most Americans agree with the verdict. An ABC News/Ipsos poll from over the weekend found that 50 percent of adults thought the verdict was correct, and a YouGov/CBS News poll found that 57 percent of adults thought so. YouGov/CBS News also found that a bare majority — 51 percent — of adults considered Trump unfit for the presidency now that he has been convicted.

However, the verdict also doesn’t seem to have changed many people’s minds about the case. For example, 51 percent of adults told ABC News/Ipsos that Trump intentionally did something illegal with regard to this case — virtually the same as the 53 percent who said so in ABC News/Ipsos’s April 2023 poll, just after Trump’s indictment. And YouGov/CBS News found that nine in 10 Americans who thought Trump was guilty before the verdict agreed with the verdict, while nine in 10 Americans who thought Trump wasn’t guilty disagreed with it.

That said, two other polls found that the verdict has made a small but significant share of potential Trump supporters less likely to vote for him. According to Ipsos/Reuters, 10 percent of Republican registered voters said they were less likely to support Trump after the conviction; HarrisX/Forbes put that number at 11 percent. Similarly, 25 percent of independents said they were less likely to vote for Trump in the Ipsos/Reuters poll, and 28 percent of independents said so in the HarrisX/Forbes poll.

As I wrote on Thursday, though, you should take more-or-less-likely polls with a grain of salt; some of those people who say they are less likely to vote for Trump may not have been very likely to vote for him in the first place, and even among supporters, “less likely to vote for” does not mean “definitely will not vote for.”

So let’s get to the money numbers: whom voters say they will support for president in the wake of Trump’s conviction. On average, the most recent national polls from the four pollsters who’ve polled since the verdict show a tied race.* That represents a 1-point average swing toward Biden from those pollsters’ pre-conviction surveys.

Keep in mind, when looking at how polls may or may not have shifted over time, it’s important to compare only surveys from the same pollster. Individual pollsters make different methodological decisions that can impact what their polls say; therefore, if one pollster shows Trump up 3 points one week and another pollster shows Biden up 3 points the next week, that is just as likely to be due to differences in methodology as it is an actual shift in the race.

Interestingly, at least according to these surveys, the shift toward Biden isn’t because Trump is losing support; it’s because Biden is gaining it. On average, Biden’s support went from 42 percent in these four pollsters’ pre-conviction polls to 43 percent after it. By contrast, Trump’s support stayed flat at 43 percent. This is the opposite of what hypothetical polls asking “how would you vote if Trump is convicted” said before last week, which was that Trump would lose support but Biden wouldn’t gain it. This just underscores how unpredictable the public reaction to Trump’s conviction was — and is (it’s still early and we don’t have that much data yet).

Now for the caveats, and there are several. Although the fact that three out of the four pollsters showed a shift toward Biden makes us more confident that this is, in fact, real movement, the shifts in both the Ipsos/Reuters and Morning Consult polls were within the margin of error — meaning they could have just been due to random chance. That said, Echelon Insights did something useful: It surveyed the same voters both before and after the conviction, removing the possibility that its 2-point shift toward Biden was due to getting a slightly more Democratic sample the second time around.

It’s also possible that these shifts are an illusion caused by something called (deep breath) differential partisan nonresponse bias. Basically, in the wake of bad news for Republicans and/or good news for Democrats, Republicans may be less excited about responding to surveys and Democrats may be more excited to — which can lead to polling numbers that are a bit better for Democrats than the true state of public opinion. (After the news fades, this excitement factor often does too, leading the polls to stabilize.) It’s easy to imagine this bias being at play right after the Republican candidate for president is convicted of a crime.

Even if Biden’s improvement is real, though, another thing to bear in mind is that these are just four polls. Over the coming days and weeks, we’ll undoubtedly get even more data showing how Americans are (or aren’t) changing their minds after the conviction, and it may not say the same thing that these early surveys do. Keep an eye on our polling averages for the best idea of whether and how much the race has shifted; our averages are trained only to move in response to new, different polling numbers once they have seen enough of them to be convinced that the shift is real. And we’ll publish additional stories if and when new polls provide more perspective.

Finally, of course, bear in mind that Americans are still digesting the news of Trump’s conviction. It’s possible that not enough time has elapsed for its full impact to be reflected in the polls. Americans may also change their minds about how much they’re changing their minds after pairing this news with additional information, such as whether Trump is sentenced to jail time in July or non-Trump news like what happens with the economy. There are still five months until Election Day, and the way the polls shift today is not the final word on where they will stand in November.

Footnotes

*When a poll released multiple versions of the same question, we used the version among registered voters over the version among adults and the version that asked about third-party candidates over the version that asked a head-to-head.