Donald Trump has been a convicted felon for a little over a week and he has already lost a small but significant chunk of support, according to the polls. 

In recent days, the New York Times and Sienna College recontacted 2,000 voters who had taken their surveys previously and found Trump’s advantage among that sample had declined by one point following his conviction for falsifying business records. The presumptive GOP nominee had led Joe Biden by three points before the jury’s verdict, but now bested the president by just two. 

A similar recontacting study by the Republican firm Echelon Insights yielded the same basic result, with respondents giving Biden a two-point lead over Trump after previously splitting their support evenly between the two candidates. A Reuters/Ipsos poll taken immediately after Trump’s conviction showed Biden leading with 41 percent of the vote to his rival’s 39 percent. The last survey taken by the pollster before the jury’s verdict had shown the race tied. 

These shifts were not outside the margins of error in the Reuters/Ipsos and Echelon Insights surveys, while the Times’s pollsters said that they could not calculate such a margin for their recontacting survey. Nevertheless, the fact that the same shift was recorded across three different surveys lends credence to its validity. 

RealClearPolitics’s polling average, meanwhile, shows a smaller but non-negligible dip in Trump’s support: On the day of the verdict, Trump led Biden by 0.9 percent; by June 6, that had dropped to 0.5 percent.

There’s good reason to think that Trump has taken a hit from several days of headlines about his criminal efforts to conceal an affair with adult film actress Stormy Daniels. But it’s less clear whether this shift toward Biden will fade or snowball in the months between now and Election Day.

On the one hand, major news stories often trigger polling spikes that dissipate when a new shiny object captures the media’s attention. And historically, this has been especially true of scandalous headlines about Donald Trump.

On the other hand, Trump’s lead over Biden has long rested on a shaky foundation: In many polls, the Republican has owed his advantage to unusually high support among politically disengaged, Democratic-leaning voters. There’s always been cause to suspect that these voters would eventually find their way back into Biden’s camp, and the available data indicates that Trump’s conviction might have served as a catalyst for returning this bloc to the Democratic fold.  

Trump’s invincibility to the taint of scandal is often overstated. He is unusually unpopular for a Republican politician and underperformed his party in both the 2016 and 2020 presidential elections, when House Republicans’ share of the vote nationwide far exceeded Trump’s.

Nevertheless, it is the case that Trump has repeatedly seen his approval decline in the immediate aftermath of a reputationally damaging event, only to regain this lost ground in fairly short order.

Consider the Access Hollywood tape. One month before the 2016 presidential election, the Washington Post published a video in which Trump told a host of that TV program that when he saw attractive women, “I don’t even wait. And when you’re a star, they let you do it. You can do anything … Grab ’em by the pussy. You can do anything.” In the audio, the Christian right’s champion also touted his attempts to seduce a married woman.

On the day that tape was published, Trump trailed Clinton by 4.7 points in RealClearPolitics’s polling average. A little over a week later, Clinton’s lead had swelled to 7.1. 

And yet, by Election Day, Clinton’s polling advantage had declined to 3.2 points, and her actual margin in the popular vote was just two. It’s plausible that the rapid decline of the Access Hollywood tape’s salience reflected the reopening of the FBI’s investigation into Hillary Clinton’s email server on October 28, 2016, as well as Wikileaks’ release of emails linked to her campaign chairman John Podesta on October 7. But this just underscores the fact that emerging news stories can erase the political relevance of older ones. 

Trump’s political rebound after January 6 might be even more telling. On that day in 2021, Trump’s net disapproval was 10.8 points in FiveThirtyEight’s polling average. Two weeks after he instigated an insurrection at the US Capitol, his net disapproval hit 19.3 points. But by June of that year, it was back down to 11.7. Today, Americans disapprove of Trump by a 12.3-point margin.

The dip in Trump’s support since his conviction is much smaller than those he suffered after the Access Hollywood tape or the January 6 insurrection. Indeed, it’s so small that it could be illusory, a reflection of what pollsters call “nonresponse bias”: When bad news comes out about a politician, voters who oppose them often become more eager to discuss politics (and thus more likely to take a pollster’s call) while voters who support the embattled candidate become less interested in such discussions. This can produce shifts in polling that don’t actually reflect changes in support. 

Recontact surveys like those conducted by the Times and Echelon Insights try to account for this by polling the exact same pool of respondents. Inevitably, though, some previously polled voters don’t take the follow-up survey while others do. Partisanship can potentially influence who does or doesn’t stay in the sample.

This said, it seems more likely than not that there has been some genuine movement toward Biden, but given the small scale of the shift, it would not be surprising if it faded over time, especially if Trump manages to escape a prison sentence. In the estimation of Republican pollster Kristin Soltis Anderson, some Trump-leaning voters might have a harder time backing a GOP candidate facing a prison sentence than one whose felony merely resulted in a fine. 

There is another way to interpret the post-conviction polls. 

Trump’s lead in national surveys has rested on an extraordinary degree of support from various Democratic-leaning demographic groups. In 2020, Biden won voters under 30 by 23 points, Hispanic voters by 35, and Black ones by 79, according to the Democratic data firm Catalist. A recent Fox News poll showed Trump leading Biden by 10 with young voters, trailing him by only 36 with Black voters, and by only 5 with Hispanic ones.

Trump’s support within these demographics has been concentrated among the politically disengaged. In the Times’s most recent polling, Biden leads Trump among voters who cast a ballot in 2020 by two points, while trailing him among those who sat out that election by 14. And the Biden 2020 voters who’ve been most likely to defect to Trump in the polls have been those who pay relatively little attention to politics and did not cast a ballot in the 2022 midterms.

This isn’t the sturdiest foundation for a majority coalition. Beyond the fact that, by definition, low-propensity voters can’t be relied on to turn out on Election Day, historically Democratic voters who say they support Trump but aren’t paying much attention to politics would seem especially likely to switch allegiances once they actually tune in. After all, such voters have no longstanding ideological or identity-based reservations about supporting the Democratic Party. And virtually all the reasons for a voter to prefer Biden over Trump in 2020 still apply in 2024. 

It therefore isn’t entirely surprising that, in its recontact poll following Trump’s conviction, the Times found that the shift toward Biden was “especially pronounced among the young, nonwhite and disengaged Democratic-leaning voters who have propelled Mr. Trump to a lead in the early polls.” Of those respondents who had previously told the paper that they voted for Biden in 2020 but were going to back Trump in 2024, roughly a quarter said that they are now planning to vote for Biden in November.

Trump’s margins with young and nonwhite voters looked a bit suspect. It would be extremely unusual for a party to gain so much support from historically unfavorable demographic groups in just four years. And this would be all the stranger in an election where both parties were running the same presidential candidates as last time.

Given that reality, there’s reason to suspect that Biden’s post-conviction gains may stick, less because the Republican nominee’s conviction has had a transformative impact on public opinion than because it prompted discontented, disengaged Democratic voters to tune back in to politics for a moment and remember why they dislike Donald Trump. 

Ultimately, it’s impossible to know whether last week’s verdict will have a durable impact on the 2024 race. Trump’s legal woes could lose salience in the weeks and months ahead while other concerns come to the forefront. If Biden’s debate performance later this month affirms the electorate’s suspicions about his senility, then that may restore Trump’s margins with the president’s historically Democratic skeptics. 

For the moment, though, it looks as though Trump’s conviction may have hastened disaffected, disengaged Democrats along their path back to Biden’s coalition.