Polling industry insiders in the U.S. agree on one thing: The criminal conviction of Donald Trump has not had any obvious effect on the presidential election.

And they say there's now enough data to assess the fallout — two full weeks after the precedent-smashing guilty verdict of the former president.

"Yeah, I think we can draw preliminary conclusions," said Carl Bialik, vice-president of data science and U.S. politics editor at the YouGov polling firm.

What's clear is that Trump being branded a felon has not had the dramatic impact that pre-conviction surveys suggested it might.

Those older polls suggested he might lose several percentage points, see swing voters flee, or have a staggering 16 per cent to 24 per cent of his supporters reconsider.

Those polls were dealing in hypotheticals. On May 31, reality struck when a Manhattan jury found Trump guilty 34 times of falsifying business records to pay off a porn star.

So in the real world, what was the impact?

Trump led most, but not all, surveys before the verdict; that hasn't changed. If there has been an effect, it's been so small that pollsters disagree on whether it's actually happened — in other words, if it was a methodological rounding error.

Bialik's assessment? Maybe "a point or two" of change — visible if "you kind of squint."

Marc Trussler, director of data sciences at the University of Pennsylvania's opinion-research centre, sees a slightly larger impact, a roughly two-point shift to Biden. 

But another pollster dismisses all of it. "Absolutely no movement whatsoever," said Patrick Murray, director of the polling centre at Monmouth University.

WATCH | Mixed messages from Trump:

Trump unsure public ‘would stand’ for jail time, house arrest

Former U.S. president Donald Trump told Fox News he would be ‘OK’ with house arrest or jail time following his conviction, but he doesn’t think the public ‘would stand for it.’ Trump’s opponents considered the comments as being able to incite violence.

Here's what the latest surveys actually say.

The New York Times found a two-point shift toward Biden after the verdict. YouGov also found Biden with his first lead in months. 

On the other hand, Biden's lead disappeared in YouGov surveys when the survey questions included third-party candidates. To boot, Trump was back ahead in YouGov's latest poll. In fact, Ipsos even has Trump with his best numbers in months.

Another reputable pollster, the Marist Poll, just found Trump up two points in crucial Pennsylvania.

Put it all together and what you get is — maybe, at the very most — some movement within the statistical margin of error. And that movement might be durable — or maybe not. 

"There might have been a statistically insignificant bump for two days after the verdict," said Murray, who says one- or two-point shifts happen in every poll, and can't be attributed to any one event.

"But if there was anything it's already evaporated."

Now here's where most of these analysts will pause to add a critical caveat about American elections: They are notoriously close. A battle of inches. And in such contests, every inch of data, every methodological scrap, might matter. 

The margin of victory in the swing states Arizona, Georgia, Wisconsin and Pennsylvania in 2020 was less than 1.5 percentage points in every case.

Every shift is big when the electorate is divided 50-50, Trussler said, noting that even a half-point shift would have doubled Biden's victory margin in Wisconsin in 2020.

Critical group, wavering support

Trussler sees one specific result in the Times polling to think this shift is significant; a sudden dip in Trump's support from a key sliver of the electorate.

A reason Trump has been leading most surveys is that, in recent months, less-engaged voters have shifted his way — people who don't usually follow politics, only vote sometimes, and who overwhelmingly favoured Biden in 2020, but not anymore.

Among respondents from previous polls who'd said they backed Biden in 2020, but had switched to Trump, the Times found a staggering one-quarter had suddenly reverted to Biden.

"That group is really critical," Trussler said.

He says the reason these voters matter is they're persuadable, certainly more than most political news junkies who are generally attached to one party.

What Trussler finds most telling about this is it signals soft support for Trump from this group, who are about to be bombarded by political ads and news entering the campaign season.

These voters might be dissatisfied with the economy, and with Biden, but they really haven't spent much time thinking about the pros and cons of Trump.

The verdict is a rare chance to grab their attention, and the initial indications are that it has made them reconsider, Trussler says.

"It's easy [for them] to think about supporting Trump when not faced with the actual Trump — the real version of Trump," he said. 

"I would expect them to move back toward Biden, remembering why they didn't support Trump [in 2020]." 

Unpredictable side-effects

One thing the analysts agree on is, with five months left until election day, Trump's conviction may feel like distant news by then. Other events will have unfolded, from the debates and conventions to inevitable surprises. Trump won't learn his sentence until next month.

In the meantime, the conviction can still play out in unpredictable ways.

To Republicans' benefit, it has stoked their fundraising. 

To their potential detriment, it has triggered a reaction on the right that has unnerved some moderates. To protest the conviction, Republicans are now talking about systematically blocking government appointments in the Senate, stalling spending bills and defunding prosecutions — with no telling how that ultimately plays with voters. 

Again, Murray doubles down on his doubts. He's far from convinced there's been any impact with the less-engaged swing voters.

His own survey work shows a two-point shift toward Biden from April, before the verdict, to June, after the verdict, among voters who dislike both major-party candidates. This is not statistically meaningful, he added.  

As for the pundits and politicos who speculated feverishly after the verdict about whether it might move voters, he says they could have spared themselves the suspense.

Anyone who was asking about the verdict's impact on voters, he said, "and who didn't have an inkling that the answer was, 'Absolutely nothing,' has not been paying attention to American politics for the past nine years."

If the Jan. 6 attack on the Capitol didn't "move the needle, then there's nothing else that's going to move the needle at this point."

Murray acknowledges that, in a close election, minuscule shifts matter. Including the scant number of votes that might shift from this verdict.

Just don't expect to see it in a poll, he said.