For pollsters hoping to accurately predict an election with Donald Trump at the top of the ticket, the third time may have been the charm — but it was still somewhat unsatisfying.
Opinion pollsters waved and missed in 2016 and 2020, systematically underestimating the level of support for Trump. But in 2024, the polls corrected — and both the popular vote and swing state results came within the margin of error for aggregate predictions.
Pollsters say they have finally met the challenge of pinning down Trump voters, an elusive slice of the voting population that has caused polls to veer off target in previous election cycles when Trump was on the verge.
“In the past, we have had many Trump supporters who simply refused to answer our questions,” said Republican pollster Whit Ayres. “We call, ‘I’m from the New York Times or The LA Times or The Washington Post, and I’m doing a survey,’ and they say, ‘Well, to hell with you,’ click.”
This is not to say that there were no flubs. Notably, Gold Standard pollster Ann Selzer, who is legendary in the industry for being right, ended her long relationship with the Des Moines Register this year after her company published a poll a few days before Election Day showing that Vice President Kamala Harris defeated Trump by 3 points in Iowa.
The poll results were met with both shock and skepticism, with even Harris’ campaign warning against reading too much into the results. Trump ultimately defeated Harris in Iowa by more than 13 points.
Since 2016, when polls predicted a landslide victory for Hillary Clinton, pollsters have faced problems with underestimating Trump’s base, a challenge unique to elections that feature Trump as the polls in the 2018 midterm elections and 2022 were relatively accurate.
Turnout tends to be higher in presidential election years — especially among those in Trump’s base who tend to distrust institutions and can be difficult to oversee.
“The same Trump voters who don’t trust experts, don’t trust the media, don’t trust science — don’t trust pollsters either,” said Democratic pollster Paul Maslin. “And we found in several states that they are simply opting out.”
To course-correct, researchers “jumped through a lot of hoops” this year, Ayres said, to reach a larger number of Trump voters and predict the outcome more accurately. They have adjusted their modeling of the likely electorate, weighted certain demographic groups more heavily and adjusted their outreach strategy to non-college-educated voters, who tend to lean Republican.
That was reflected in the latest polls showing five of the swing states in a dead heat, with Trump comfortably ahead in Arizona and Nevada. The polls also accurately predicted Trump’s strength against white voters and Harris’ softness against black and Latino men, contributing to Trump’s decisive victory.
The president-elect ultimately won all the battlegrounds, an outcome that pollsters attribute to undecided voters who split unevenly — and perhaps surprisingly — toward Trump.
“If you look at the past, the undecided have generally not favored Republicans,” said Republican Party pollster Brent Buchanan. “This year, Trump took advantage of being undecided.”
Although the results on election night did not deviate from what pollsters had predicted, they were still surprising in magnitude. In the run-up to November 5, many journalists and experts are bracing themselves for a long and arduous process that would take days to resolve due to the closeness of the polls in swing states.
Although the final vote counts were within the margin of error of polls in battleground states, they still tended to underestimate his support by about three points. That could be because Trump took advantage of voters who made their decision at the last minute, after the final surveys had been conducted.
In the final week of the campaign, Trump’s campaign received the coveted endorsement of Joe Rogan, to which Maslin credited Trump’s last-minute surge. Buchanan’s models showed that Trump had a three-point lead among voters who made their choice in the final week of the election.
Pollsters have largely blamed Trump for variable inaccuracies in 2016 and 2020. “It’s a Trump issue, not a Republican issue,” said Ayres, who has four decades of polling experience in the Republican Party. Looking to 2028 — the first presidential election without Trump in 12 years — Eyres predicted that “we can reasonably expect whatever problems we had to make Trump voters disappear.”
Others weren’t so sure.
“It’s not like, ‘Oh, great, fine, the election was pretty good, this time we can put this to bed,'” Maslin said. “No, of course not. It’s an ongoing question. It will remain that way, and it must remain that way.”
