Democratic pollsters acknowledge 'major errors' in 2020 surveys

The Democratic Party’s top pollsters acknowledged on Tuesday that they “failed to live up to” their own expectations in their polling of the 2020 elections, saying that “major errors” led them to believe that Democrats would have a better Election Day than what eventually materialized.

In a memo, five of the largest Democratic firms  ALG Research, Garin-Hart-Yang Research Group, GBAO Strategies, Global Strategy Group and Normington Petts  pointed to a litany of potential issues in their polling, from failing to accurately predict who would actually turn out in the election to misreading the political attitudes of certain groups of voters. 

But ultimately, the pollsters said, the analysis of 2020 failed to yield any clear answers.


“There were several factors that may have contributed to polling error in 2020, and there is not a single, definitive answer  which makes solving the problem especially frustrating,” the polling firms wrote in the memo. 

The memo, which was first reported on Tuesday by Politico, highlights the challenges for Democrats in diagnosing what went wrong in November. 

Democrats won the White House, maintained control of the House and, two months later, went on to win the Senate majority. But the results were still a far cry from the kind of electoral blowout that they had hoped for  and that polling led many to believe they would see. 

Data collected in the weeks and months before Election Day showed President BidenJoe BidenCaitlyn Jenner says election was not 'stolen,' calls Biden 'our president' Manchin, Biden huddle amid talk of breaking up T package Overnight Energy: 5 takeaways from the Colonial Pipeline attack | Colonial aims to 'substantially' restore pipeline operations by end of week | Three questions about Biden's conservation goals MORE well ahead of former President TrumpDonald TrumpCaitlyn Jenner says election was not 'stolen,' calls Biden 'our president' Overnight Health Care: FDA authorizes Pfizer vaccine for adolescents | Biden administration reverses limits on LGBTQ health protections Overnight Defense: US fires 30 warning shots at Iranian boats | Kabul attack heightens fears of Afghan women's fates | Democratic Party leaders push Biden on rejoining Iran deal MORE, while House Democrats appeared to be on the precipice of expanding their majority.

But Biden’s margin of victory was far smaller than what many Democrats expected. The memo from the five pollsters noted that “thanks to the quirks of the electoral college, the difference between a new administration and four more years of Donald Trump was merely 43,000 votes cast across Wisconsin, Georgia and Arizona.”


House Democrats, meanwhile, ended up suffering a particularly tough series of losses, even as they kept their majority. Republicans picked up 12 seats in the lower chamber, putting them easily within striking distance of recapturing the House in 2022. 

The Senate, likewise, only went to Democrats after a pair of wins in two runoff elections in Georgia on Jan. 5. Even at that, the upper chamber is split 50-50, leaving Democrats no room for defections from within their own party.

The memo released on Tuesday noted that low-propensity voters  those who are expected to turn out rarely  broke more for Republicans than for Democrats. But the larger culprit, the pollsters said, was likely a failure to accurately capture the attitudes of certain voters.

That’s largely because many Trump voters simply did not participate in polling, the memo says. The former president spent months sowing distrust in institutions and trying to discredit polling that showed him well behind Biden. Consequently, many of his supporters simply didn’t respond to pollsters. 

“What we have settled on is the idea there is something systematically different about the people we reached, and the people we did not,” the pollsters said. “This problem appears to have been amplified when Trump was on the ballot, and it is these particular voters who Trump activated that did not participate in polls.” 


The firms acknowledged that the refusal by some voters to participate poses a particular problem that they don’t have a solution for. They said that they would “embark on a number of experiments over the course of this year” that will seek to correct past errors in their methodology. 

But they also cautioned against the impulse to “overcorrect” in the wake of 2020, noting that polling error has been far more significant in presidential races  especially when Trump has been on the ballot. 

“Our industry must figure out how to improve, and it is not going to be easy,” the pollsters wrote in the memo. “Polling was very accurate in some places and inaccurate in others, and the explanation for why is not yet clear.” 

“We believe polling plays a critical role in our democracy and gives a voice to the American people. And we believe we can, and must, do much better.”