'Cost of living was the defining issue' 

Mallory Newall, senior vice-president, Ipsos 

Polls correctly predicted a strong Democratic performance, telling the story of an electorate increasingly dissatisfied with the economy and capturing a Democratic edge on enthusiasm and the generic congressional ballot. 

Our polling consistently showed that cost of living was the defining issue of the election. Voters sent a clear signal that concerns about affordability and the economy were top of mind, helping fuel Democratic gains in states like Texas, Ohio, and North Carolina, and underscoring a strong anti-incumbent mood and a desire to hold governing parties accountable. 

'Polls identified a wide enthusiasm gap' 

Scott Keeter, senior survey advisor, Pew Research Center 

In general, our early observations are that the pre-election polls did a good job telling the story of the electorate’s mood this year: deep dissatisfaction with the president – particularly his economic policies, including among some of his 2024 voters.  

Polls also consistently identified a wide “enthusiasm gap” – with Democratic supporters more interested, engaged – and likely to show up – than Republican supporters. It appears that pre-election polls did understate Democratic support in many of the major statewide races for governor and US Senate, though poll watchers are still assessing the magnitude of the phenomenon. Stay tuned for more details as we finish our preliminary analysis. 

'What happens when an affordability crisis meets a crisis of trust' 

Jen Musil, global president, research, The Harris Poll 

These results are what happens when an affordability crisis meets a crisis of trust. All year, our polling showed Americans across party lines saying they could barely keep up with the cost of groceries and gas, and in our recent research, cost of living was the clear number one voting issue, with government corruption close behind in second. Voters connected those two concerns, and then they voted on both. 

'Right story, wrong numbers'

Christopher Barnes, president, Escalent

Tuesday’s election saw pollsters get the story right and the numbers wrong. The majority of polling from prominent pollsters said Democrats were headed for strong midterm results, and they were. But in race after race, the final averages understated the size of the wave, in some of the most watched contests by nine points or more.

'Modelling likely voters is not precise'

Scott Clement, polling director, The Washington Post

Polls showed Democrats were more interested in voting, but modelling likely voters is not precise, and some polls showed a sizable range of estimates across different turnout scenarios.

Relatedly, Democrats were significantly more likely than Republicans to answer polls, which can be a sign of higher turnout interest but also a warning sign for bias. In our own Kansas poll, Democrats were more than twice as likely to complete the survey after answering an interviewer’s call, and the survey found signs that President Donald Trump’s 2024 voters were underrepresented. The poll found Democrat Adam Hamilton leading by nine points, and he ended up winning by five. 

'Any claim of a lasting Republican realignment emerging from 2024 was premature'

Jon Cohen, founder and chief executive, TrueDot.ai

Democrats benefited from an across-the-board swing driven by widespread concerns about affordability and growing aversion to Trump. Just 37 percent of voters approved of the job he was doing as president. The shifts were particularly large among young people and Hispanic voters. Democrats also won independents decisively and drew support from Republicans outside the MAGA core. Any claim of a lasting Republican realignment emerging from 2024 was premature.

Republicans controlled Washington and owned the consequences of growing voter frustration. Voters expected relief on living costs and didn’t think they were getting it. The war in Iran compounded the party’s problems, but no single issue explains a defeat this broad. Support won in 2024 wasn’t a permanent commitment to the Republican Party.

Overall, pre-election polls understated Democratic strength by about four points on average. The storylines were accurate, but there’s substantial error to understand and address. We continue to face substantial sampling challenges: reaching a representative cross-section of voters and accounting for differences between people who answer surveys and those who don’t. We also need to separate changes in vote choice from changes in who turned out. Weighting and turnout models have to capture both. We can identify the errors now; explaining their causes takes more work.