Wedgewood Polls has admitted fabricating US election surveys that were published by major news outlets and incorporated into polling forecasts. A former Senate candidate says the false figures damaged her campaign, raising serious questions about misinformation and the reliability of political reporting.
A US polling organisation whose election surveys were used by major news outlets, including The New York Times and Newsweek, has admitted fabricating polling data ahead of the November 2026 midterm elections. Wedgewood Polls claimed the deception was a social experiment designed to expose weaknesses in the polling industry. The admission followed an investigation by election analysis website FiftyPlusOne, while a former US Senate candidate says one of the fabricated surveys seriously damaged her campaign.
How fabricated election polls reached major news organisations
Wedgewood Polls began publishing US election surveys in June 2026, presenting itself as an independent polling organisation.
Its figures were subsequently reported by publications including Newsweek, The Hill, Breitbart, The Daily Beast and the Washington Examiner.
The results also entered election polling databases maintained by The New York Times, political analyst Nate Silver, VoteHub, FiftyPlusOne and Decision Desk HQ.
These organisations do not simply report individual polls. Several use polling data to calculate averages, analyse electoral trends and generate election forecasts.
That meant fabricated surveys could influence the information presented to journalists, political campaigns, donors and voters.
On 20 October, Wedgewood publicly acknowledged producing fabricated results after investigators questioned the authenticity of its underlying data.
The organisation claimed it had been conducting a social experiment to demonstrate how easily an apparently credible polling company could gain recognition.
However, the admission followed questions from investigators who were already preparing to publish evidence of serious irregularities.
Wedgewood’s explanation of its motives has not been independently established.
How investigators uncovered the fabricated data
The breakthrough came from FiftyPlusOne, a US election analysis organisation.
Its researchers became suspicious of a Wedgewood survey concerning the Mississippi Senate election.
The survey placed Republican Senator Cindy Hyde-Smith just four percentage points ahead of Democratic challenger Scott Colom.
Investigators requested the original survey data and examined the records supposedly generated by the polling platform Pollfish.
They discovered inconsistencies that were difficult to reconcile with genuine survey responses.
Among the warning signs were respondent identification numbers that did not follow the expected patterns and transaction identifiers that appeared inconsistent with when interviews had supposedly taken place.
To test their suspicions, FiftyPlusOne commissioned its own survey using the same polling platform and questionnaire, spending $900 to obtain 600 responses.
The resulting data displayed markedly different characteristics.
Researchers also checked their findings against records from another polling organisation and information supplied by Pollfish.
When Wedgewood was confronted with the discrepancies, it admitted fabricating polling data.
The investigation demonstrates why examining the evidence behind a survey can be more revealing than simply assessing whether its headline figures appear plausible.
Major media outlets remove false polls
Following the admission, news organisations and polling analysts began removing Wedgewood’s results from their reporting and forecasting systems.
Newsweek confirmed that it had retracted three articles and corrected others.
The New York Times removed 15 Wedgewood surveys from its polling project after explaining that the organisation had previously answered questions about its methods, personnel and funding.
Decision Desk HQ removed 17 Wedgewood polls from its race-level polling lists and averages.
It subsequently reran its forecasting models and reported that removing the fabricated data did not change its projected congressional seat totals or its probabilities for which party would control either chamber.
Nate Silver also removed Wedgewood’s results. His election model showed some changes in individual race estimates, although the overall forecast was little changed.
These distinctions matter.
The fabricated surveys entered recognised forecasting systems, but that does not establish that they changed the predicted outcome of the election or altered how voters intended to vote.
The wider issue is how apparently legitimate polling information was able to circulate through established news and analysis services before its authenticity was challenged.
Former Senate candidate says fake poll damaged her campaign
The controversy has also raised questions about possible consequences for political candidates.
Michigan state Senator Mallory McMorrow, who suspended her campaign for the Democratic nomination for the US Senate on 5 July 2026, says a fabricated Wedgewood poll harmed her campaign.
A Wedgewood survey released on 24 June had placed McMorrow on just 9% support, substantially below two rival candidates.
McMorrow subsequently said the figures contradicted her campaign’s internal polling.
Following Wedgewood’s admission, she described how the published results had affected the campaign.
Donations dried up. Volunteers were devastated and pleading for guidance on what to do.
She also said journalists had pressed her to withdraw.
The former candidate argued that an exercise presented as an experiment had real consequences for people who had invested time, money and effort in her campaign.
However, it has not been independently established that the fabricated Wedgewood survey caused McMorrow to withdraw.
Other genuine polling had also placed her behind her rivals.
At the time of her withdrawal, Associated Press reporting identified substantial outside campaign spending supporting rival Haley Stevens as another major factor.
McMorrow’s account nevertheless illustrates the potential consequences of publishing unreliable election data.
Even when figures are subsequently withdrawn, campaigns may already have faced questions from donors, supporters and the media.
Why did established polling organisations accept the surveys?
One of the most important aspects of this case is that the fabricated polling did not enter every database without scrutiny.
FiftyPlusOne says it had contacted Wedgewood when the organisation first appeared, requesting information about its personnel and methodology.
The explanations initially appeared credible, including details of individuals said to have previous polling experience.
The New York Times similarly reported asking Wedgewood questions about its personnel, sampling methods, survey weighting and funding before accepting its results.
This means the problem was not simply the complete absence of verification.
Rather, the case exposed how traditional checks can be defeated when an organisation supplies convincing but unreliable information.
FiftyPlusOne has said it intends to strengthen its procedures for accepting new polling organisations, potentially including direct meetings with their representatives and additional examination of underlying survey records.
Did money change hands?
Another unresolved issue concerns Wedgewood’s activities beyond publishing false survey results.
FiftyPlusOne reported that the organisation had collected money from contributors to conduct crowdfunded polls while presenting itself as a legitimate polling company.
This raises questions about what contributors believed they were funding and whether the money was used as represented.
However, the available evidence does not establish a criminal offence, and any suggestion of financial fraud would require further investigation.
The organisation’s claim that it was conducting an experiment does not, by itself, resolve questions about payments, transparency or potential civil liability.
Polling experts condemn fabrication
The American Association for Public Opinion Research issued a statement on 21 October condemning fabricated polling information.
The professional organisation emphasised that legitimate public opinion research is based on actual responses from members of the public.
It also warned that deliberate fabrication undermines confidence in legitimate research.
The distinction is important because a poll that produces an inaccurate prediction is not necessarily fraudulent.
Genuine political surveys contain statistical uncertainty and may be affected by sampling limitations, turnout assumptions, changes in public opinion and other methodological factors.
Fabricating survey results is fundamentally different.
In that situation, the underlying evidence being presented as public opinion research may not exist at all.
Could artificial intelligence make fake polling easier?
The Wedgewood investigation also raises concerns about how modern technology can make misleading information appear credible.
FiftyPlusOne noted that AI-assisted tools can produce professional-looking websites, spreadsheets and polling reports with relatively little effort.
Such materials may resemble the documents published by established research organisations.
However, the investigators explicitly cautioned that using AI is not evidence of misconduct.
Many legitimate organisations use automated tools to prepare reports, analyse information or format research materials.
The danger arises when convincing presentation is mistaken for proof that the underlying information is genuine.
This is particularly relevant to journalism, where a professional-looking document or widely repeated claim can acquire apparent authority even when the original evidence remains unverified.
How can readers judge whether an election poll is trustworthy?
For readers encountering political polling, several basic checks can help distinguish transparent research from unsupported claims.
- Who conducted the poll? Look for an identifiable organisation, named researchers and a record of previous surveys.
- Who paid for it? Funding sources and commissioning organisations can provide important context.
- How was it conducted? Reliable polling reports should explain their sampling methods, interview dates, sample size and weighting procedures.
- Are other polls showing similar results? An unusual finding is not automatically wrong, but it may require closer examination.
- Is the original methodology available? Reporting should ideally link to the original research rather than relying on second-hand summaries.
The AAPOR guidance on survey research provides further information about evaluating public opinion surveys.
No single check guarantees authenticity, as the Wedgewood case demonstrates.
However, greater transparency and independent examination can make deliberate deception more difficult.
What the scandal means for election reporting
The Wedgewood case has exposed vulnerabilities in a system that depends heavily on trust between pollsters, data aggregators and news organisations.
Its fabricated surveys were reported by publications with different political viewpoints, demonstrating that unreliable information can spread across the media landscape.
Some organisations conducted preliminary checks, but the deception was not uncovered until investigators closely examined the underlying survey records.
The consequences for individual campaigns remain difficult to measure. McMorrow has described genuine concerns about damage to her campaign, while available evidence does not establish that the fabricated poll alone caused her withdrawal.
There is also no established evidence that Wedgewood’s results changed an election outcome.
Nevertheless, the episode demonstrates how misleading data can enter political reporting, influence public discussion and potentially affect campaign decisions before its authenticity is challenged.
The lesson is not that all political polling is unreliable. It is that polling data, like any other significant claim, requires evidence that can withstand independent scrutiny.
The original FiftyPlusOne investigation explains how the fabricated data was identified and sets out the organisation’s proposed changes to its verification procedures.
Sources and further reading
- FiftyPlusOne: investigation into Wedgewood polling data
- American Association for Public Opinion Research: statement on fabricated polling
- Associated Press: Mallory McMorrow’s withdrawal
Watch: Farron Balanced discusses the fake polling scandal
Political commentator Farron Balanced examines the Wedgewood controversy, its treatment by US media organisations and the allegations of damage to election campaigns.
The video includes commentary and interpretations that go beyond the independently established findings reported above.
