An American polling organization has admitted fabricating the vast majority of its election surveys, exposing how invented figures entered major polling averages and national media coverage just two weeks before the November midterms. Wedgewood Polling made the unexpected admission on Tuesday, 20 October, after investigators confronted it with evidence that its supposed voter interviews were fake.
In a public statement, Wedgewood described the operation as a ‘social experiment’ intended to expose weak oversight in the political polling industry. Its organizers said they wanted to discover how easily a fictitious polling firm could gain credibility. ‘Far too easy,’ they concluded, claiming their data had been picked up by The New York Times, FiftyPlusOne, VoteHub and other major aggregators despite lacking genuine underlying surveys.
According to Wedgewood, the team, which consists of senior professionals from the polling industry, conducted only one genuine survey, in Nebraska, as a baseline for the alleged experiment.
Wedgewood Polls on X (formerly Twitter): “The polling industry needs major reform. Our statement below: pic.twitter.com/Sj8JLYgply / X”
The polling industry needs major reform. Our statement below: pic.twitter.com/Sj8JLYgply
How Wedgewood Has Influenced the Midterms So Far
The confession followed an investigation by FiftyPlusOne, a polling and forecasting organization that had been suspicious of Wedgewood’s data for months. After receiving evidence of possible fabrication, researchers obtained Wedgewood’s raw data from a Mississippi Senate survey and compared its respondent identifiers with genuine records from Pollfish, the platform Wedgewood claimed to use. The identifiers and transaction timestamps displayed patterns incompatible with authentic interviews. When confronted, Wedgewood issued its confession rather than explaining the anomalies.
The fake polls had infiltrated some of the United States’ most prominent polling databases and election forecasting platforms, including The New York Times, Nate Silver’s Silver Bulletin, FiftyPlusOne, VoteHub, Decision Desk HQ and 270toWin. While some merely listed Wedgewood’s surveys, others incorporated the fabricated figures into their weighted polling averages and election forecasts.
Following the scandal, Decision Desk HQ removed 17 surveys and recalculated its forecasts, although it said the correction did not change its projected congressional seat totals or probabilities of party control. FiftyPlusOne similarly deleted Wedgewood’s surveys and permanently barred future submissions from its identified operators, while disputing the firm’s accusation that aggregators had conducted no vetting. Investigators also revealed that Wedgewood had collected donations to finance its supposedly legitimate surveys.
The real-world implications of the fake numbers now look particularly awkward in hindsight. In Mississippi, Wedgewood reported Republican Senator Cindy Hyde-Smith leading Democratic challenger Scott Colom by just 52 to 48 per cent in a state Donald Trump carried comfortably in 2024. Colom’s campaign immediately seized on the survey and promoted it as evidence of a virtually tied race, while Newsweek wrote a full report around it, arguing that a historic Democratic upset in the state could be possible in November.
In Michigan, former Democratic Senate contender Mallory McMorrow said a Wedgewood survey showing her plunging into single-digit support helped devastate fundraising and volunteer morale. Journalists pressed her about abandoning her campaign, she recalled, after the Wedgewood poll contradicted her internal numbers. McMorrow subsequently withdrew from the primary.
The organization also published a fictitious October Michigan Senate poll showing Democrat Abdul El-Sayed ahead of Trump-backed Republican Mike Rogers by 53 to 47 per cent. Other invented surveys showed Democrats competitive in Republican-held states, including Kansas, Ohio and Texas.
‘Other invented surveys showed Democrats competitive in Republican-held states, including Kansas, Ohio and Texas’
The two political camps drew different conclusions from the still-evolving scandal. For Republicans, it reinforces long-standing complaints that the polling and media establishment can amplify misleadingly favourable Democratic narratives, thereby dragging down conservative campaigns. Democrats have a separate grievance: McMorrow’s experience shows that fabricated numbers can damage candidates within their own party. In that particular case, it helped a radical democratic socialist candidate defeat the moderate one.
With or without Wedgewood, the American polling industry’s record offers substantial grounds for scepticism. In three consecutive presidential elections, major pollsters systematically overstated Democratic support, inflating Hillary Clinton’s advantage over Donald Trump in 2016, exaggerating Joe Biden’s lead in 2020 and once again overestimating Kamala Harris’s performance in 2024. According to the American Association for Public Opinion Research, polls overstated Democratic margins by an average of 3.1 percentage points in 2016, 4.6 points in 2020 and 2.7 points in 2024 across electoral contests.
Related articles:
The post Wedgewood Admits Fabricating Polls, Many Boosting Dems in GOP States appeared first on Hungarian Conservative.