Wedgewood Polls admitted most of its published election data was fabricated.
Why it matters: The episode shows how unverified polling can enter media coverage, polling averages and political narratives before basic data checks are complete. Campaigns, investors and news organizations face risks when plausible-looking numbers influence decisions before their provenance is established.
- Wedgewood made the admission on October 20, 2026, calling the operation an experiment exposing weak polling verification.
- FiftyPlusOne found irregularities in 511 respondent records from Wedgewood's Mississippi Senate survey conducted October 12-14.
- The New York Times removed Wedgewood surveys from its polling tracker, while Decision Desk HQ removed 17 polls from its lists and averages.
- Wedgewood was the second polling organization in 2026 to acknowledge fabricating election data, after Median Strategies.
Wedgewood said it fabricated most of its published election-polling data to demonstrate how easily unverified figures could gain credibility. "Our fabricated numbers were given instant credibility across national media platforms," the organization said.
The admission followed an investigation by FiftyPlusOne into raw data from a Mississippi Senate survey. The group said Wedgewood's respondent and transaction identifiers did not match patterns in genuine Pollfish data. In a 600-respondent replication survey costing $900, 51.1% of genuine respondent IDs began with a dash; none of Wedgewood's did. The correlation between transaction IDs and completion times was 1.0 in FiftyPlusOne's sample and -0.03 in Wedgewood's data.
Wedgewood's surveys had been cited or tracked by Newsweek, The Hill, Breitbart, The Daily Beast, The Washington Examiner, The New York Times, Decision Desk HQ, VoteHub and FiftyPlusOne. The Times removed all Wedgewood surveys from its polling project after the admission. Decision Desk HQ removed 17 polls from its race-level lists and averages, according to reporting by The Guardian.
The organization appeared online only in June 2026 and did not publicly identify employees, a location or a telephone number, according to that report. The episode followed Median Strategies' August admission that it had created fake polls in California, Wisconsin and Nevada.
The American Association for Public Opinion Research said fabricated responses undermine public trust and that legitimate polls should be based on actual respondents and disclose methodology. The identities of Wedgewood's operators, the full scope of its fabricated data and any financial or political effects remain unresolved.
By the numbers
- 511 - respondent records in the disputed Mississippi survey
- 17 - Wedgewood polls removed by Decision Desk HQ
- 60 million - approximate jump in responses implied by disputed transaction IDs, according to FiftyPlusOne
Yes, but: Wedgewood's explanation that the data was a social experiment has not been independently verified. It remains unclear whether anyone profited, whether campaigns acted on the data or whether laws were violated.
What's next: Independent verification of Wedgewood's full polling record, operators, data sources and any financial or political effects remains outstanding.