Fact Check
The claim: Median Strategies conducted legitimate voter surveys showing eye-catching results in the Wisconsin governor race, the Nevada governor race and the Los Angeles mayoral contest. The verdict: false. The organization shut down its website and said the releases were part of a short-term social experiment. The Associated Press reported that the surveys were fabricated. They should not be cited, averaged, compared with real polls or used to describe voter movement.
A real public poll should disclose enough methodology to evaluate how the numbers were produced: sponsor, field dates, population, sample size, mode of contact, question wording, weighting, geographic coverage and a contact for verification. Median Strategies’ releases lacked the kind of complete, consistent documentation reputable analysts expect. The firm had no established track record, and its striking results spread faster than basic questions about who paid for calls, where respondents came from or whether any interviews occurred.
The Wisconsin survey drew particular attention because it appeared to show movement in a high-profile governor race. Other purported polls touched Nevada and Los Angeles. Fox News reported the shutdown and admission, while AP independently verified the central facts. The episode does not prove that every surprising poll is fake. It proves that novelty and a polished graphic cannot substitute for a methodology statement and an accountable organization.
The damage is not limited to one headline. Fabricated polls can influence fundraising pitches, candidate messaging, media narratives and polling averages. Even after deletion, screenshots keep circulating without the later correction. Editors and aggregators should mark affected entries as fraudulent rather than silently removing them, so readers understand why the historical number disappeared. Campaigns that promoted the releases should correct their own posts with equal visibility.
Brutus Proof’s test takes less than a minute: find the methodology, verify the pollster’s identity, locate the sponsor and compare the result with transparent surveys. If those pieces are missing, label the number unverified and do not build a story on it. Confirmed here: the firm admitted the experiment and the cited surveys were not genuine research. Unknown: who organized the operation, the complete motive and the full reach of its data. Until those facts emerge, do not invent a partisan mastermind. The hoax is established; the authorship story is not. Newsrooms should retain screenshots and correction notes, while polling averages should disclose removals. A transparent correction repairs the record better than quietly pretending the number never influenced coverage.
Sources
- Associated Press — Mysterious polling firm shuts down and admits its election surveys were fake (08-18-2026)
- Fox News Digital — Mysterious polling firm behind fake election surveys closes, admitting ‘social experiment’ (08-18-2026)
- Off The Press — Fake Wisconsin poll was part of a social experiment (08-18-2026)