Second Fake Pollster in Two Months Admits It Invented Its Numbers

A polling firm whose surveys were cited by The New York Times, Newsweek and Decision Desk HQ admitted Tuesday that nearly all of them were fabricated, the second firm since August to confess that the polls it released to the press were never real.

Wedgewood Polls called itself a “fictional entity run by a small group of industry veterans” and said its work was an “experimental project to demonstrate the vulnerabilities” of the polling industry, PBS NewsHour reported. “We wanted to test a simple hypothesis: How easy is it for an entirely fabricated polling firm to gain traction, be taken seriously, and expose how little scrutinization exists in media and public commentary?” the firm said in its statement. “The short answer: Far too easy.”

How it was caught

The admission came hours after FiftyPlusOne, a polling and election analysis site, confronted the firm with problems in the raw data from an October 12 to 14 survey of the Mississippi U.S. Senate race, the Lynnwood Times reported. The file the firm supplied showed respondent identifiers and timestamps that did not match authentic records from the survey platform it said it had used. Wedgewood said only one of its releases, a crowdfunded survey of Nebraska’s Senate race, was based on real interviews, according to Newsweek.

The firm opened its account on X in June, PBS reported. Its website is now offline.

Aggregators have pulled the numbers. The New York Times removed 15 Wedgewood polls from its polling tracker, and Decision Desk HQ said on X that it had “removed 17 polls from our race-level polling lists and averages.” The Times added an editor’s note to an article that had featured a Wedgewood poll.

Andrew Gelman, a professor of statistics and political science at Columbia University, doubted the firm’s explanation. “Maybe so, but it’s kind of suspicious that they happened to release this statement on the very day that they got caught,” he wrote.

The numbers did real damage before anyone checked them. Mallory McMorrow, a Michigan state senator who ended her U.S. Senate campaign in July, wrote that a June Wedgewood poll showed her “plummeting into single digits,” and that afterward “donations dried up.”

The second time since August

In August, a firm called Median Strategies admitted that its polls in the Wisconsin governor’s primary and the Los Angeles mayor’s race were invented, calling them a “short-term social experiment.” The Texian Partisan reported then that newsrooms across the country had carried those numbers as fact before anyone asked to see the data.

Two months later, the same thing happened at a larger scale. Wedgewood’s polls did not stay on the margins. They went into the averages that news organizations use to tell the public who is ahead and what voters think.

What it means for Texas

Texans have heard for years what a poll says they think about independence. The Texian Partisan has reported on the problems with those surveys: a 2024 university study that leaned on undergraduates and urban respondents, and a YouGov poll this February that estimated state-level opinion from roughly 500 respondents per state. That coverage also showed how question wording alone can swing measured support for independence from 40 percent to 60 percent.

Daniel Miller, president of the Texas Nationalist Movement, made the same point on his Late Night Coffee Talk livestream in March, saying the wording pollsters use is “artificially suppressing the results in favor of independence by on average, anywhere from 20 to 30 percent.”

Wedgewood shows how much further the problem goes. A poll does not have to be badly designed to mislead. It does not have to exist at all. A firm that posted no names and no address on its website, and by its own account had real data behind almost none of its releases, placed 15 surveys in the polling tracker of The New York Times.

“Polling is broken. It is fundamentally, absolutely shattered, broken,” Miller said in March. On Thursday, he went further. “It is social engineering amplified by legacy media,” he said. “It’s propaganda pure and simple.”

That is why the Texas Nationalist Movement has never treated a poll as the answer to the independence question. A poll is a claim about what Texans think, made by people who may or may not show their work. A binding referendum is Texans answering for themselves, on a ballot with a paper trail and a certified result. Nobody can invent that result, and nobody can withdraw it the day after they are caught.