A little-known polling outfit that repeatedly boosted far-left Democratic candidates admitted fabricating survey numbers as a social experiment, leaving party strategists furious over real campaign damage.
Fox News Digital reported that Wedgewood Polling released a Tuesday-night statement confessing it used “fabricated numbers” while billing itself as an election survey company. The firm cast the stunt as an “experimental project to demonstrate the vulnerabilities, lack of oversight, and overall low barrier to entry in the modern political polling industry.”
Those numbers did not sit in a vacuum. Campaigns, donors, aggregators, and reporters treated them as real signals in contested Democratic primaries and general-election matchups. Progressive hopefuls got the lift. Rivals watched money and momentum dry up.
Wedgewood insisted the project “was never intended to sway an election” and framed it as a “wake-up call” on data integrity and vetting. Democrats who lived with the fallout rejected that excuse outright.
In Michigan’s high-profile Democratic Senate primary, Wedgewood numbers circulated while Abdul El-Sayed, Mallory McMorrow, and Rep. Haley Stevens competed. El-Sayed’s campaign cited the firm’s polls. El-Sayed advanced. McMorrow did not recover the ground she lost when the fake figures landed.
McMorrow said the surveys helped choke off donations and left reporters pressing her to quit. A Democrat strategist who said she ran a super PAC backing McMorrow put the money damage in blunt terms.
Jennifer Holdsworth said the Washington Examiner environment of false polling left donors skittish at the worst possible moment:
"this poll was used by several donors to pull back funding at a time we were being outspent by $80 million."
McMorrow posted her own accounting of the human cost after the admission.
"Sure, a point was proven. Polling is broken. But tens of thousands of people poured their hearts, time, energy, hard-earned money into something they believed in - for a year-and-a-half, and it was treated like a game,"
She added that trust is already at rock bottom and that media outlets hooked on poll drama need to separate legitimate work from junk. The New York Post reported the firm later admitted only one of 17 published surveys this cycle was real, and that a June Wedgewood poll putting McMorrow in single digits cratered fundraising and pushed her toward the exit against El-Sayed.
Breitbart News likewise reported the fabricated data helped drive McMorrow out after a late-June survey showed her at 9 percent, at odds with her campaign’s internal numbers. In her fuller recounting, she said donations dried up, volunteers were devastated, and reporters hounded her with calls to drop out.
Major outlets and models did not catch the fraud in time. The New York Times removed 15 Wedgewood polls released between June and October after the admission. The Times said Wedgewood had answered vetting questions about pollster identities, sampling and contact methods, weighting, and funding before the confession.
Decision Desk HQ pulled 17 Wedgewood polls from its projections. It said the cleanup did not change projected seat totals or chamber-control probabilities in the Senate or House. Political Polls, a nonpartisan aggregator, had cited Wedgewood in Wisconsin’s gubernatorial picture, showing Democratic Assembly member Francesca Hong with an edge over Rep. Tom Tiffany. Hong later lost the Democratic primary to the establishment-backed candidate.
FiftyPlusOne was already digging when the statement dropped. Co-founder G. Elliott Morris wrote that his outlet had been investigating Wedgewood for faking data and had emailed for comment that morning, ahead of planned findings the next day.
"We were investigating Wedgewood for faking their data and emailed them for comment this morning, ahead of publishing the findings tomorrow. They are trying to cover up their fraud with the 'we were doing a social experiment' excuse."
Former Obama staffer Tommy Vietor called the conduct “lying and fraud.” Election-integrity fights already run through broadcast rules and ballot procedures, from FCC scrutiny of alleged fake polls before the midterms to courtroom battles over what IDs count at the polls. Fake survey shops only add another layer of noise voters are asked to ignore.
Not every aggregator played along. RealClearPolitics explained it never included the Wedgewood numbers because the operation lacked credentials, a track record, staff directories, and a legitimate web presence. RCP polling director Kavon W. Nikrad said the site keeps a very high standard for averages, “which this pollster did not begin to meet.” >
That standard mattered. The fake surveys reached The New York Times, FiftyPlusOne, Decision Desk HQ, VoteHub, and other pipelines that shape horse-race coverage. National Review argued the episode underscores the need for skeptical appraisal of data, not a blanket decision to ignore all polling. Ranges and probabilities still beat made-up point estimates dressed up as science.
Wedgewood’s own statement, as carried in later coverage, put the hypothesis in the open: how easy is it for an entirely fabricated polling firm to gain traction, be taken seriously, and expose how little scrutiny exists in media and public commentary? The firm’s short answer was that it proved far too easy.
Veteran Democratic strategist Mike Nellis, a former Kamala Harris advisor, did not buy the moral-high-ground pitch. He told Fox News Digital the operators should face real accountability.
"These people are absolute scumbags,"
Nellis continued that they lied about polling data, got caught, and were scrambling because they had committed fraud. He said the people responsible should be run out of American politics and, if possible, prosecuted to the full extent of the law. He also flagged crowdfunding and asked whether the money would be returned.
Wedgewood Polls had run a GoFundMe seeking $5,000 for Nebraska governor and Senate polling. It raised $500. No public accounting in the available reporting shows a full return of those dollars, and no charges were described as filed. Nellis warned both parties and the press should rethink verification, “because there's likely other people that are doing things like this.”
Michigan was not the only theater. In Wisconsin, the Hong-Tiffany framing gave a socialist-labeled Democrat an early narrative edge she could not convert in the primary. In Texas, a July Wedgewood survey showed state Rep. James Talarico leading Attorney General Ken Paxton among Hispanics, 64 percent to 30 percent, the kind of subgroup shock that travels fast through activist media and donor chats.
Talarico’s Senate push already draws national Democratic money and attention, including the kind of Hollywood fundraising swing that signals how seriously national players take the race. Inflated early polls make that cash easier to raise and harder for rivals to dismiss.
Party discipline around that Texas contest remains uneven. Progressive stars hedge while committees decide where dollars actually go, a tension visible when Jasmine Crockett declined to lock in help for Talarico even as the broader map tightens. Junk data only muddies those choices further.
RealClearPolitics noted similar distortion risks in Louisiana, Mississippi, and Maine, not just Michigan. Another outfit, Median Strategies, had already walked a comparable “social experiment” path. The pattern is low barriers, thin vetting, and outsized narrative power once a number hits an aggregator.
Control of Congress is heading into a November showdown. Horse-race models and seat-probability charts feed cable panels, campaign budgets, and voter morale. When fabricated surveys slip into The New York Times and Decision Desk HQ before anyone pulls them, the cleanup comes after fundraising windows close and primary voters have already absorbed the storyline.
Republicans, including President Trump, have long questioned poll veracity. The 2016 cycle still looms as the textbook case of surveys showing Hillary Clinton with an edge that vanished on Election Night. Democrats spent years treating that skepticism as bad faith. This time the loudest fraud charges are coming from their own strategists, after progressive candidates harvested the benefit of numbers that were never real.
White House and Capitol handicapping already assumes turbulent House math. Quiet planning for a possible Democratic majority, including high-level outreach of the sort seen when Jared Kushner sat down with Hakeem Jeffries, only raises the premium on clean data. Fabricated primaries and subgroup shocks are not a harmless prank in that environment.
State election rules still try to police the basics of who votes and how, fights that include New Hampshire’s college-ID ban at the polls. Survey shops face no comparable gate. A crowdfunded brand with polished answers to a vetting form can move national coverage until an independent outlet forces a confession.
Holdsworth’s $80 million outspent gap, McMorrow’s dried-up small-dollar stream, and seventeen polls yanked from a major desk are not abstract lessons. They are campaign facts created by invented respondents. Wedgewood called it a demonstration. The candidates who lost time, money, and volunteers called it something else.
Media brands that rode the pollercoaster now scrub archives and promise tighter checks. The cheaper fix is the one RealClearPolitics already practiced: demand a track record before you print the number. Until that becomes the industry floor, any activist with a spreadsheet and a press list can rent a headline, and far-left campaigns will keep cashing the temporary boost while someone else pays the bill.
Voters already distrust the circus. Firms that fake the scoreboard, and gatekeepers who wave them through, earn every ounce of that distrust.