A series of fabricated election polls that circulated in competitive races across the country is raising questions about whether false data influenced campaign fundraising, donor confidence, and perceptions of candidates’ chances in primary and general elections.

Wedgewood Polls admitted Tuesday to falsifying survey data after election analysis website FiftyPlusOne uncovered irregularities in its research, prompting polling aggregators to remove its findings. Stay informed for what's next in the midterms.

Get unlimited access to Washington Examiner's midterm coverage, analysis, and political insights. Already a member?

The surveys covered Senate and gubernatorial contests across the country, including several races that could determine control of Congress. The fabricated numbers appeared on prominent tracking websites and social media pages and were cited by national news outlets, giving them exposure beyond Wedgewood’s own publications.

FiftyPlusOne’s investigation focused on a Mississippi Senate survey after researchers identified inconsistencies in respondent and transaction identification numbers. Wedgewood subsequently acknowledged that its polling was fabricated.

The revelations raise questions about how easily false polling can influence political narratives and campaign decisions, particularly as candidates rely on surveys to gauge which races are competitive.

Michigan: McMorrow’s polling upset Michigan offers one of the clearest examples of how the fabricated polling could have affected a campaign.

A Wedgewood survey at the end of June of the Democratic Senate primary showed Abdul el Sayed leading with 46% support, followed by Rep. Haley Stevens (D-MI) at 38%, and Mallory McMorrow at just 9%. The figures appeared on polling trackers, including at the New York Times, as the candidates competed for the nomination.

McMorrow trailed significantly behind el Sayed and Stevens in other polls, but a large portion of respondents were undecided. While there was no guarantee the bloc of undecided voters would flock to McMorrow, having 20% to 40% of voters not settled on a candidate should indicate the race is far from decided.

Additionally, hypothetical Wedgewood matchups of Democratic candidates against Republican hopeful Mike Rogers showed McMorrow putting up less of a fight, losing by 7 percentage points. El Sayed, however, led Rogers in the potential general election by 5 percentage points.

W. JAMES ANTLE III | NEW FAKE POLLING SCANDAL RILES DEMOCRATS AHEAD OF MIDTERM ELECTIONS The poll showing McMarrow falling behind appeared in the New York Times, VoteHub, FifityPlusOne, and Decision Desk HQ.

Other surveys from the same time period painted a much tighter race and showed McMorrow putting up more of a fight. A Quantus Insights poll showed McMorrow trailing Rogers by just 1 point in a hypothetical general election race.

McMorrow said the Wedgewood survey contradicted her campaign’s internal polling and undermined her fundraising efforts. She also said volunteers became discouraged and that reporters questioned whether she would leave the race. She ultimately suspended her campaign July 5.

Mississippi: Colom campaign touts competitive numbers In Mississippi, Wedgewood published an October survey showing Democrat Senate hopeful Scott Colom gaining ground on incumbent Sen. Cindy Hyde-Smith (R-MS), with the latter leading by just 4 percentage points, 52% to 48%. The poll did not include undecided voters.

The numbers appeared in coverage by the Magnolia Tribune and were promoted by Colom’s campaign, which touted the race as “tied.”The survey followed Sabato’s Crystal Ball moving the contest from “Safe Republican” to “Likely Republican.”Earlier polling, however, painted a slightly different picture. A survey conducted by a Colom-affiliated PAC showed Hyde-Smith leading by 6 percentage points in a 47%-41% split, with 12% of respondents undecided.

Senate Majority PAC, which is aligned with Senate Minority Leader Chuck Schumer (D-NY), moved forward with a $1.2 million ad buy in a state that was viewed as a shoo-in for Republicans days after the bogus poll was released.

The findings could have strengthened the case for Democratic investment in a state where the party has historically struggled in statewide elections. However, Colom’s campaign also cited separate polling and strong fundraising numbers, and there is no evidence establishing that Wedgewood’s survey changed investment decisions.

The Mississippi survey ultimately became the subject of FiftyPlusOne’s investigation, which exposed the fabricated data.

Maine: Fabricated polls reinforce Collins’ vulnerability In Maine, Wedgewood published two surveys showing Democrat Senate candidate Troy Jackson leading incumbent Sen. Susan Collins (R-ME), adding to an already difficult election environment for the five-term incumbent.

A July survey placed Jackson ahead by 5 percentage points, 48% to 43%, just days before then-nominee Graham Platner suspended his campaign after multiple women came forward with sexual abuse allegations.

Wedgewood released an additional survey from the same time period showing Collins leading Platner by 4 points. At the time, Maine Democrats were considering who would replace Platner on the ballot.

Other polling from the same time frame showed the Platner leading by 4 points. Additional surveys released by Platner’s campaign just prior to his exit showed Jackson leading Collins by 5 points. A University of New Hampshire poll conducted roughly one week after the Wedgewood poll showed Jackson leading Collins 49% to 46%.

While Jackson had been viewed as the candidate to fill Platner’s shoes in the closely watched race, the timing of the polls is notable, considering no other candidates seeking to replace Platner were considered by the fake pollster. Platner’s campaign sponsored polling that included Jordan Wood, Sheena Bellows, Janet Mills, and Nriav Shah as his potential replacements.

The fabricated results could have reinforced an existing narrative about Collins’s vulnerability rather than introducing an unexpected shift in the race.

Wisconsin: Hong held lead before primary loss Wisconsin’s Democratic gubernatorial primary offers another example of how Wedgewood’s fabricated surveys could have shaped perceptions of a race that ultimately produced a major upset.

Wedgewood published two surveys in July showing state Rep. Francesca Hong, a socialist, leading the Democratic field. The first, shared July 5, placed Hong at 30%, former Lt. Gov. Mandela Barnes at 28%, and Lt. Gov. Sara Rodriguez at 19%.

A second survey, released July 21, showed Hong expanding her lead to 38%, followed by Barnes at 30% and Milwaukee County Executive David Crowley at just 9% after reviving his campaign.

2026 MIDTERM ELECTIONS LIVE UPDATES: LATEST NEWS, RESULTS, AND ANALYSIS The findings appeared on the election-tracking websites 270toWin and PollingSource, and the earlier survey was promoted by conservative MacIver Institute. Hong’s lead was not exclusive to the fabricated surveys, as legitimate polls showed her closing in the week of the primary, although the surveys underestimated Crowley’s support.

Crowley ultimately defeated Hong in the primary, receiving roughly 39.8% of the vote to Hong’s 39.3%.

Kansas and Ohio: Unexpected Democratic jumps Some of Wedgewood’s fake polls suggested Democrats were gaining unexpected ground in Republican-leaning states, potentially creating a misleading impression of races that warranted additional attention and financial investment.

In Kansas, a Sept. 24 survey showed Democratic Senate candidate Adam Hamilton leading incumbent Sen. Roger Marshall (R-KS) by 2 percentage points, 50% to 48%. The findings were particularly striking in a state that has not elected a Democrat to the U.S. Senate since 1932.

The Senate Leadership Fund, a GOP-affiliated PAC, pumped $8 million into the state to support Marshall shortly after the poll was released.

The same survey showed Republican gubernatorial hopeful Ty Masterson leading Democrat Cindy Holscher by 4 points, suggesting a closer contest for the governor’s mansion.

In Ohio, an Aug. 14 Wedgewood survey showed Democrats leading both major statewide contests. Former Democratic Sen. Sherrod Brown held a 4-point advantage over Sen. Jon Husted (R-OH), 48% to 44%. The race had already been considered competitive, but earlier polling showed a much tighter race with Husted holding a slight advantage.

Other fabricated polls reflected legitimate polling trends Not all of Wedgewood’s bogus surveys presented surprising results. Several closely resembled findings from established polling organizations, allowing the false data to blend into the broader election polling landscape.

In Iowa, an Aug. 30 survey showed Rep. Ashley Hinson (R-IA) leading Democrat Josh Turek by 2 percentage points in the Senate race, 50% to 48%. Other polls showed Hinson leading by a slightly wider margin, but the race is considered a toss-up. The same survey placed Democrat Rob Sand 5 points ahead of Republican Zach Lahn in the gubernatorial contest, similar to a Suffolk University poll that showed Sand leading by 4 points.

In Texas, Wedgewood’s Aug. 1 survey showed Democrat James Talarico leading Republican Ken Paxton by 2 points in the Senate race, closely matching a July survey that placed Talarico ahead by 4. The fabricated poll also showed Gov. Greg Abbott (R-TX) leading Democrat Gina Hinojosa by 6 points.

An October Wedgewood survey in Georgia showed Sen. Jon Ossoff (D-GA) leading Republican Mike Collins by 10 points, while Republican Rick Jackson held a 2-point advantage over Democrat Keisha Lance Bottoms in the gubernatorial contest. Other polling similarly showed Ossoff ahead and a competitive governor’s race.

In Nevada, Wedgewood placed Gov. Joe Lombardo (R) ahead of Democrat Aaron Ford by 3 points, reflecting the closely contested race.

Although fabricated, the surveys largely reinforced existing perceptions of the contests, potentially making the false findings less conspicuous when published alongside legitimate polling.