2026 Election Rumor Rundown

This blog post was written by the Election Rumor Research team at the University of Washington’s Center for an Informed Public and crossposted on the CIP’s Election Rumor Rundown Substack newsletter.

Key Takeaways

• On Tuesday, Wedgewood Polls released a statement admitting that their polling was fake and justifying it as part of a social experiment to draw attention to low oversight in polling.
• News of this admission — the second case of “fake polls” this cycle — was quickly amplified and wrapped into a number of different rumors and narratives. Conversation and framing contests around this news continue.
• Partisan actors across the spectrum have used the revelation to declare that polls were underrepresenting support for their parties and potentially causing voter suppression for their candidates.
• One risk around this discovery is that the existence of fake polls will fuel distrust in election polling across the political spectrum and could be used to bolster “rigged election” claims after election results are announced.

Figure 1: Cumulative graph showing the number of X posts about Wedgewood and “fake polls.” Individual posts are plotted on the curve at the minute they were posted, sized by # of reposts. The graph shows how the narrative gets started in the mid-afternoon (PDT) on October 20, takes off that evening (more detail on the red box in Figure 3 below), before fading down overnight, and then surges again the next day with help from right wing influencers and GOP political elites. https://faculty.washington.edu/kstarbi/SpotlightViz.html

On Tuesday, October 20, at 3:19 p.m. PDT, Wedgewood Polls — a small polling firm that had launched in June 2026 — released a statement on X, announcing their polling data was fake. In their statement (Figure 2), they framed their work as a “social experiment” to demonstrate the proliferation of low quality political polls, and called for “major reform” to the polling industry. They also called out specific media outlets (MS NOW) that had reported on their polls and other polling organizations (The New York Times, VoteHub, and FiftyPlusOne) that Wedgewood alleged had included their polls in their polling aggregators.

The admission by Wedgewood (whose X account has now been deleted) was allegedly in response to an exposé piece published by forecasting and polling aggregation outlet FiftyPlusOne, who had been posting that they had found inconsistencies in the methods and data used by Wedgewood Polls.

This Election Rumor Deep Dive explores the context, major rumors, and implications of the revelation around Wedgewood Polls.

Figure 2: A press statement from Wedgewood Polls shared in a since-deleted October 20 X post says their operation is “an experimental project to demonstrate the vulnerabilities, lack of oversight, and overall low barrier to entry in the modern polling industry.”

This is not the first, nor likely to be the last, case of fake polls being filtered into political conversations. In August, another company called Median Strategies released fake polling data that Los Angeles Mayor Karen Bass touted as evidence that her re-election campaign was “gaining momentum.” Once the poll was discovered to be fake, the company released a statement quite similar to the excuse Wedgewood used: the fake poll was allegedly a “social experiment” designed to test how polling information could enter the political conversations without independent verification.

Wedgewood and Median Strategies are not the only recent cases. At the time of writing, there have been reports that another polling company, SoCal Strategies, has removed their entire online presence after investigation by FiftyPlusOne. This is notable in part because individuals alleged to be behind Wedgewood Polls claimed prior affiliation with SoCal, suggesting that SoCal may also be a fake pollster.

Fake polls have real impact. A state senator in Michigan vying for the U.S. Senate blamed a Wedgewood poll for impacting her ability to raise money, causing her to drop out of the Democratic primary race. As the Brennan Center has noted, fabricated polls have been and can be used to inflate or deflate a party’s perceived strength, and can potentially lay the groundwork for challenging the legitimacy of an unfavorable result.

The impacts of fake polls don’t just impact candidates or audience perceptions of election outcomes, they also impact the credibility of news outlets and poll aggregators themselves, potentially undermining the perceived legitimacy of real poll numbers. In the Wedgewood case, The New York Times countered they did not use any of Wedgewood Polls’ data in their own polling and forecasting because they work with a reputable pollster, but had just listed a few of their polls on its tracking pages — not integrating it into their own forecasting averages. Despite this defense, even small amounts of amplification can be used by political opponents to cast doubts on legitimate polling, further sowing distrust about polling and election outcomes more generally.

Figure 3: A zoomed-in version of the cumulative graph, featuring the first few hours of conversation. Posts from highlighted accounts (white circles with usernames) are featured in the write-up below. Interactive version of this graph is available here: https://faculty.washington.edu/kstarbi/SpotlightViz.html

The Polling Community Makes Sense of the ‘Fake Poll’ Confession

Early X conversations around the confession featured several accounts dedicated to conducting or reporting on political polls. At 3:22 p.m. PDT, just three minutes after Wedgewood’s confession, @PollTracker2024 posted a copy of Wedgewood’s statement along with this: “So it turns out Wedgewood polls is just a bogus polling firm.” That post catalyzed a series of reactions from others in the community.

Some claimed that they had prior suspicions about Wedgewood. Others voiced anger about what they had done. A few began to examine their reporting alongside other data to speculate about whether their fake polls had impacted other polls or betting markets. Zachary Donnini from the independent forecasting outlet, VoteHub, bemoaned the situation and described on X how he had phoned the allegedly college-aged Wedgewood forecasters earlier in the summer, extending them the benefit of the doubt given many forecasters start as hobbyists.

Not long after Wedgewood’s X post and confession, at 4:23 p.m. PDT, G. Elliott Morris, one of the co-founders of FiftyPlusOne, posted to X (Figure 4), quoting Wedgewood’s now deleted X post, about an upcoming story on Wedgewood.

Figure 4: A screenshot of G. Elliott Morris’s October 20 X post announcing FiftyPlusOne’s investigation into Wedgewood Polls.

Morris’ X post questioned Wedgewood’s “social experiment” excuse. He included a link to FiftyPlusOne’s reporting, which explained how he and his colleagues had, in recent days, been closing in on Wedgewood for manufacturing polling data. The article also noted some of the damage that Wedgewood had done, explaining how its fabricated polling data had been amplified by established media outlets and prominent election forecasters, influencing election reporting and predictions.

Right-Wing Influencers Amplify the Controversy

Early conversations about Wedgewood and its fake polls were mostly concentrated among polling experts and political wonks. About two hours after the confession, the story began to get a boost from right wing influencers. At 5:29 p.m. PDT @Geiger_Capital, a verified X account with more than 300,000 followers, published the first of two high engagement posts drawing attention to the case. This post went viral, receiving 656,000 views, 4,800 reposts, and 17,000 likes (Figure 5).

Figure 5: A screenshot of Geiger Capital’s October 20 X post amplifying Wedgewood Polls’ press release, emphasizing that The New York Times and Nate Silver had posted the fake polls.

Geiger Capital’s tweets specifically called out The New York Times and prominent poll expert Nate Silver for allegedly integrating Wedgewood’s data into their own aggregate polling statistics. A follow-up post by Geiger Capital stressed that left-leaning media outlet MS NOW had also amplified the polls. Geiger Capital’s posts promoted a perspective on the story that soon became prominent on the right: that the existence of this fake polling data and its integration into other polling data suggest that the polling industry cannot be trusted. (Interestingly, this echoed the framing from Wedgewood’s confession.) Later, others would extend this framing to suggest that polling numbers, especially those that show Democrats leading Republicans in many U.S. House and Senate races, are not accurate.

Shortly after @GeigerCapital’s two posts, an account called @DSA_Watch, picked up the story and alleged in an X post — without evidence — that Wedgewood was “pro Abdul El-Sayed” and that it had falsely inflated his numbers to support his candidacy. About 20 minutes later, right-wing content creator Eric Daugherty, posted the first of two Wedgewood-related messages. This one highlighted recent numbers from the fake polling firm that suggested Democrats were leading in red and purple states, suggesting that these — and perhaps other similar polling numbers — were not trustworthy.

Figure 6: A screenshot of an October 20 X post from Eric Daugherty highlighting where “FAKE POLLSTER” Wedgewood Polls had shown Democrats ahead in important midterm races.

About 15 minutes after that, Polymarket’s X account posted about the Wedgewood fake data to its 2 million followers, reaching the biggest audience thus far. Overall engagement in the story began to surge at that time and didn’t taper off until much later that evening. The story resurged the next morning with some help from Fox News and, on X, from right-wing influencer @GuntherEagleman (Figure 7), an account we’ve seen receive a lot of engagement this cycle around election rumoring.

Figure 7: A screenshot of an October 21 X post from Gunther Eagleman amplifying a Fox News video discussing Wedgewood Polls and its potential impact on GOP fundraising and volunteers.

Within 24 hours, the right-wing media ecosystem had succeeded in drawing attention to the story and setting some of the most salient frames: that poll numbers around the midterms, especially poll numbers favoring Democrats, cannot be trusted. Later, more conspiratorial narratives would emerge alleging that polls were being intentionally faked to boost Democrats to set the stage for the left to contest elections results that go in the Republican’s favor.

Republican Political Elites Respond by Leaning In

The day after the Wedgewood announcement at 7:26am (PDT), Federal Communications Commission (FCC) Chairman Brendan Carr leaned into the story, publishing an X post noting a “surge in entirely Fake Polls” and threatening broadcasters who are found to have shared data from these fake polls. Initially, there was some confusion among people outside the right wing mediasphere and polling communities about what had motivated Carr’s statement. However, our data and this timeline suggest his announcement was directly downstream from Wedgewood’s confession and its amplification on the political right.

Later in the day, U.S. House Speaker Mike Johnson leveraged the Wedgewood confession to levy a broader criticism against political polls, alleging that Republicans are being under-sampled with the potential to suppress GOP turn-out, and ending with encouragement to Republican voters to ignore the polls and go vote (Figure 8).

Interestingly, Johnson’s post highlights a Fox News article featuring a quote from a Democratic strategist who alleged that Wedgewood’s fake polls had boosted far-left candidates against centrist Democrats during the primaries.

On late Thursday night, the controversy caught the attention of President Trump, who posted on Truth Social that the “failing New York Times is ‘ARTIFICIAL’” for posting the fake polls (Figure 9).

Figure 8: A screenshot of U.S. House Speaker Mike Johnson’s October 21 X post suggesting that mainstream media is “delighted” that Wedgewood Polls could depress GOP turnout.

Figure 9: An October 22 screenshot of a Truth Social post from President Trump discussing Wedgewood Polls.

The Center-Left Reacts With Suspicions About Impacts on Primary Races

A narrative that Wedgewood’s fake data had impacted primary races began to emerge from parts of the political left. On Tuesday evening, Mallory McMorrow, a Michigan state senator and former Democratic candidate for U.S. Senate, claimed in an X thread that the data from the Wedgewood polls was likely a catalyst for the suspension of her campaign, as funds and support for her campaign dried up, claiming “it killed us” (Figure 10). We saw this rumor both on X and Bluesky, as an example of the real life impacts of using polling data for decision making during this election cycle.

Figure 10: An October 20 screenshot of an X post from Mallory McMorrow, former Democratic primary candidate for U.S. Senate this year, discussing the impact Wedgewood Polls had on her campaign.

In response, others on the center-left questioned how Wedgewood’s fake data might have boosted further-left candidates during the primaries, therefore negatively impacting Democrats’ electability in the general election — potentially as part of an intentional Republican strategy. For instance, Amy Siskind, an activist, writer, and early supporter of Hillary Clinton, speculated on X about how Wedgewood polling might have caused Democrats to run “the least electable of the three candidates running in the primary in Michigan.”

Allegations of Prediction Markets Insider Trading

Conversation around the Wedgewood case also included speculation that individuals behind the fake Wedgewood polls were likely using them as part of a prediction market scheme. On Bluesky and X, an account called BeshearStan claimed to have found a Kalshi account of an individual purportedly connected to Wedgewood, posting a screenshot of an account that was up $4764 in profits, claiming that it had been “insider trading using the fake polls” (Figure 11). So far, there has not been any reported confirmation that this account was linked to Wedgewood Polls. On X, one user remarked that if it were indeed a Kalshi account from Wedgewood Polls, the account had failed to benefit from its own insider trading. Two days later on October 22, Kalshi issued a press release stating that Wedgewood’s releases and admission of fakery did not meaningfully impact any markets. So far, these allegations appear to be just speculation, but do underscore anxiety about the financial incentives to fake polling posed by potential manipulation of prediction markets.

Figure 11: A screenshot of BeshearStan, a Bluesky poster, claiming to have found Wedgewood’s Kalshi account, showing a profit.

Between now and Election Day, we may see more fake polls emerge and/or be identified. Generative AI tools have made fabricating relatively realistic looking survey data and methodology statements cheap and easy; FiftyPlusOne flagged that the data reported by Wedgewood was likely AI generated, a claim that seems to have been buried under more popular rumoring around polls downplaying Republican voter enthusiasm. Social media allows toplines and out-of-context screenshots of polling results to circulate widely before anyone verifies whether the underlying source and methodology are legitimate.

Moreover, the rise of online prediction markets introduces new financial incentives: while there is so far limited evidence that any of the exposed fabricated polls meaningfully have moved betting markets, the prospect of making money off of a thinly-traded race by releasing a well-timed fake poll provides a motive beyond merely garnering attention.

The Wedgewood Polls case has shown that some of these polls can penetrate the political information ecosystem beyond social media, making their way into mainstream polling aggregators, being cited by campaigns in their messaging, and even affecting campaign fundraising and morale. This risk becomes more acute in the final stretch of the election cycle, when demand for data from both media and campaigns is high and there is less time left for journalists, aggregators, or campaigns to vet new releases before votes are cast.

Election officials, journalists, and campaigns should also watch for the opposite dynamic: documented cases of fabrication can be used to cast doubt on legitimate polling or on legitimate election results. Here it helps to separate three categories of polls that are easily, and sometimes deliberately, conflated:

• fake polls like those released by Wedgewood Polling that rely on entirely fabricated data;
• low-quality polls that were actually conducted but rely on weak methodology; and
• legitimate polls with unwelcome results for particular candidates.

After the election, the conflation among these categories could allow gaps between earlier polls and election results to be framed as evidence that the outcomes of particular races are suspect, or at worst, part of larger election fraud conspiracies.