Jia Lissa
Prediction markets, frequently lauded by proponents as impartial barometers of public sentiment and even "truth machines," are increasingly being scrutinized as recent trading activity on platforms like Polymarket raises serious questions about their integrity. While these markets are often touted for their ability to offer clearer forecasts than traditional polling, exemplified by Kalshi’s accurate prediction of the 2024 US presidential race, some of their trading patterns reveal a more confounding and potentially manipulative landscape, particularly on offshore entities like Polymarket. Recent analyses have uncovered hundreds of millions of dollars in highly unusual transactions, predominantly involving long-shot candidates in global elections, pointing towards sophisticated market manipulation tactics such as wash trading and "activity farming."
The Ethiopian Enigma: A Deep Dive into Anomalous Activity
A striking example of this perplexing activity emerged from two global election markets hosted on Polymarket’s offshore platform, collectively generating over $600 million in trading volume. This figure far surpasses the typical engagement seen in most political prediction markets. What made this volume particularly unusual was its concentration: the vast majority of transactions involved contracts tied to candidates with negligible chances of victory, who ultimately did not win their respective races.
The Ethiopian federal elections, held this past summer, provide a stark illustration. Prime Minister Abiy Ahmed’s Prosperity Party achieved a decisive victory, securing 438 parliamentary seats and paving the way for Ahmed to serve another five-year term. His closest competitor managed to win only 13 seats, underscoring the overwhelming nature of Ahmed’s triumph. In conventional prediction markets, trading volume—the total value of contracts bought and sold—is typically a strong indicator of perceived probability, with frontrunners attracting significantly more activity than long-shot contenders. However, the Ethiopian market deviated sharply from this established norm, presenting an inverted reality that immediately drew suspicion.
Astonishingly, only $173,000 was wagered on Prime Minister Ahmed’s widely anticipated win. In stark contrast, a staggering $41 million was traded on his chief of staff, Shimelis Abdisa, becoming the country’s next leader—an outcome that was, in reality, highly improbable. The total trading volume for the Ethiopian market alone swelled to over $285 million, with nearly all of this immense sum placed on losing candidates. This disproportionate allocation of capital to improbable outcomes challenged the fundamental premise of prediction markets as efficient information aggregators.
Further investigation by experts unveiled an even more peculiar layer to these transactions. Chris Park, a former Polymarket trader who now spearheads MSR Decode, a prediction-market forensic analysis firm, revealed that "Nearly all of it—about $284 million—came from pairs of accounts repeatedly buying and selling with each other." This pattern of self-dealing, where the same entities or closely linked accounts engage in a continuous cycle of buying and selling to themselves, raised immediate red flags among market observers and experts. Rajiv Sethi, an economics professor at Barnard College specializing in prediction markets, unequivocally stated, "It’s obviously some form of volume manipulation." Such seemingly pointless back-and-forth transactions, devoid of genuine market discovery or profit-seeking, underscore a concerted effort to artificially inflate trading metrics.
Polymarket’s official response to these findings was swift but limited. Shana Bautista, the platform’s head of global intelligence and investigations, informed WIRED that her team "immediately flagged this activity, investigated, and blocked the accounts from trading." However, the platform declined to elaborate on critical follow-up questions, particularly concerning the substantial volume these accounts were permitted to generate before any corrective action was taken. This lack of transparency has fueled concerns about the platform’s proactive monitoring capabilities and the effectiveness of its internal controls against market abuse.
The timing of these anomalous trades compounded the strangeness. Polymarket’s operational rules stipulated that the Ethiopian market would not be officially resolved until October, following the prime minister’s formal appointment and swearing-in. This meant the market remained active for several months after Ahmed’s victory was already a widely acknowledged and undeniable fact. For an astonishing 123 days post-election, bursts of seemingly irrational trading persisted, prompting bewildered Polymarket users to post comments such as "THIS NOT RESOLVED?" and "fuckkkk," indicative of their frustration and confusion over the market’s continued open status despite a foregone conclusion. Sethi aptly termed this phenomenon a "zombie market," where trading continues on events that have already concluded or are no longer relevant, further eroding the market’s credibility as a forecasting tool.
The Romanian Red Flag: Wash Trading Allegations Intensify
The Ethiopian case was not an isolated incident. Another market, focused on a 2024 election in Romania, presented "even more dramatic" trading anomalies, according to Sethi. The initial Romanian election was called off due to fears of foreign interference, necessitating a new race several months later. Crucially, instead of closing the original market in light of the cancellation, Polymarket opted to keep it running concurrently with a new market established for the rescheduled election.
It was during this period that Sethi’s sophisticated algorithms, designed to detect potential wash trading on Polymarket, first flagged suspicious activity in the older market. His analysis revealed an alarming statistic: for certain weeks, "99.6 percent of volume in this market… has been wash trading."
Wash trading is a deceptive practice in which individuals or entities simultaneously buy and sell the same asset to create a false impression of high trading volume and liquidity. In reality, no genuine change in ownership or market interest occurs; money simply moves back and forth between accounts controlled by the same party. This manipulative tactic is strictly prohibited on regulated financial exchanges in the United States and is explicitly banned under Polymarket’s own terms of service. Sethi and his colleagues had previously published a study finding widespread evidence of wash trading on Polymarket’s offshore platform, lending further weight to these recent observations.
While wash trading offers a compelling explanation for the irregularities observed in both the Ethiopian and Romanian political markets, the anonymous nature of cryptocurrency transactions on Polymarket’s offshore entity means definitive proof of trader identity remains elusive. Sethi cautions that "What appears to be wash trading isn’t always wash trading," acknowledging the complexities of forensic analysis in decentralized environments. However, the sheer scale and patterns of activity strongly point towards manipulation.
Unpacking the Motives: "Activity Farming" and the Token Lure
To gain deeper insight into the Ethiopian market’s peculiar behavior, MSR Decode conducted a detailed analysis for WIRED. Their findings suggested that the most probable explanation for the anomalous trading was "activity farming." This term describes a strategy where individuals deliberately inflate their trading volumes to position themselves for potential future benefits, rather than to profit from market predictions.
The firm’s suspicion centers on the possibility that traders were motivated by the prospect of a Polymarket cryptocurrency token. In 2025, Matthew Modabber, Polymarket’s chief marketing officer, publicly stated on a crypto-themed podcast that there "would be a token," though specifics remained vague. The crypto industry has a precedent for "airdrops," where new tokens are distributed to early users or those who have demonstrated significant activity on a platform. High trading volume could be a key criterion for receiving a larger allocation of such a token, creating a powerful incentive for artificial activity. Sethi also finds this motivation plausible, suggesting traders could be driven by anticipated "token allocation" to accounts exhibiting higher volumes.
Andre Guettler, director of Ulm University’s Institute of Strategic Management and Finance and a researcher on prediction market structures, confirmed this perspective, stating, "These are textbook examples on how market participants farm a potential airdrop." This behavior is well-documented within the broader cryptocurrency industry, where users engage in seemingly uneconomic actions to maximize their potential earnings from token distributions.
MSR Decode’s analysis of the Ethiopian election market further debunked the idea that sophisticated algorithmic traders were behind the activity. The firm scrutinized over 38,000 "trade loops" that occurred after the election—instances where individual wallets rapidly bought and sold contracts to each other. Their findings were conclusive: over 98 percent of these loops resulted in neither side gaining nor losing money, effectively demonstrating a zero-sum game designed for volume generation rather than profit. The median time that accounts held onto the contracts before reselling them was a mere nine seconds, a pattern inconsistent with genuine market participation aimed at capitalizing on price movements.
Broader Patterns of Concern: Beyond Specific Elections
These types of unusual trading patterns are not entirely new to Polymarket. Earlier this year, Barron’s reported on similar anomalies in markets predicting the 2028 US presidential election. Much of the trading activity in these markets centered on unlikely candidates such as Elon Musk and New York City mayor Zohran Mamdani. Neither individual is constitutionally eligible to run for the US presidency, as they were born outside the country. This further reinforces the idea that some trading volume on Polymarket may be disconnected from genuine forecasting or political analysis, instead serving other, potentially manipulative, purposes.
Polymarket’s primary competitor, Kalshi, has also faced public scrutiny regarding unusual market activity in recent months. Allegations of suspicious activity on its recently launched perpetual futures market for cryptocurrency grew so intense on X (formerly Twitter) that Kalshi felt compelled to publish a blog post explicitly denying claims of wash trading. While Sethi, who extensively researches these markets, agreed with Kalshi that those specific allegations were unfounded, he posited that the unusual volume could be explained by Kalshi’s specific incentive structures for market makers, which inadvertently create opportunities for other traders to profit from small price discrepancies. This distinction highlights the varied nature of market anomalies, some stemming from manipulation, others from unintended consequences of market design.
The Rise of Polymarket: Growth Amidst Scrutiny
Despite, or perhaps in conjunction with, these controversies, Polymarket has experienced explosive growth. The company recently secured an impressive funding round exceeding $1 billion, led by 1789 Capital, a venture capital firm where Donald Trump Jr. serves as a partner. This significant investment has propelled Polymarket’s valuation to an estimated $21 billion. Furthermore, the platform announced earlier this summer that its annualized revenue had surpassed $1 billion. Polymarket’s rapid expansion and success are often measured by its trading volume, a metric that is frequently touted as evidence of active user engagement and market dynamism. This reliance on volume as a key performance indicator potentially creates an environment where artificial inflation of this metric becomes an attractive strategy for some participants.
Regulatory Landscape and Future Implications
The Commodity Futures Trading Commission (CFTC), the federal agency tasked with regulating prediction markets in the United States, declined to comment on whether it is currently investigating these specific election trades on Polymarket. Historically, the CFTC has conducted numerous investigations into suspected instances of market manipulation on Polymarket, including several exclusively reported by WIRED. However, these past inquiries primarily focused on potential insider trading—where individuals trade based on non-public information—rather than the wash trading or "activity farming" seen in the Ethiopian and Romanian markets.
The CFTC has a strong track record of imposing substantial fines for wash trading in other financial sectors. As a former CFTC employee noted to WIRED, "It seems to me that the CFTC would want to know about weird activity in a very large market." The agency’s mandate is to ensure fair and transparent markets, and the scale of the alleged manipulation on Polymarket, particularly given its rapid growth and high valuation, would logically fall within its purview, even if the platform operates offshore. The challenge for regulators, however, lies in jurisdiction and enforcement against entities operating outside direct US regulatory control, complicated further by the pseudo-anonymous nature of blockchain-based transactions.
The implications of these findings are significant for the broader prediction market industry. If platforms designed to aggregate collective wisdom and predict future events are instead being used for volume manipulation and speculative gaming, their fundamental promise as "truth machines" is severely undermined. Such practices erode trust among legitimate users and investors, distort market signals, and ultimately call into question the very utility and integrity of these innovative financial instruments. As prediction markets continue to gain prominence and attract substantial capital, the need for robust oversight, transparent operations, and effective mechanisms to deter and punish manipulative practices becomes paramount to safeguarding their future credibility and ensuring they fulfill their stated purpose.
