Trading volumes for certain products on prediction market platforms Kalshi and Polymarket are raising questions from some industry observers, who fear the numbers may be inflated.
On Polymarket, observers have flagged a quirk on the company's international exchange, which is not overseen by U.S. regulators: Markets that feature multiple contracts will see greater activity on those with lower odds of occurring versus those with higher probabilities. It's seen on markets related to elections, sports and central bank decisions.
Meanwhile, a user on X on Sept. 20 flagged a strange pattern on Kalshi's perpetual futures market for the ether cryptocurrency. Trades of around $5,500 made up a large share of transactions on the contracts, and that 24-hour trading volume was abnormally high compared to actual resting liquidity.
A CNBC analysis found that on Sept. 20, nearly half of the dollar volume traded on Kalshi's ether perpetuals came from trades sized between $5,495 and $5,505.
Both developments create concerns among some observers that the companies are overseeing inflated volumes or, at worst, are experiencing wash trading, in which traders collude to buy and sell an asset to create a false sense of economic activity. Both Kalshi and Polymarket deny wash trading occurs on any of their products. They also deny that the respective patterns represent inorganic activity.
A designated contract market, the type of exchange prediction markets are, "has a duty to protect market integrity by monitoring market conditions, price movements, and volumes in real time for abnormalities," said Tamika Bent, a partner at K & L Gates. "You'd expect DCMs to be surveilling unusual volume and looking into any signs of market disruption." Polymarket is currently raising at a valuation north of $20 billion in the private market after the launch of its U.S. exchange in May, while Kalshi is reportedly in talks to raise funds that would value the company at $40 billion following its perpetual futures product debut in June.
Both have used surging trading volumes as a metric to underscore the exchanges' growing popularity, helping justify those valuations.
But as the companies reportedly explore moves to the public markets as soon as next year, the accuracy of these volume figures is in the spotlight. "To the extent that a material share of reported perpetual volume is manufactured... headline volume and its trajectory may overstate the underlying trading demand on which such a valuation would rest," Andre Guettler, a professor of finance at Ulm University in Germany, wrote in a working paper regarding the patterns on Kalshi's perpetual futures' volume. "This distinction matters most for retail investors, who are the natural buyers of a prediction-market exchange at a public listing." Polymarket's head of revenue and analytics Kyle Gesuelli told CNBC that the popularity of its low-odds contracts isn't from wash trading or inorganic activity.
Rather this is driven by highly active traders, referred to as "sharps," discovering mispricing in various contracts, he said. "It's actually healthy for markets because it brings pricing imbalances back into balance," Gesuelli said.
Kalshi denied last week that it has any wash trading on the platform, stating that it tracked hundreds of users involved in the trades flagged originally on social media. However, experts CNBC spoke to said that the ratio between daily volumes and actual sitting liquidity on the contracts is worrying and represents an inefficient market structure that creates inorganic activity.
Jack Such, a spokesperson for Kalshi, said the company has "zero concerns" about that ratio. The Wall Street Journal reported the Commodity Futures Trading Commission is examining the trades on Kalshi's ether perpetual future contract. CNBC could not independently verify the report.
A spokesperson for the CFTC said that the agency doesn't confirm or deny matters related to investigations. "We have a zero tolerance policy when it comes to any sort of manipulative trading, including wash trading, insider trading, or fraud in our markets," CFTC Chairman Michael Selig said in an appearance on CNBC's " Squawk on the Street " last Wednesday. "When you have new types of markets evolving, you're going to see fraud with that." Low odds draw popularity Barron's reported in April that activity on contracts related to who will win the presidential election in 2028 on Polymarket's international exchange saw unusual activity on low odds markets.
A contract that asked if JD Vance will be the Republican presidential nominee in 2028 had less volume than one for Elon Musk, despite Musk not being eligible for the presidency. Similar strange volume patterns were found on markets related to prospective nominees for the Democrats in the 2028 contest.
That isn't the same on its rival Kalshi's markets, where low-odds candidates for the presidency in 2028 have less volume than candidates with higher odds. Additionally, a CNBC analysis of trades throughout September determined the pattern wasn't detectable on 2028 presidential nominee-related contracts on Polymarket's U.S. exchange, which is regulated by the CFTC.
Unusual activity also occurred on Polymarket international's sports-related contracts. For the 2026 FIFA World Cup, victor Spain had $152 million traded on contracts related to its odds to win the soccer tournament.
Spain's total was less than that of the $158 million traded on Egypt's odds, which never crossed 0.5%, and it was slightly smaller than Morocco's, which never had a greater than 2% chance of winning. Perhaps most striking is a contract that asks who will be the next Prime Minister of Ethiopia.
Prime Minister Abiy Ahmed, who won that election and has 98% odds on the contract, has about $170,000 traded on his chances. Meanwhile, Gedion Timothewos, whose probability has remained below 3% for months, has almost $56 million traded on his odds.
The contract is still active, even as the election passed in June. "Those are ridiculous long shots," said Joe Saluzzi, head of equity market structure research at Themis Trading. "It sounds like someone's trying to put some very, very low risk volume up there…
Why would anybody be trading that?" A CNBC analysis shows volume for the Ethiopian election jumped over 6.7 times from June 21, when reports confirmed Ahmed won the election, to Sept. 25, according to data pulled from Dune Analytics. The highest reported daily volume for the market was on July 30, recording over $15.3 million, over a month after reports noted who won.
A Polymarket spokesperson told CNBC that the contract will resolve when the government elected this summer is officially sworn in. That is slated for Oct. 5.
Gesuelli said that the low odds contract activity is more common on the company's international exchange because it has a greater number of sharps, who often are using complicated software and algorithms for their trading activity to make profits off of small mispricings. The U.S. exchange is more dominated by casual retail traders, he said. Worries about wash trading on Polymarket aren't new.
A study first released in November 2025 by researchers at Columbia University found that trading patterns it determined were indicative of wash trading accounted for 60% of Polymarket international's weekly volume in December 2024, though it declined to 20% in October 2025. Lead author Allen Sirolly said the measure fell to a negligible amount by April 2026.
A Polymarket spokesperson said expanded surveillance and the introduction of fees to the exchange have weakened chances for market manipulation. But Sirolly said the lingering popularity of low-odds contracts is concerning.
Many prediction market observers have chalked up the strange discrepancies on Polymarket's international exchange to the hints it has given in the past to conducting a cryptocurrency token "airdrop." Jay Maliava, co-founder and CEO at prediction market trading term