Prediction markets, such as Polymarket and its main rival Kalshi, have flourished during Trump’s second term.
Gabby Jones/AFP via Getty Images
hide title
Gabby Jones/AFP via Getty Images
Polymarket has investigated “dozens” of bets in its prediction markets related to the midterm elections, according to the company’s head of global research and intelligence, Shana Bautista.
At least two of these Polymarket investigations have been referred to its regulator, the U.S. Commodity Futures Trading Commission, according to a person familiar with the investigations but not authorized to speak publicly about them.
These investigations into insider trading and insider manipulation in elections come at a time when prediction market Polymarket (and its main competitor, Kalshi) has skyrocketed in popularity over the past year. But the fast-growing industry also faces significant legal challenges from states and political pushback from lawmakers who argue that betting on the outcome of elections poses serious risks to the democratic process.
In April, Kalshi banned and fined a handful of political candidates for betting on themselves. A month later, NPR reported that campaign staff had made “thousands” by using insider polling information to bet on their own candidates on platforms like Polymarket and PredictIt.
Polymarket’s Bautista, a former FBI official and cryptocurrency analyst, told NPR that his 16-person team and a new election surveillance system are actively monitoring election betting ahead of the Nov. 3 midterm elections.
“If you have information that you have a duty to protect, don’t deal with it,” Bautista said. “It’s explicitly against our terms of use on both platforms, but it’s also against the law. If you decide you’re smarter and can do it, we’ll find you.”
Bautista joined Polymarket in June. She oversees information collected from both the U.S.-regulated version of Polymarket, which requires users to provide their real identities, and its older brother, Polymarket International, which is essentially unregulated and where trading can be done anonymously and in cryptocurrency. Despite this, Bautista said cryptocurrency transactions are rarely completely anonymous.
“There has never been a case where we haven’t been able to identify, potentially, who these people are,” Bautista said.
It said its election surveillance system pulls “gigabytes” of data on each social media user, data from the Federal Election Commission, cryptocurrency companies, banking partners, statistical models and other sources of public information. If a user encounters a certain number of internally defined red flags, Bautista receives an alert on his phone.
“We receive hundreds of alerts a day,” Bautista said.
When asked about specific detection criteria and thresholds, Bautista did not comment citing the security of the system. But he said insider trading and market manipulation look the same regardless of the financial market.
“We are looking at threat vectors and risk areas that have existed for decades in traditional markets,” Bautista said. “But we also recognize that when it comes to prediction markets, we’re looking at some of the old tried and true tactics, but being applied to a new vector through prediction markets.”
In May, Kalshi, Polymarket’s main rival, announced its own election surveillance system that also uses FEC data. Their system supposedly prevents any campaign staffer on the FEC list from betting on an election. In July, Kalshi said he had blocked “dozens” of employees from trading in the election, although at least one employee told NPR that they managed to fix this problem. The employee shared records of his recent exchanges and spoke to NPR on condition of anonymity for fear of the consequences for his future employment.
Prediction markets for elections are especially vulnerable to market manipulation and insider trading, said Rajiv Sethi, an economics professor at Columbia University’s Barnard College who has studied prediction markets for decades.
“[Election prediction markets] “They have the potential to be a self-fulfilling prophecy,” Sethi said. “Elections have the property that what you think is going to happen can actually affect what happens if you think that a candidate is losing momentum or that the campaign is faltering.”
Bautista said his Polymarket team and surveillance system are learning from past cases and putting their feet in the shoes of potential insider traders and market manipulators.
“If I were concerned that someone was trying to materially move an electoral market to make it look like one option is more likely to happen, I identify a pattern where they open several different accounts and do so in short bursts and I see device IDs consisting of different trades placed in short succession with new wallets in five or four different accounts,” Bautista said. “If you worked that case and then acted, the findings from that tactic would go back into the process to create the detections.”
As prediction markets have grown in popularity over the past two years, the industry’s federal regulator, the CFTC, has done little to police these new financial markets, largely leaving that job to the companies themselves.
