Prediction markets, primarily Polymarket, have become an arena for a new class of traders whose operations raise serious questions. Since the beginning of the year, the volume of transactions with signs of insider trading on this blockchain platform has reached nearly $200 million. This refers to 34,000 suspicious operations identified by the Polysights service between August 2025 and June 2026. All of them passed through the global platform, where every bet is visible on the blockchain, making analysis transparent but not always unambiguous.

How "Insider Trading" Works on Prediction Markets

The story of 21-year-old TikToker Cayden Booth is illustrative. In February, he arrived at the Super Bowl stadium 12 hours before the game, timed the duration of the anthem rehearsals, and bet over $50,000 that the performance would last 117 seconds. The anthem lasted 104 seconds — the bet paid off. Booth faced accusations of "insider trading," although he formally violated no rules. His case demonstrates how blurred the line itself is: some see this as resourcefulness, others as a threat to market integrity.

Data on truly suspicious trades looks far more serious. Polysights evaluated each operation based on eight criteria, including bet size, account age, odds at entry, and volume concentration on individual events. This approach identifies atypical behavior but proves nothing. Some participants may have simply been better prepared for the event or taken a successful risk.

Geopolitics as a Driver of Suspicious Activity

The main source of the surge in activity was bets on geopolitics. The daily volume of suspicious trades peaked in late February amid events around Iran. Winnings are distributed extremely unevenly: more than half of all money went to 1% of the most profitable wallets, with 57% of them created less than a day before the trades.

Insiders are acting with increasing sophistication. A group of 38 linked addresses bet on U.S. actions in Iran and Venezuela with 98% accuracy, earned $1.6 million, and withdrew funds through a single Coinbase account. This is no longer a "lucky guess" but systematic work with access to confidential information.

Reaction from Platforms and Authorities

Some stories have already reached the courts. In the U.S., charges were brought against serviceman Gannon Ken Van Dyke — according to the investigation, he earned over $400,000 on Polymarket using secret data about the operation in Venezuela. A similar case was opened in Israel: authorities accused a reservist and a civilian of betting based on classified military information about the country's actions in Iran.

The Kalshi platform has not remained in the shadows. It reported to the regulator about a White House teleprompter operator who allegedly profited from bets on Donald Trump's speeches. Earlier, Kalshi had already flagged congressional candidates who participated in bets on their own elections.

U.S. authorities are also imposing restrictions. The White House sent a warning to employees in March, and the Senate unanimously banned such bets for its members and staff in April. Goldman Sachs management closed employee access to prediction markets — exceptions were made only for sports and entertainment bets.

Polymarket tightened its rules in late March. The platform banned transactions using stolen information and illegal insider data, and also provided law enforcement with information on nearly 100 wallets.

My Analyst's View

Prediction markets are a powerful tool for information aggregation, but their anonymity and global nature make them vulnerable to abuse. The problem is not the technology, but the lack of clear rules of the game. Until platforms and regulators develop unified standards, we will see more and more such stories. Insiders will always look for loopholes — the market's task is to make their search as costly and risky as possible.