On-chain review weakens, but does not settle, Senate-bet insider claims

Blockchain and trading-record analysis has weakened, but not resolved, allegations that a bettor used advance knowledge of polling to profit from U.S. Senate prediction markets in Nebraska and Michigan.

The scrutiny centred on wagers made shortly before New York Times/Siena polls were released. The bettor put about $191,000 behind Republican Pete Ricketts in Nebraska, generating more than $77,000 in profit, and about $623,000 behind Democrat Abdul El-Sayed in Michigan, generating more than $180,000.

The timing had been flagged by Vanderbilt University researchers, as we reported in October. But The Block’s review, described by Bloomingbit and Koin Bulteni, linked Polymarket activity to the public account “dubdubdub2” and found a broader pattern of large, directional wagers that were held through to election outcomes.

Its matching of Kalshi deposits and trading records also suggested that a large Nebraska position may have been placed about 10 minutes after the relevant poll was published. The conclusion is necessarily tentative: Kalshi’s public records do not identify the account behind an individual trade.

The linked wallets had traded for roughly two years and moved tens of millions of dollars through decentralised-finance platforms including Aave and Pendle. They also showed earlier large positions in the New York mayoral race and the Super Bowl, as well as about $2.99 million deposited into Hyperliquid before the Senate bets, followed by a loss of roughly $783,000 over 26 hours.

Rajiv Sethi said the pattern looked more like wagers on ultimate election results than attempts to capture brief poll-driven price movements. Vincent Grégoire said the bets could have reflected a proprietary forecasting model, private polls or other information. Both said insider information could not be definitively excluded, and on-chain records alone cannot establish what informed a trade.

Polymarket said it was investigating the trades, while the New York Times and Siena Research Institute said they would review the allegations. The episode adds to wider questions about prediction-market surveillance: Polymarket had already investigated dozens of election-related bets and referred at least two cases to the Commodity Futures Trading Commission, according to OPB.