Polymarket Probes Late Bets on Israel's Election

Polymarket is investigating suspicious, profitable trades placed shortly before polls closed in the Israeli election, The Wall Street Journal reported on 29 October 2026. The bets cluster around one outcome: that Benjamin Netanyahu and his Likud party would perform better than their opponent.

The activity stands out for its timing and its source. According to Bloomberg, hundreds of new accounts placed bets on Netanyahu and Likud's performance in the final stretch before polls closed. Quartz reported that some of the traders involved cashed out more than $100,000 in total.

Brand-new accounts that all lean the same way in a narrow window draw scrutiny in any market. Coordinated activity of that kind is hard to explain as many people independently reaching the same view.

For readers who do not trade these contracts, the mechanics matter. A prediction market contract pays a fixed amount if a specified outcome occurs and nothing if it does not. Its price, quoted between zero and one, is read as the market's implied probability of that outcome. A trader who buys a contract the market has priced as unlikely, and who turns out to be right, collects the full payout against a small outlay. That asymmetry is the attraction, and it is also why a cluster of well-timed buys can look like information rather than opinion. The open question is whether anyone traded on knowledge the public did not have, or whether the timing was coincidental.

The election was close enough to make late money look meaningful. The main challenger to Netanyahu was Gadi Eisenkot, according to The New York Times. NBC News reported that exit polls projected no clear winner, with neither Netanyahu nor Eisenkot gaining a parliamentary majority. The seat count was still moving. Earlier projections had Netanyahu's bloc at 53 seats, against the 68-seat coalition he led in the outgoing Knesset, per CNN. With more than 84 percent of ballots counted, Al Jazeera reported a projection of 54 seats for Netanyahu's bloc and 49 for Eisenkot's alliance.

That gap is the reason a market on this election could move sharply in its final hours, and the reason the trades are hard to dismiss as noise. Thin, fast-moving markets on a contested vote can reward anyone who acts before the price catches up, whether that person has inside information or simply a faster read of the count. Separating those cases requires account-level data the platform holds and the public does not, which is why an internal investigation, rather than outside commentary, is where the answer will come from.

For anyone pricing political risk, the episode is a reminder that prediction markets carry their own integrity questions. Their prices are only as informative as the trading behind them. If a small group of accounts can move a probability late in a vote, the market is measuring that group as much as the electorate. The investigation will show whether these trades were an anomaly on one platform or a weakness that other event markets will need to address.