The U.S. midterm elections turned prediction market platforms into a kind of political casino, where participants bet real money on the victory of individual candidates and parties. Despite the growing popularity of this type of trading, leading platforms Kalshi and Polymarket failed to predict the scale of the Democratic Party’s victory.
This was reported by The Wall Street Journal in an article by journalists Kevin Dugan and Katherine Long, published on November 4, 2026, and distributed in the newspaper’s Japanese edition.
One of the participants was 64-year-old Thomas Morrow, who bet around $1,200, approximately 190,000 Japanese yen, on Republican candidates winning.
Morrow was convinced that his political predictions were correct and viewed the bets as an opportunity to prove his position to his friends.
“This is a great opportunity to clearly prove to my friends that my judgment is correct,” he said on November 3, Election Day.
Just a few hours later, the situation had changed. By the morning of November 4, Morrow had already lost around $640, or more than half of the amount he had initially bet, after Democrats secured a convincing victory in the midterm elections.
Prediction markets such as Kalshi and Polymarket allow users to trade contracts whose value depends on a specific event occurring. In political betting, this could be the victory of a particular candidate, an election result, or control of a given legislative institution.
Unlike traditional opinion polls, in which participants express their preferences, prediction markets require a financial commitment. Contract prices reflect participants’ assessments of the likelihood of a particular outcome.
However, this does not guarantee the accuracy of the predictions. The results of the U.S. midterm elections show that even markets trading significant sums can underestimate the scale of political changes.
According to The Wall Street Journal, many participants use such platforms primarily for entertainment rather than as a professional investment tool.
The popularity of prediction markets also raises the question of how reliable financial bets can be as an indicator of public sentiment. Although they bring together the expectations of many participants, they remain dependent on the behavior, preferences, and financial means of the people who trade.
The case of the U.S. elections also highlights the risks for users who turn their political beliefs into financial bets. Even strong confidence in a particular election outcome does not protect against losses when voters cast their ballots differently than expected.