Why prediction markets like Polymarket and Kalshi often fail the average investor

By Mark Hulbert

Big traders don't always know best

There is a 19% probability that the Dow Jones Industrial Average DJIA will eclipse 59,000 at some point between now and the end of the year, according to a contract that trades on the Kalshi prediction market.

That represents a 23% gain over the DJIA's current level.

Is this probability too high or too low? On the one hand, President Donald Trump recently predicted that the stock market "is going to go through the roof." On the other hand, the stock market is extremely overvalued, and, year-to-date, it has already far exceeded the average return for midterm election years.

Many fans of prediction markets believe that they are the best gauge of probabilities, such as how high the DJIA will rise this year. But a recent study out of the University of California, Berkeley, calls this belief into serious question. It found that prediction markets are unduly influenced by their biggest traders, and these so-called "whales" do worse, on average, than those who place smaller bets.

The study, "How Wise is the Crowd? Bias and Edge in Prediction Markets," was conducted by five researchers at UC Berkeley: Avaneesh Deleep, Dhruv Suresh, Jenny Bai, John Lee and Harsh Dhawan. They reached their conclusions after analyzing trading histories of more than 5,000 individual prediction market contracts on Polymarket and Kalshi between October 2025 and this February. For each contract, the researchers collected its order book at 15-minute intervals, enabling them to determine the impact of trade size on a contract's price.

(Polymarket has a data partnership with Dow Jones, the publisher of MarketWatch.)

To appreciate what the researchers found, it's useful to step back and review what the Efficient Market Hypothesis (EMH) would have us believe. It holds that, since asset prices reflect all available information, the price is the best estimate of fair value. The EMH doesn't imply that the market is never wrong, but it means that, on average across all securities, the price is right.

This is the source of the well-known tendency for crowdsourced opinions to be remarkably accurate, significantly better than any of us individually. It's less well-known that this tendency exists only when the crowd is large and diverse and not dominated by a few whales. It's this precondition that is not met by the prediction markets.

The absence of this precondition wouldn't be so consequential if the average whale's bets were no better or worse than those of the smallest investors. But that's not the case, according to the authors of this new study. Whales tend to have a worse record than smaller traders. This is illustrated in the accompanying chart, which shows the perfectly inverse correlation between trader size and the probability of profit - known as "edge."

We should not be surprised by this result. One analyst who predicted it from first principles was Joachim Klement, a managing director at U.K. investment bank Panmure Liberum. Writing in February, before this new study was published, he argued that "prediction markets allow forecasters to express the level of confidence in their forecasts by betting larger sums on their views. But if individual 'whales' start to bet large sums on an outcome, they materially shift the odds in line with their preferred outcome. Hence, prediction market forecasts reflect more the views of the wealthiest punters rather than a crowdsourced range of opinions."

Robert Arnott, founder of Research Affiliates, has for years made a similar argument in the stock market. "Investors often assume the market's biggest companies are also the fastest-growing businesses," he and his colleague Lillian Wu recently wrote. "The evidence suggests otherwise ... Over the last 35 years, firms ranked just below the largest 500 U.S. stocks tended to grow operating cash flow much faster, yet their stock returns were only modestly better ... Mega-cap firms kept pace with their smaller, faster-growing peers because investors were willing to pay more for them."

The bottom line: Prediction markets are being seriously skewed by what their whales are hoping will be the case rather than what they truly believe will come to pass. That would mean that these markets, rather than being a reliable guide to the future, have come to represent a triumph of hope over experience.

Mark Hulbert is a regular contributor to MarketWatch. His Hulbert Ratings tracks investment newsletters that pay a flat fee to be audited. He can be reached at mark@hulbertratings.com.

-Mark Hulbert

This content was created by MarketWatch, which is operated by Dow Jones & Co. MarketWatch is published independently from Dow Jones Newswires and The Wall Street Journal.

(END) Dow Jones Newswires

07-13-26 0930ET