New York State Department of Health Commissioner James McDonald talking in New York last year about measles outbreaks
Jim Franco/Albany Times Union via Getty Images
Betting on the spread of measles in the US has gained traction among gamblers. Since the beginning of the year, nearly $9 million has been wagered on future cases of the disease through prediction markets like Kalshi and Polymarket. These predictions show potential in accurately modeling the disease’s spread.
Prediction markets operate by buying and selling shares based on potential outcomes. Participants bet “yes” or “no” on future events, with the cost of a share fluctuating based on the balance of existing bets.
To illustrate, if 86% of bets on an event are “yes,” the price of a “yes” share is 86 cents. Should the event occur, successful bettors earn $1 per share, while those who bet incorrectly lose their investment.
The concept of prediction markets has its roots in academic research. In 1988, University of Iowa economists Robert Forsythe, George Neumann, and Forrest Nelson devised a market-based method to forecast US federal elections, allowing participants to place small wagers on potential outcomes.
Their election predictions were notably accurate. In 2003, infectious disease researcher Philip Polgreen from the University of Iowa encouraged the integration of disease predictions into the market, emphasizing their educational and public benefit.
Recently, prediction markets have expanded into the commercial sector, with companies like Kalshi and Polymarket offering these services. Although legal in the US and regulated by the Commodity Futures Trading Commission, they are facing increased scrutiny from government bodies.
Critics have raised ethical concerns over bets placed on conflicts such as the Iranian and Ukraine wars, labeling them as immoral. In February, a trader known as Magamyman won $553,000 on Polymarket by accurately predicting the removal of Ayatollah Ali Khamenei, who was killed on 28 February 2026. This success led to suspicions among US Congress members about the potential use of insider information.
With measles cases rising across the US, a betting market has emerged for the disease as well. While the ethics are questionable, Spencer J. Fox from Northern Arizona University, who models covid-19, influenza, and RSV, views these markets as potentially valuable data sources.
For example, in June 2025, prediction markets anticipated around 2000 measles cases by year-end, with the actual figure reaching 2288. “I’ve seen many worse forecasts from our models,” Fox commented.
Epidemiologists typically use varied data sources like vaccination rates, genomic information, and climate data for disease prediction. Fox notes that while measles isn’t usually forecasted scientifically due to its probabilistic nature, new data streams are always sought after.
Emile Servan-Schreiber, CEO of the prediction market company Hypermind, attributes the accuracy of these predictions to the “wisdom of crowds,” where diverse amateur insights compensate for a lack of expertise.
However, Fox cautions that prediction markets cannot replace scientific models, as they lack the range and detail of traditional forecasts. “You would have to make thousands of bets a week for all of the different forecasts that we’re making,” he explains.
He also emphasizes the need for expert input in predicting rare events. “If we don’t invest in the expertise for forecasting infectious diseases now, we’re going to be caught flat-footed for the next covid-19.”
Kalshi and Polymarket did not respond to New Scientist‘s request for comment.
Topics:
