Kalshi is offering up to $1m in research grants to study how prediction markets performed during the 2026 US midterm elections, inviting researchers to examine everything from election-night price movements to the accuracy of market forecasts.
The initiative follows an election cycle that generated nearly $3.2bn in trading volume on Kalshi’s midterm markets alone as of Nov. 10. The election also saw claims about the accuracy of prediction markets draw pushback from prominent political forecasters.
Kalshi opened applications Nov. 10 for its 2026 Midterms Research Grant program, which will provide up to $1m in funding for as many as 50 projects. Individual grant amounts will vary, and the exchange is not obligated to award the full amount.
Research to examine market accuracy and election-night trading
Kalshi identified several areas of particular interest, including how quickly contract prices reacted to early election returns and race calls. Those movements were tracked throughout NEXTPredict.io‘s 2026 midterms live coverage.
Other suggested topics include the performance of prediction markets in congressional races with limited polling, the impact of campaign developments on prices, and differences in trading activity across states and races.
The exchange is also inviting proposals examining how closely market-implied probabilities matched actual election outcomes, a measure known as calibration.
The research could help address questions raised following the election, when Kalshi CEO Tarek Mansour said market favorites had won approximately 98% of the first 487 races decided. Forecaster Nate Silver challenged broader claims about prediction markets’ forecasting abilities, arguing that leading election models had been more confident and quicker to identify likely outcomes.
A separate Dartmouth study published ahead of the midterms also examined the relationship between election-market pricing and political bias.
Researchers are not limited to Kalshi’s suggested subjects and can propose other questions involving prediction market data from the 2026 midterms.
Applications open through Nov. 18
Applicants must submit a research proposal of no more than two pages outlining their question, methodology, data requirements and intended public output, which could include a research paper, report, dataset, dashboard or code.
The deadline for direct applications is 11:59 p.m. ET on Nov. 18. Researchers can also enter by posting their idea on X by Nov. 18, with an additional two days to complete the application form.
Applicants must be US residents aged 18 or older, with certain exclusions for political campaign personnel, election officials and Kalshi affiliates. No university affiliation, Kalshi account or trading activity is required.
Kalshi plans to evaluate submissions between Nov. 21 and 26 and notify selected recipients around Nov. 27.
Selected researchers must publish their findings and disclose Kalshi’s financial support. Under the official program rules, recipients retain editorial independence, with Kalshi prohibited from directing or approving their conclusions.
The exchange may provide publicly available trading data to support the projects but does not guarantee access to any particular dataset.