Prediction markets place odds on election results. Officials say they could erode public trust.

Prediction Markets Draw Scrutiny From Election Officials Over Regulation and Public Perception

Add betting markets to the list of items under review by election officials this year.

This article was originally published by Votebeat, a nonprofit news organization covering local election administration and voting access. Votebeat reports on voting access and election administration across the U.S. The analysis was distributed in Votebeat’s weekly newsletter.

A 2004 paper on the history of presidential election betting markets opens with a partial quote from Henry David Thoreau’s 19th-century essay “Civil Disobedience”: “All voting is a sort of gaming, like checkers or backgammon, with a slight moral tinge to it, a playing with right and wrong, with moral questions; and betting naturally accompanies it.”

Not all observers accept the extension to betting. The growth of online prediction markets such as Kalshi and Polymarket has renewed discussion of the distinction.

Prediction markets operate by selling contracts that pay out based on whether specified events occur. Examples include contracts on which party will hold a majority in the U.S. House or Senate. Contract prices adjust with trading volume and are interpreted by participants as reflecting collective estimates of probabilities. Payouts are fixed at settlement.

Regulatory authority over these platforms remains unsettled. Many states maintain statutes that restrict election-related wagering, according to a Pew Research Center review issued in June. Several of those statutes predate current platforms and have seen limited recent judicial review.

Kalshi responded to a Wisconsin advisory that cited a 19th-century statute barring voters from participating in races on which they have placed bets. A Kalshi representative posted on social media that the advisory constituted “active voter suppression.”

Betting activity on election outcomes has continued. An NBC News review recorded nearly $200 million in trading volume on Kalshi and Polymarket contracts tied to midterm results. Election officials have expressed concern that published odds could affect perceptions of election integrity, especially if certified results diverge from market-implied probabilities, and that the structure could create financial motives for interference.

Dean Logan, registrar-recorder and county clerk for Los Angeles County, California, addressed the topic during a webinar hosted by the Partnership for Large Election Jurisdictions. Logan stated that in the June Los Angeles mayoral primary, early returns diverged from market-implied probabilities and that this divergence prompted questions about standard ballot processing and canvassing steps. He added that election administration operates in an environment shaped by perception and that jurisdictions must prepare responses to resulting inquiries.

Prediction market operators have stated that prices aggregate dispersed information and have produced forecasts that compare favorably with polling averages in multiple cycles. Historical analysis in the 2004 paper by economists Paul Rhode and Koleman Strumpf examined data from 1868 to 1940 and concluded that the markets of that period generated accurate forecasts prior to the development of scientific polling.

Contemporary platforms differ from surveys in participant composition and weighting. Polls draw from defined geographic samples of eligible voters. Prediction markets accept participants without geographic restriction, and larger positions exert greater influence on prices.

Election officials have cited risks that odds could be moved by coordinated trading or that financial exposure could encourage attempts to affect results. Kalshi and Polymarket did not provide comments for this article.

Jim Allen, elections director for Delaware County, Pennsylvania, had already prohibited direct election betting by staff and later extended the prohibition to prediction-market contracts. He compared the activity to game officials wagering on contest outcomes. Allen reported that the updated policy prompted discussion during poll-worker training, including one participant’s question about placing a small wager on turnout levels. The response given was that such activity is not permitted.

Separate analyses have identified more than 210,000 trades across Polymarket and Kalshi between 2024 and 2026 that exhibited characteristics of potential manipulation and generated roughly $143 million in profits. No documented instances have established that these trades changed certified election outcomes or produced measurable shifts in aggregate public confidence metrics.

Prediction markets have recorded higher accuracy than polling aggregates in several 2024 federal and state races, according to post-election evaluations by academic researchers. Officials continue to monitor the platforms while existing statutes undergo review for applicability to modern contract structures.