Prediction Market

A prediction market is a market in which participants buy and sell contracts tied to the outcome of a future event, such as an election, a sports result or an economic data release. A typical contract pays a fixed amount, often one dollar, if a stated outcome happens and nothing if it does not. Because traders compete to buy underpriced contracts, the price tends to settle at a level read as the crowd's estimate of the probability of that outcome.

Prediction markets have been studied by economists for decades as tools for gathering dispersed information, drawing on ideas linked to the efficient market hypothesis. They overlap with sports betting and financial derivatives, and their legal treatment depends heavily on how a given country classifies event contracts. In the United States, regulated venues such as Kalshi operate under the CFTC, while crypto-based platforms such as Polymarket and Augur use blockchains to settle trades.

Imagine a jar of tickets for the question "Will it rain in the city on Saturday?" Each "yes" ticket is worth one dollar if it rains and nothing if it does not. If people are happy to pay 70 cents for a "yes" ticket, the crowd is roughly saying there is a 70% chance of rain. If someone knows more, for example by checking a better forecast, they can buy or sell tickets and move the price.

The appeal is that people put money behind their beliefs, which encourages them to think carefully. The price becomes a running summary of everyone's best guess, updated every time new information arrives.

For a simple yes-or-no contract paying 1 dollar, the price can be read directly as an implied probability:

If a "yes" contract trades at 0.65 dollars and pays 1 dollar, the implied probability is 0.65 ÷ 1 = 65%. A trader who buys 100 "yes" contracts pays 100 × 0.65 = 65 dollars. If the event happens, they receive 100 dollars, a profit of 35 dollars before fees; if it does not, they lose the 65 dollars. Matching "no" contracts trade at roughly 1 − 0.65 = 0.35 dollars.

Markets use either an order book, matching buyers and sellers, or an automated market maker that quotes prices from a formula. When the event is over, a resolution process decides the outcome. Centralised venues use their own rules and data sources; blockchain platforms may use an optimistic oracle or a token-holder vote.

• Iowa Electronic Markets. Run by the University of Iowa since 1988 for research and teaching, with small trading limits, it is one of the oldest real-money examples.
• Augur. Launched on Ethereum in 2018, Augur was an early decentralised prediction market in which anyone could create a market, an example of permissionless market creation. It attracted limited lasting activity.
• Polymarket. A crypto-based platform founded in 2020 that settles trades in the stablecoin USDC. In January 2022 it settled with the CFTC over offering unregistered event contracts and agreed to block US users. It drew very large volumes around the 2024 US presidential election and has since sought to re-enter the US market through a regulated entity.
• Kalshi. A US exchange approved by the CFTC as a designated contract market, offering contracts on economic data, weather and other events. In 2024 it won a federal court case allowing it to list contracts on US congressional elections, and it later expanded into sports-related contracts, which brought disputes with state gambling regulators.

Supporters argue that prediction markets can produce useful forecasts for elections, economic releases and business questions, and that they let people and firms practise hedging against specific events. Some companies have run internal markets to forecast project deadlines or sales. Researchers use their prices as data on how expectations change in response to news.

Whether a prediction market is a financial exchange, a form of gambling or something else is contested. In the US, event contracts fall under commodities law, but the CFTC may bar contracts that involve gaming or are contrary to the public interest, and states regulate sports betting separately; how these rules fit together remained in dispute as of November 2026. Many other countries ban or license prediction markets as gambling. Critics warn that markets on elections or violence could create incentives to influence outcomes, that thin markets can be moved by a single large trader, that ambiguous questions lead to disputed resolutions, and that heavy users can lose money much as gamblers do. Insider trading on event outcomes is a further concern. Prices are estimates, not guarantees, and should not be treated as certain forecasts or as advice to trade.