how do prediction markets work (2026)

Prediction markets work by letting people buy and sell contracts tied to the outcome of a real-world event, such as an election, a Fed rate decision, or whether a movie wins an Oscar. Each contract resolves to a fixed value (usually $1) if the event happens and to zero if it doesn’t. The price of the contract at any given moment, somewhere between 0 and $1, reflects what traders collectively think the odds of that outcome are.

Buy a “yes” contract at 30 cents and the event occurs, you collect $1, a profit of 70 cents per contract. If it doesn’t happen, you lose your 30 cents. That price of 30 cents is effectively the market’s estimate of a 30% chance the event happens.

The basic mechanics of a prediction market

Every prediction market contract is built around a specific, verifiable question with a defined resolution date. Examples include “Will the S&P 500 close above 7,000 by December 31, 2025?” or “Will the Federal Reserve cut rates at its next meeting?” The question has to resolve cleanly to yes or no, based on a source everyone agrees on, like an official vote count or an index closing price.

Traders can take either side:

• Yes contracts pay $1 if the event happens, $0 if it doesn’t.
• No contracts pay $1 if the event doesn’t happen, $0 if it does.

Because yes and no prices always add up to roughly $1 (minus fees), buying “no” is functionally the same as betting against “yes.” A yes contract trading at 65 cents implies a no contract trading around 35 cents.

How prices move

Prices shift with buying and selling pressure, the same way stock prices do. If a poll shows a candidate gaining ground, traders buy “yes” contracts on that candidate winning, pushing the price up. If a company reports bad clinical trial data, contracts betting on FDA approval drop. There’s no central authority setting the price; it’s set by whoever is willing to trade at that level.

Most platforms use an order book model, matching buyers and sellers directly, similar to a stock exchange. Some smaller or crypto-based markets use automated market makers instead, where a pricing algorithm adjusts odds based on the ratio of money on each side.

Where the price comes from: aggregating information

The core idea behind prediction markets goes back to economist Friedrich Hayek’s 1945 argument that prices aggregate scattered, decentralized information better than any single expert or model can. Every trader brings their own information, whether that’s a poll they trust, insider knowledge of an industry, or a hunch based on local conditions. When thousands of people put money behind those views, the resulting price tends to reflect a more complete picture than any one source.

This is why prediction market prices are often compared to polls or forecasts. A poll asks people what they think will happen. A prediction market asks people to put money on it, which tends to filter out casual or biased opinions since bad bets cost money.

Who runs these markets and how they make money

In the US, several platforms currently operate prediction markets, including Kalshi, Polymarket, ForecastEx (owned by Interactive Brokers), and PredictIt. They differ in structure and legal status:

• Kalshi is regulated by the Commodity Futures Trading Commission (CFTC) as a designated contract market, similar to a futures exchange.
• ForecastEx contracts are also CFTC-regulated and available through Interactive Brokers and Robinhood.
• Polymarket operates on blockchain infrastructure and settles trades using cryptocurrency; its US legal status has shifted over time.
• PredictIt runs under a no-action letter from the CFTC, with caps on how much any one trader can invest per contract.

These platforms typically make money by charging a small fee per trade, taking a cut of winnings, or charging a spread between the buy and sell price. Fees vary widely; some platforms charge as little as a penny per contract, while others take a percentage of profits when a contract resolves.

A concrete example

Say a market asks: “Will it rain in Chicago on July 4th?” The “yes” contract trades at 40 cents, meaning the market estimates a 40% chance of rain.

• You buy 100 “yes” contracts for $40 total.
• If it rains, each contract pays $1, so you receive $100, a $60 profit.
• If it doesn’t rain, the contracts expire worthless, and you lose your $40.

You don’t have to hold the contract until resolution. If the forecast shifts and the “yes” price rises to 60 cents before July 4th, you can sell your contracts for a $20 profit without waiting to see if it actually rains.

What prediction markets are used for

Beyond entertainment or speculation, prediction markets have practical uses:

• Elections. Political betting markets date back to at least 1884 on Wall Street, and academic research by economists Paul Rhode and Koleman Strumpf found betting turnover in presidential elections has historically rivaled campaign spending.
• Corporate forecasting. Eli Lilly used internal prediction markets to help forecast which drugs in development were most likely to clear clinical trials.
• Economic indicators. Markets tied to Fed decisions, inflation reports, and jobs numbers let traders and analysts gauge probabilities the same way they’d read a Treasury yield curve.
• Academic research. The University of Iowa’s Iowa Electronic Markets, launched in 1988, was one of the first modern platforms used to study how markets forecast election outcomes compared with traditional polling.

Are prediction markets legal in the US?

The legal landscape has been shifting. The CFTC blocked Polymarket from accepting US traders in 2022 and tried to shut down Kalshi and PredictIt earlier in the decade. PredictIt won a court case in 2023 that let it keep operating under its existing no-action letter. Kalshi won an injunction against the CFTC in October 2024, which let it continue offering election-related contracts while litigation continued, a decision widely read as opening the door for regulated election betting markets. Interactive Brokers and Robinhood both added election-related contracts through ForecastEx shortly after.

Rules still vary by platform and by state, and some state regulators have pushed back against prediction markets they consider unlicensed gambling. Check a platform’s current regulatory status and any state-specific restrictions before trading.

Risks to understand before trading

• You can lose your entire stake. Unlike a stock, a losing contract goes to zero, not just down in value.
• Markets can be thin. Low-volume contracts can have wide bid-ask spreads, making it expensive to enter or exit a position.
• Resolution disputes happen. Ambiguous event questions occasionally lead to disagreements over how a contract should settle.
• Regulatory status can change. A platform legal today could face restrictions tomorrow, affecting your ability to withdraw funds or close positions.

Anyone trying prediction markets for the first time should start with a platform’s smallest available contract size, read the exact resolution criteria before betting, and treat the exercise as speculation rather than investing.