What Is a Prediction Market? Event Contracts, Prices, and Payouts, Explained

A prediction market is an exchange where you trade on the outcome of a future event. Each market asks a yes-or-no question — Will the Fed cut rates in September? Will this film win Best Picture? — and sells contracts that pay out a fixed amount, usually $1, if you’re right and nothing if you’re wrong. The price of that contract, somewhere between 1¢ and 99¢, is the market’s live estimate of the probability the event happens. That’s the entire machine: a price that is also a forecast.

How an event contract works

Every prediction market runs on the same basic instrument, called an event contract. It has three parts:

A question with a verifiable answer. Good markets are written so there’s no argument about the result: “Will the Federal Reserve cut its target rate at the September meeting?” resolves from the Fed’s own announcement. Every market publishes resolution criteria in advance — the fine print that says exactly what counts as “yes.”

Two sides you can buy. You can buy “yes” or you can buy “no.” Because one of them must end up correct, their prices are two views of the same probability: if yes trades at 72¢, no trades at about 28¢.

A fixed payout at resolution. When the event resolves, the winning side redeems at $1 per contract and the losing side expires at zero. Buy yes at 72¢ and it happens: you make 28¢ per contract. It doesn’t: you lose your 72¢.

You don’t have to wait for resolution, either. Contracts trade continuously, like a stock — if you bought yes at 40¢ and news pushes the price to 65¢, you can sell and take the gain without ever finding out how the story ends.

The price is a probability

This is the idea that makes prediction markets worth anyone’s attention. A contract priced at 72¢ is the crowd saying there’s roughly a 72% chance the event happens — because if traders believed the true odds were higher, buying at 72¢ would be profitable and they’d bid the price up, and if they believed the odds were lower, they’d sell it down. The price settles where the money disagrees.

A prediction market price is a forecast with money behind it.

That’s the difference between market odds and a poll or a pundit. A poll asks people what they think and costs nothing to answer carelessly. A market asks people what they’ll pay, and being wrong is expensive. It doesn’t make markets infallible — thin markets misprice things, and crowds share blind spots — but it makes the number honest in a way opinion can’t be.

A worked example

Say a market asks whether the Fed cuts rates at its next meeting, and yes is trading at 62¢.

You think a cut is more likely than that — closer to 80%. You buy 100 yes contracts for $62. Two weeks later, a soft inflation report lands and the market reprices to 78¢. You now have two options: sell your 100 contracts for $78 and lock in a $16 profit no matter what the Fed does, or hold to resolution. If the Fed cuts, your contracts redeem for $100 (a $38 profit). If it doesn’t, they expire worthless and you’re out $62.

Notice what the market did along the way: it turned every trader’s read of the inflation report into a single public number, instantly. That repricing — from 62¢ to 78¢ — is information you can watch, whether or not you ever place a trade.

Where event contracts trade

As of August 2026, the U.S. landscape has a few distinct flavors:

Regulated exchanges. Kalshi operates as a CFTC-designated exchange, with dollars in and dollars out like a brokerage. The sportsbook giants have entered the same lane: DraftKings runs its own regulated event-contract exchange, and FanDuel has partnered with CME Group on one.

Crypto-native venues. Polymarket, the largest global venue, runs on crypto rails and has been re-entering the U.S. market through a licensed exchange it acquired.

Brokerage apps. Robinhood and Interactive Brokers offer event contracts inside ordinary brokerage accounts, sourced from regulated exchanges.

Which platforms can operate where — and whether some of their markets count as trading or as gambling — is actively being fought over by regulators, states, and courts right now. It’s the biggest story in this space, and it changes month to month.

How this differs from sports betting and stocks

Versus a sportsbook: a sportsbook sets the odds and takes the other side of your bet — the house has a position against you. A prediction market is an exchange: you trade against other people, prices move with the order book, and the venue earns fees rather than your losses. You can also exit early by selling, which a bet slip doesn’t allow.

Versus stocks: a share of stock is open-ended ownership with no expiration. An event contract is binary and terminal — it resolves to $1 or $0 on a known question by a known process. There’s no dividend, no decade-long compounding; there’s a probability, a price, and a deadline.

Why the prices matter even if you never trade

Prediction market prices have become a public forecasting instrument. Newsrooms, analysts, and increasingly AI assistants cite them as the live consensus on elections, Fed decisions, and geopolitical risk — often before polls or experts catch up, because traders reprice the moment news breaks. Reading those prices well — knowing when a market is deep enough to trust and when it’s a thin market twitching — is a skill, and it’s a large part of what this site exists to teach.

What to watch out for

Honest basics before you trade a dollar: fees vary by platform and can quietly eat the edge on small-margin trades. Liquidity varies wildly — a big election market is deep, a niche culture market may barely trade, and crossing a wide spread costs you. Resolution risk is real: your contract pays out based on the market’s written criteria, not your reading of the news, and ambiguous wording has burned traders before. And the legal landscape is unsettled — where you can trade depends on where you live, and it’s changing fast. We track and explain all four of these in depth across this site.

Quick answers

Do yes and no prices always add up to $1? Approximately, not exactly — the gap is the bid-ask spread, and on some venues, fees. If yes+no ever sells for meaningfully less than $1 combined, traders arbitrage the gap shut.

What happens if the event is ambiguous? The market resolves by its pre-written criteria, decided by the exchange’s resolution process. Reading those criteria before buying is the single most underrated habit in this hobby.

Is this legal? On CFTC-regulated exchanges, event-contract trading is federally regulated — but several states are challenging parts of it in court, especially sports markets, and access differs by state and platform. It’s a moving target we cover continuously.

Are prediction markets accurate? They have a strong track record on big, liquid questions and a weaker one on thin markets — a full, numbers-based answer is its own article, coming soon.