Prediction markets, also known as betting markets, information markets, decision markets, idea futures, or event derivatives, are open markets that enable the prediction of specific outcomes using financial incentives. They are exchange-traded markets established for trading bets in the outcome of various events. The most common form of a prediction market is a binary option market, which will expire at the price of 0 or 100%.
Prediction markets can be thought of as belonging to the more general concept of crowdsourcing which is specially designed to aggregate beliefs on particular topics of interest, where the market price can indicate what the crowd thinks the probability of the event is. Traders with different beliefs trade on contracts whose payoffs are related to the unknown future outcome and the market prices of the contracts are considered as the aggregated belief.
Prediction markets are considered gambling by many governments, and are banned in some locations. Some users and researchers have reported that prediction markets are similar to gambling and can cause addiction.
History Before the era of scientific polling, early forms of prediction markets often existed in the form of political betting. One such political bet dates back to 1503, in which people bet on who would be the papal successor. Even then, it was already considered "an old practice". According to Paul Rhode and Koleman Strumpf, who have researched the history of prediction markets, there are records of election betting in Wall Street dating back to 1884. Rhode and Strumpf estimate that average betting turnover per US presidential election is equivalent to over 50 percent of the campaign spend.
Economic theory for the ideas behind prediction markets can be credited to Friedrich Hayek in his 1945 article "The Use of Knowledge in Society" and Ludwig von Mises in his "Economic Calculation in the Socialist Commonwealth". Modern economists agree that Mises' argument, combined with Hayek's elaboration of it, is correct."Biography of Ludwig Edler von Mises (1881–1973)", The Concise Encyclopedia of Economics Prediction markets are championed in James Surowiecki's 2004 book The Wisdom of Crowds, Cass Sunstein's 2006 Infotopia, and Douglas Hubbard's How to Measure Anything: Finding the Value of Intangibles in Business.Douglas Hubbard "How to Measure Anything: Finding the Value of Intangibles in Business" John Wiley & Sons, 2007
Milestones One of the first modern electronic prediction markets is the University of Iowa's Iowa Electronic Markets, introduced during the 1988 US presidential election. HedgeStreet was the first prediction market to seek approval by the Commodity Futures Trading Commission as a designated contract market after the Commodity Futures Modernization Act of 2000, and it was granted in 2004. The exchange was acquired by the United Kingdom–based IG Group and rebranded to Nadex in 2007; Nadex was then acquired by Crypto.com in 2021. In July 2003, the U.S. Department of Defense publicized a Policy Analysis Market on their website, and speculated that additional topics for markets might include terrorist attacks. A critical backlash quickly denounced the program as a "terrorism futures market" and the Pentagon hastily canceled the program. In 2005, an article in Nature stated how major pharmaceutical company Eli Lilly and Company used prediction markets to help predict which development drugs might have the best chance of advancing through clinical trials by using internal markets to forecast outcomes of drug research and development efforts. Also in 2005, Google announced that it had been using prediction markets to forecast product launch dates, new office openings, and many other things of strategic importance. Other companies, such as HP and Microsoft, also conduct private markets for statistical forecasts. Starting around 2022, mainstream adoption of prediction markets Polymarket and Kalshi began. In October 2024, the Kalshi prediction market won a lawsuit against the Commodity Futures Trading Commission, allowing it to relist its election prediction markets. Kalshi's court victory led to a much broader range of prediction markets offered on their platform and by competitors.
Core concepts and mechanics
General mechanics Prediction markets are financial markets made up of binary contracts that resolve based on whether certain events happen or not. These contracts are usually exchange traded through a free floating order book system. The price of such contracts are set between $0.01 and $1 and represent the odds of an event occurring. Each event will have a “Yes” or “No” tradable contract. For example, if a “Yes” contract around an event occurring has a market price of $0.93 then the market is implying that there is a 93% that this event will take place. In the same way the “No” contract in the same market will have a price of $0.07 and thus the market thinks this event has a 7% chance of occurring. The free-floating central limit order book (CLOB) has shown to be an incredibly efficient mechanism for matching pure supply and demand by giving participants the ability to submit trades at whatever price they choose and only being able to take on a trade or prediction if another market participant disagrees.
Pricing and microstructure of markets
Prices in prediction markets are determined within one of the two predominant structures. The dominant structure of modern commercial markets is the central limit order book (CLOB), where participants place limit orders, accept limit orders, and prices are generated by supply and demand. Kalshi operates a fully centralized limit order book as a designated contract market regulated by the Commodity Futures Trading Commission (CFTC). As for Polymarket, it has a "hybrid-decentralized" order book where trades are matched off-chain and settled on-chain on Polygon. Previously, Polymarket operated using an automated market maker (AMM), built on Hanson's scoring rule, until late 2022 when the firm switched to a full order book.
The other structure is the automated market maker, most notably the Logarithmic Market Scoring Rule (LMSR) introduced by Robin Hanson. While in an order book a transaction is matched between two traders, an LMSR market maker keeps a probability distribution over possible outcomes and offers to make trades at prices defined by the cost function, providing guaranteed liquidity even in the case of low trading activity. For each outcome, the price is equal to the exponentiation of the ratio between the number of shares outstanding for that outcome and the liquidity parameter b, normalized so that prices for all the outcomes add up to one. The liquidity parameter regulates the sensitivity of prices to the trades and at the same time limits the maximum loss for the operator of the mechanism, which is known beforehand and bounded to b ln n for n outcomes. The two structures involve a trade-off since an order book requires a willing counterparty for each transaction, while LMSR guarantees liquidity at the expense of a loss which is known in advance and bounded.
Academic literature has examined the meaning of prices in terms of probabilities. Justin Wolfers and Eric Zitzewitz showed that the market clearing price equals the mean trader belief under logarithmic utility and a symmetric distribution of beliefs, and that prices are close to the mean trader belief in many cases, but that they may be biased estimations of these beliefs. Charles Manski argues that prices partially identify mean beliefs, while Wolfers and Zitzewitz argue that prices are still meaningful given reasonable risk preferences.
Granular exchange-level data allowed studying the microstructure of prediction markets in detail. In a working paper concerning Kalshi, Constantin Bürgi, Wanying Deng, and Karl Whelan studied transaction-level data on over 300,000 contracts. They showed that prices were informative and that they became more accurate as the closing date of the market approached, with the mean absolute pricing error decreasing greatly on the last day of the market. There also was a systematic favorite–longshot bias: low-priced longshot contracts performed worse than their prices and lost money on average, while high-priced favorite contracts overperformed relative to their implied probabilities.
The degree of this bias depends on price: contracts priced at ten cents or lower lost over 60% on average, while contracts priced above fifty cents earned some profit. For all the contracts, the average return before fees is approximately −20%. The returns also differed depending on whether participants posted a resting limit order (maker) or accepted such an order (taker). The makers lost on average 9.64%, while takers lost 31.46%. The bid–ask spread in these markets arises from the order-matching process and not from administrative setting by the exchange; for the sample period, Kalshi charged a fee of 0.07 × P × (1 − P) per taker contract and charged makers no fee.
In an efficient market, the prices of the complementary contracts, such as a "Yes" contract and its counterpart "No" contract, or all the mutually exclusive outcomes of an event, should sum up to a dollar. Incoherence in prices generates arbitrage opportunities, and traders' competition for these opportunities leads to the restoration of coherence, a process which is usually regarded as evidence of the informational efficiency of the market. Empirical research on Polymarket has identified such opportunities in practice. Analysing on-chain historical order book data, Oriol Saguillo and colleagues distinguish market rebalancing arbitrage, which arises within a single market or condition, from combinatorial arbitrage, which spans several related markets, and estimate that approximately 40 million US dollars of arbitrage profit was realized on the platform.
Resolution and settlement of event contracts
Following the occurrence of the event under consideration in the contract, the winning outcome is determined and the contract is settled. The procedures of resolution vary greatly between regulated markets and decentralized ones.
In regulated exchanges, settlement happens according to a rulebook filed with a government regulator prior to the start of trading operations. Every single Kalshi contract is bound by a rulebook filed with the CFTC, which identifies authoritative data sources (Source Agencies) and a Payout Criterion for a "Yes" resolution. The Kalshi markets team relies on the results officially stated by these source agencies, settling each winning contract with a payment of one dollar while all the other contracts expire worthless. Kalshi operates as a designated contract market under the Commodity Exchange Act.
In decentralized systems, the settlement of the contract is determined by an oracle that transfers the information about the result of the event to the blockchain. For example, Polymarket makes use of UMA (Universal Market Access), an "optimistic oracle," which means that the proposed resolution is considered true unless challenged within a dispute window. In case of a dispute, the Data Verification Mechanism (DVM) is launched: a Schelling-point mechanism in which UMA stakers commit votes in secret over a 24-hour period and reveal them during the following 24 hours. According to UMA documentation, a dispute resolves when at least 65% of staked UMA both votes and agrees on a single outcome, and a vote that does not meet this threshold rolls into the following round. Stakers who abstain or vote against the majority are slashed, and the forfeited stake is redistributed to the majority voters. UMA also reports that its optimistic oracle currently resolves 99.8% of requests without escalating them to the DVM.