History of Prediction Markets: From 1884 to 2026
The complete history of prediction markets from 19th century election betting to Polymarket. Key milestones, platforms, and the evolution of crowd forecasting.
Prediction markets feel like a modern invention, but people have been wagering on future events for centuries. The idea that market prices can aggregate information about uncertain outcomes has a long and fascinating history, stretching from Wall Street curb markets in the 1880s to the blockchain-powered platforms of today.
Understanding this history helps explain why prediction markets work, how they have evolved, and where they are headed next.
The Early Days: Wall Street Election Betting (1884-1940s)
The first organized election betting markets in the United States operated on Wall Street and at the Curb Exchange (now the American Stock Exchange) starting in the 1884 presidential election. These were not informal bar bets. They were structured markets with published odds, reported daily in major newspapers including the New York Times.
From 1884 to 1940, Wall Street election betting was a mainstream activity. Trading volumes sometimes exceeded those of the stock market on election eve. The markets were remarkably accurate, correctly predicting the winner of 11 of 15 presidential elections during this period.
These early markets died out in the 1940s as scientific polling emerged (Gallup began polling in 1936) and anti-gambling sentiment led to legal restrictions. For decades, organized prediction markets essentially disappeared from the American landscape.
The Academic Revival: Iowa Electronic Markets (1988)
Prediction markets were reborn in academia. In 1988, professors at the University of Iowa launched the Iowa Electronic Markets (IEM) as a research project. Operating under a no-action letter from the CFTC, the IEM allowed participants to trade real-money contracts on presidential elections with account limits of $500.
The results validated what the 19th century markets had shown: crowd-sourced prices were remarkably accurate forecasts. A comprehensive study found that IEM prices were closer to actual election outcomes than 74% of contemporaneous polls. The IEM demonstrated that prediction markets could work with modern technology and rigorous academic oversight.
Key IEM Findings
| Finding | Significance |
|---|---|
| Markets beat polls 74% of the time | Empirical proof of forecasting superiority |
| Small markets still accurate | Even with few hundred traders, prices were calibrated |
| Real money mattered | Play-money versions were less accurate |
| Prices adjusted to news quickly | Information incorporation was near-instantaneous |
The Intrade Era (2001-2013)
Dublin-based Intrade brought prediction markets to a global audience. At its peak, Intrade offered contracts on elections, Supreme Court decisions, economic indicators, geopolitical events, and even entertainment outcomes. The platform attracted traders from around the world and generated significant media attention.
Intrade's most famous moment came in the 2008 U.S. presidential election, when its markets correctly predicted the winner of all 50 states. Major news organizations began citing Intrade prices as authoritative indicators of likely outcomes, bringing prediction markets into mainstream consciousness.
Intrade closed in March 2013 following regulatory pressure from the CFTC (which sued the platform for offering commodity options to U.S. customers) and internal financial irregularities. Its closure left a significant gap in the prediction market landscape that would not be fully filled for years.
The DARPA Controversy and Government Interest
In 2003, the U.S. Defense Advanced Research Projects Agency (DARPA) proposed a "Policy Analysis Market" that would have allowed trading on geopolitical events, including potential terrorist attacks and political assassinations. The program was immediately condemned by senators who called it a "terrorism futures market," and it was cancelled within 24 hours of public disclosure.
Despite the political backlash, the underlying research showed that prediction markets could provide valuable intelligence signals. Several U.S. intelligence agencies quietly developed internal prediction markets for analyst forecasting, and research into their applications continued.
The Fragmented Middle Period (2013-2020)
After Intrade's closure, the prediction market landscape fragmented. Several platforms attempted to fill the gap:
- PredictIt (2014): Launched by Victoria University of Wellington under a CFTC no-action letter. PredictIt focused on political markets with low trading limits ($850 per market). It remains active but with limited liquidity.
- Augur (2018): One of the first blockchain-based prediction markets, built on Ethereum. Augur was technically innovative but suffered from poor user experience, high gas fees, and minimal liquidity.
- Kalshi (2020): Became the first CFTC-regulated prediction market exchange in the United States. Kalshi focuses on event contracts across economics, politics, and culture.
The Polymarket Revolution (2020-Present)
Polymarket, founded by Shayne Coplan in 2020, rebuilt prediction markets for the modern era. Using USDC on the Polygon blockchain, Polymarket combined the accessibility of web-based trading with the transparency and global reach of decentralized infrastructure.
The platform's breakout moment was the 2024 U.S. presidential election. Polymarket processed over $3.5 billion in election-related volume, with its prices proving substantially more accurate than polling averages. The platform attracted hundreds of thousands of new users and established itself as the go-to source for real-time probability estimates on major events.
By 2026, Polymarket dominates the global prediction market landscape with monthly volumes consistently exceeding $3 billion across thousands of active markets spanning politics, economics, crypto, sports, technology, and more.
What Comes Next?
The prediction market industry in 2026 is growing rapidly. Several trends are shaping its future:
- Regulatory clarity: Governments worldwide are developing frameworks for prediction market regulation, providing stability for platforms and participants.
- Institutional adoption: Hedge funds, media organizations, and governments are increasingly using prediction market data for decision-making.
- AI integration: AI-powered trading bots and analytical tools are bringing new levels of sophistication to prediction market trading.
- Mainstream awareness: The 2024 election cemented prediction markets in public consciousness. More people know what they are and how to use them than ever before.
FAQ
Why did early election betting markets disappear?
A combination of factors: the rise of scientific polling (starting with Gallup in 1936), anti-gambling legislation, and the disruption of World War II all contributed to the decline of organized election betting in the 1940s.
Was Intrade a scam?
No. Intrade was a legitimate prediction market that operated for over a decade. Its closure was due to regulatory pressure and internal financial mismanagement, not fraud. Most users received their funds back during the wind-down process.
Why are blockchain-based prediction markets better?
Blockchain provides transparency (all trades are publicly verifiable), global access (anyone with internet can participate), and trust (resolution and payouts are enforced by smart contracts). These properties address many of the issues that plagued earlier platforms like Intrade.
Ready to trade on real prediction markets?
Put your knowledge to work. Trade on thousands of real-money markets covering politics, crypto, sports, and more.
Start trading on Polymarket