The History of Prediction Markets

Prediction markets did not arrive with the internet. People have priced the future with real money for well over a century, and today’s apps are the latest chapter in a long story that runs from Wall Street betting rings through university research labs to crypto exchanges and play-money games.

The through-line never changes: let people put something on the line, aggregate their bets into a single price, and read that price as a forecast. What changed over the decades is who was allowed to play, how the markets were built, and how the law treated them. Here is how we got from there to here.

Before the internet: a century of election betting

Long before anyone spoke of "prediction markets," large sums changed hands on American elections. From roughly the 1860s into the 1940s, organized betting on presidential races ran openly in the financial district of New York, with brokers matching wagers the same way they matched stock trades. In several elections the money involved rivaled trading in the securities themselves.

These markets were not a sideshow. Newspapers printed the odds, and the implied favorite was correct far more often than not. With no scientific polling to compete against, the betting price was frequently the best public forecast available. The practice faded in the 1940s as the modern opinion poll arrived and anti-gambling enforcement tightened, but it had proven a point that would be rediscovered decades later: a market can forecast.

1988: the Iowa Electronic Markets

The modern academic version began at the University of Iowa. The Iowa Electronic Markets have run since 1988 as a real-money research market operated by the university, letting students and researchers trade contracts tied to elections and economic events.

The stakes are deliberately small, but the money is real, and that is the point: the Iowa markets were built to study whether a market could forecast better than polls. Across many election cycles they often did, especially far from election day. They remain one of the longest-running prediction markets in the world and set the template for almost everything that followed.

1996: the Hollywood Stock Exchange

If Iowa proved the academic case, the Hollywood Stock Exchange proved the popular one. Launched in 1996, it let anyone trade play-money "shares" in films, actors, and awards, with prices meant to predict box-office takings and Oscar winners. Because it used virtual currency rather than cash, it sidestepped gambling law entirely and grew a large, engaged community.

It also showed that play money can still produce sharp forecasts. When thousands of movie fans compete for bragging rights and a leaderboard spot, the resulting prices tracked real outcomes surprisingly well, a lesson that today’s free-to-play forecasting apps took to heart.

1999 to 2013: the rise and fall of Intrade

The first prediction market to reach a truly mass online audience was Intrade, founded in 1999 by the Irish entrepreneur John Delaney. Trading real money on everything from elections to economic data, it became the number journalists quoted on election night.

Intrade’s accuracy and visibility made it the public face of prediction markets through the 2000s. But it operated in a legal gray zone for United States traders. In November 2012 the CFTC sued Intrade for offering illegal off-exchange commodity option contracts. It closed to United States customers the following month, and in March 2013 it shut down entirely and moved to settle its outstanding contracts. Its collapse became the cautionary tale that shaped how the next generation approached regulators.

That legal question, who may run and use these markets, still defines the field. See are prediction markets legal for where the law stands today.

2014: PredictIt and the academic exemption

Into the gap left by Intrade stepped PredictIt, which launched in 2014 under a CFTC no-action letter granted to Victoria University of Wellington in New Zealand, framing the platform as a research project.

PredictIt gave United States political junkies a legal, if tightly capped, place to trade on elections, with small position limits that fit its academic framing. For most of the 2010s it was the go-to real-money election market in the country, and its prices became a fixture of political commentary, keeping the Intrade idea alive under a very different legal structure.

The modern era: exchanges, crypto, and apps

The current wave is the most diverse yet, split across several models. Some are fully regulated exchanges, some run on crypto, and some use play money, but they all descend from the same idea the Wall Street betting rings first put to work.

On the regulated side, Kalshi became the first CFTC-regulated exchange built specifically for trading on the outcome of events, bringing prediction markets fully inside United States financial regulation.

On the crypto side, Polymarket grew into one of the largest prediction markets by volume, settling trades on a blockchain and drawing global attention during major elections.

For play-money forecasting, Manifold lets anyone create a market on any question and trade with virtual currency, lowering the barrier to entry to zero.

And for rigorous community prediction without a market at all, Metaculus scores forecasters on their accuracy over time, showing that reputation can motivate good predictions just as money does.

If you want the mechanics behind all of these, prediction markets explained walks through how a price becomes a probability.

Where this leaves you

The history rhymes. Every era found a new way to let people stake something on the future and read the crowd’s price as a forecast. Real-money exchanges, crypto platforms, and play-money apps are simply different answers to the same regulatory and design questions the Wall Street betting rings first raised.

For a tour of the platforms you can actually use today, see the best prediction apps.

Clutch carries that lineage into a free, play-money app: predict real news and sports with in-app credits, climb a leaderboard, and watch your own track record build. Get the app and add your prediction to the crowd.