A short history of prediction markets
From betting on elections on Wall Street to Iowa's academic experiment and today's global platforms — how prediction markets grew up.
Outcomer Team · Aug 3, 2026
Prediction markets can feel like a very modern idea — apps, live odds, crypto rails. But the core mechanism, letting people trade on what will happen and reading the price as a probability, is more than a century old. Here is how the history of prediction markets runs, and why each chapter still shapes how the markets work today.
Betting on elections before the pollsters
Long before scientific polling existed, Americans wagered on presidential elections in the open. From the 1880s through the 1930s, large, organised betting on election outcomes took place on Wall Street, and newspapers reported the odds as a serious read on who was ahead. In many years those markets called the winner well before election day.
They faded for two reasons: the rise of scientific opinion polling in the late 1930s gave people a new way to gauge public sentiment, and legal betting moved elsewhere. But the basic insight survived — a price that people back with real money tends to aggregate information better than a show of hands. That is the same logic behind the wisdom of crowds.
Iowa turns it into a science
The modern era starts in 1988 at the University of Iowa. Three economists — Robert Forsythe, George Neumann and Forrest Nelson — set up what became the Iowa Electronic Markets (IEM) as a research project, launching it for that year's US presidential election. Unlike the old Wall Street betting, IEM was small, transparent and built for study, run by the university's business school with tiny stake limits.
The results were hard to ignore. Across elections from 1988 onward, IEM prices beat traditional polls a large majority of the time. That academic track record gave prediction markets intellectual credibility and turned "can a market forecast better than a poll?" into a testable question rather than a hunch. If you want the short version of the answer, see are prediction markets accurate?.
The 2000s: mainstream attention and a political misstep
Two very different events defined the next chapter. First, Intrade, which ran from 2001 to 2013, brought prediction markets to a wide audience by listing contracts on elections, awards and world events. For a stretch, "what is Intrade saying?" was a normal question in newsrooms.
Second came a cautionary tale. In 2003, a US defence research agency floated a project nicknamed the Policy Analysis Market, which would have let traders bet on geopolitical events in the Middle East. Critics attacked it as a market in terrorism, and it was cancelled within about a day of becoming public in late July 2003. The episode showed how the framing of a market matters as much as the mechanism — and why serious platforms think carefully about which questions they list.
The modern platform era
By the 2010s the pieces were in place for prediction markets to scale. Regulated, research-linked venues appeared, offering event contracts under formal oversight. Then in 2020, Polymarket launched with a very different design: contracts denominated in a dollar stablecoin and settled on a blockchain, which let anyone with a wallet trade global events without a traditional broker. Alongside regulated US venues, it pushed trading volumes and headlines to levels the old markets never reached.
That is why the platform choices you see today can feel so varied — some are exchange-regulated, some are on-chain, some are play-money. If you are comparing the big names, Polymarket vs Kalshi walks through how two of them differ.
What the history teaches
A few threads run through all of it. Prices that carry real stakes have out-forecast polls for over a century. The mechanism is neutral, but the choice of questions is not — reputation lives or dies on listing sensible, resolvable markets. And accuracy comes from participation: the more informed people trade, the sharper the estimate. None of that has changed from Wall Street in 1900 to a stablecoin contract in 2026.
Understanding where these markets came from makes the price on your screen easier to read — it is not a bookmaker's number, it is a crowd's best estimate, refined by everyone willing to back their view. If you are new to the mechanism itself, start with what is a prediction market.
The best way to build intuition is to try it. On Outcomer you can practise with virtual money — place a few trades, watch how prices move as events unfold, and see the century-old logic work in real time before you ever risk a cent.