Frequently Asked Questions
How accurate are prediction markets as forecasts?
Liquid markets calibrate well — implied probabilities track actual outcomes closely on elections, macro data, and major sports. Thin or late-resolving contracts calibrate worse. Multiple academic studies have found prediction markets at least match and often outperform polling on election forecasting.
Are prediction markets better than polls?
On most US presidential elections since 1988, prediction-market prices have tracked the final outcome more closely than poll averages, especially in the final weeks. Polls remain stronger on demographic breakdowns and turnout questions; prediction markets aggregate everyone’s information into one price.
Where can I get prediction market data?
Kalshi, Polymarket, and ForecastEx all publish public APIs with live pricing and volume. Third-party aggregators like Adjacent News and Manifold collect cross-venue. Settlement data is published end-of-day through standard exchange channels.
What is Polymarket’s daily trading volume?
Polymarket runs the highest peak volumes on individual political events, sometimes clearing hundreds of millions of dollars on a single contract during US election cycles. Daily volume varies widely; the Data section tracks weekly volume reports as they’re published.
What is calibration in prediction markets?
Calibration measures how well market-implied probabilities match actual outcome frequencies. A perfectly calibrated market has 70-cent contracts resolving "yes" exactly 70% of the time. Most academic studies find liquid prediction markets are within a few percentage points of perfect calibration.
Are prediction markets used as macro indicators?
Yes. ICE has called Polymarket data a key macro indicator, and macro research desks at major banks now reference Kalshi prices on Fed decisions, jobs prints, and CPI alongside survey data.