Prediction markets largely picked midterm winners but missed Democratic gains

Politics · First reported 6 Nov, 02:55 · Updated 03:49

Image: Wikimedia Commons · NASA Johnson Space Center · Public domain

Kalshi and Polymarket reported high accuracy in forecasting individual races, but an analysis found both markets and polls underestimated Democratic candidates. Experts differed on how much the markets added beyond polling and vote-count data.

Kalshi said $2.5 billion was traded on its midterm markets; Polymarket reported nearly $700 million in wagers across its U.S. and international platforms.
Both platforms said candidates favoured by their markets won or led in about 98 per cent of races before polls closed or the day before Election Day.
Betting Intelligence identified 14 Democratic upsets among more than 400 races and concluded that polls and markets identified winners equally often while both underestimated Democrats.
In South Carolina, markets gave Republican Darline Graham about a 70 per cent chance before Democrat Annie Andrews won; in Texas, Gina Hinojosa won despite a market estimate of 40.1 per cent the previous day.
Experts said markets reacted to vote counts and polling, and that neither markets nor polls predicted the scale of Democratic victories. One economist also pointed to their differing estimates in Mississippi's close Senate race.

Summary written by Lux AI from 3 articles

Key facts

Kalshi midterm trading: $2.5 billion, according to Kalshi.
Polymarket midterm wagers: Nearly $700 million across its U.S. and international platforms, according to Reuters.
Upsets identified: 14, all won by Democrats, among more than 400 races analyzed by Betting Intelligence.
Mississippi Senate count: Scott Colom led Cindy Hyde-Smith by 1,803 votes as of Thursday afternoon; the race had not yet been called by The Associated Press.