Prediction Markets Call Midterms Fast, Beat Networks to Key Races

Prediction markets didn’t just watch the 2026 midterms — they called the shots early. Before the first polls closed at 6 p.m. ET Tuesday, Democrats were favored to flip both the U.S. House and Senate on the platforms, and were narrow favorites to take at least 53 Senate seats, with a 54th seen as a toss-up. Democrats have now won 54 Senate seats, including Iowa, which markets had at 50-50 odds, and South Carolina, where Republicans were narrowly favored. Results are still outstanding in Mississippi and Alaska, where markets give Democrat Scott Colom and Mary Peltola 90% odds of winning.

Polymarket bragged on X that it projected 10 key races ahead of CNN, Fox News and the Associated Press by a median 31 minutes. CEO Shayne Coplan said it’s now routinely hours between when prediction markets call a race and when TV networks do. “No knock to them, but now there’s Polymarket, and many people are realizing they want to know what’s happening in real time,” he posted. Kalshi was confident making calls when a candidate’s odds were above 99% and trading on the opposite candidate “dried out,” spokesperson Jack Such said.

The platforms moved fast when vote counts shifted. In South Carolina, Democratic Senate candidate Annie Andrews was seen as the favorite on Kalshi shortly after 7:30 p.m., just 30 minutes after polls closed. In Texas, where Democratic state Rep. Gina Hinojosa ousted Republican Gov. Greg Abbott, Hinojosa’s odds on Polymarket jumped 34 percentage points in five minutes, reaching 73% after they stood at 39%.

It wasn’t a flawless run. Right before the midterms, prediction markets missed major global elections, including Brazil’s presidential race, where Luiz Inácio Lula da Silva had only about 20% odds to win on Kalshi before polls closed. They also whiffed in high-profile Democratic primaries over the summer, giving extremely high odds to progressives in Wisconsin and Michigan, CNBC reported. Abdul El-Sayed narrowly won his Michigan primary before winning Thursday’s general election.

Academics say a low-odds winner doesn’t mean the markets are broken. Rajiv Sethi, an economics professor at Barnard College, compared it to tossing two coins and getting two heads in a row — a 25% chance nobody would find shocking. Dartmouth’s Eric Zitzewitz said casual observers may see 55-45 odds and think the candidate at 55 is way ahead. Rutgers statistics professor Harry Crane called this “the first true prediction markets election,” and experts expect the platforms to remain a force through the next presidential election.

The success won’t quiet calls for regulation. Michelle Kendler-Kretsch of the Anti-Corruption Data Collective pointed to a Reuters/Ipsos survey finding most Americans oppose using prediction markets to trade on elections, saying risks and vulnerabilities didn’t change overnight just because the markets were accurate this time. Flip Pidot, a board member at the Coalition for Political Forecasting, said financially motivated number crunchers around the world now deliver signals much faster. CNBC disclosed it has a commercial relationship with Kalshi, including customer acquisition and a minority investment.