Election Trading: Midterm Accuracy Beats Broadcast Calls

Online trading platforms called the results of the 2026 midterm elections hours before television networks did.

Democratic candidates won a majority of contested seats, a shift that observers linked to voter frustration with President Donald Trump and the Republican Party.

Before polls closed, traders on platforms like Kalshi and Polymarket priced in these outcomes with high precision.

Kalshi reported that its markets correctly predicted 98 percent of House, Senate, governor, attorney general, lieutenant governor, and secretary of state races.

The company measured accuracy by comparing market favorites at 5 p.m. EST, just before polls closed in Indiana and Kentucky, to the final results.

Specific accuracy rates included 95 percent for House races and 94 percent for Senate races.

Kalshi noted that competitive seats in Texas, Maine, and Kansas all went to Democrats, matching the platform's projections.

Polymarket stated its markets were accurate in 98.4 percent of resolved races when compared to data from the day before the election.

That platform reported roughly 97 percent accuracy in Senate and governor races, and 98.6 percent accuracy in House races.

Shayne Coplan, CEO of Polymarket, posted on X that the gap between market calls and network calls now spans hours.

He explained that networks wait for absolute certainty to maintain suspense, while traders want real-time information.

Trading volume on Kalshi election markets nearly doubled over two years, rising from $1.3 billion in 2024 to $2.5 billion in this cycle, the company said Wednesday, Nov. 4, 2026, in a statement regarding the results. A Washington Post analysis estimated nearly $2 billion was wagered on Tuesday's midterms across platforms like Kalshi and Polymarket. Ross Dahlke, a professor of computational social sciences at the University of Wisconsin-Madison, said prediction markets react quickly to new information, unlike traditional media which prioritizes fact-checking. He noted that the high accuracy rate is evident. However, Dahlke cautioned that initial numbers only include resolved markets, which may be easier to predict than those still in progress. Many races were considered fairly certain before Election Day. Matthew Bakowicz, a business professor at American University, said the accuracy creates an incentive for casual observers to become active traders. He added that as these platforms attract more money, lawmakers will likely focus more on market integrity. Congress has voiced concern about insider trading and gambling addiction in these markets. The platforms have had notable misses in other elections. Polymarket showed Flávio Bolsonaro with a greater than 80 percent chance of winning the Brazilian presidential race before he lost to Luiz Inácio Lula da Silva. Kalshi traders priced Spencer Pratt at 78 percent to advance in the Los Angeles mayoral primary, but Karen Bass and Nithya Raman won the nomination. In August, Francesca Hong appeared favored to win the Wisconsin gubernatorial Democratic primary with a 95 percent chance on Kalshi, but David Crowley won the nomination instead.