STU BURGUIERE: Put your money where your mouth is: Prediction markets won the midterms

Billions of dollars were spent this election season to lie to you.

More than $12.8 billion was spent on political advertising in the 2026 midterms making this the most expensive election cycle in American history. In my home state of Texas, the Senate race drew preliminary estimates of nearly $573 million placing it among the most expensive non-presidential contests on record.

Despite all the differences between races and candidates, a few things were dependable. Every campaign said it was going to win. Every analyst said their numbers were right. Every super PAC said its candidate had the best policies and character. And the mainstream media picked its favorites, cut to commercial, and served up a dozen ads of warring politicians accusing each other of high crimes and misdemeanors.

TRADERS POUR NEARLY $1.3 BILLION INTO PREDICTING WHO WILL CONTROL CONGRESS

A massive industry exists to bend your view of reality toward whatever our would-be leaders in Washington and the state capitals want you to believe. Americans need a way to cut through that noise. No wonder prediction markets are booming.

As with any new technology, prediction markets have drawn a lot of attention and a lot of criticism. Senators wrote letters warning about election contracts. States moved to sue or ban them outright. And somewhere, a podcaster is still on hour 133 of explaining how it's all rigged.

POLLSTERS MISSED THE MARK AGAIN — BUT THIS TIME, IT WASN'T REPUBLICANS THEY UNDERESTIMATED

Then Election Day settled the argument. Across all tracked federal contests, Kalshi's markets had the eventual winner out front in 98% of races. The markets not only correctly priced non-competitive incumbents they also accurately reflected the razor-thin odds in core battlegrounds.

This wasn't a one-time fluke. Kalshi studied 2,243,741 of its resolved markets, dating from its launch in 2021 through mid-2026. It found that prices became remarkably well calibrated as events approached. In plain English, the prices meant what they said. Events priced at 30 cents happened about 30% of the time; events priced at 90 cents, about 90%.

Outside researchers have reached the same conclusion. Brier.fyi, an independent site that measures prediction-market accuracy, saw a similar pattern across nearly 74,000 Kalshi markets. In a study published last month in the Proceedings of the National Academy of Sciences, researchers examined more than 1 million resolved forecasts across four platforms and found events happened about as often as their prices said they would. A Washington Post review of this year's primaries showed that candidates won about as often as their market prices suggested.

THE BATTLE FOR CONGRESS GETS A WALL STREET-STYLE SCOREBOARD POWERED BY REAL-MONEY BETS

Despite that long track record, critics will seize on upsets like Democrat Annie Andrews' defeat of Republican Sen. Darline Graham in South Carolina as p