Money has a way of sharpening opinions. That is the idea behind prediction markets, where traders buy and sell contracts tied to real-world outcomes, and it explains why political analysts now watch these platforms as closely as they watch the polls during the 2026 midterm primary season.
The shift became hard to ignore on March 3, when Texas opened the primary calendar. Kalshi recorded roughly $1.1 million in trading volume on the Texas Senate races alone, dwarfing the $80,000 changing hands on the same market at rival platform Polymarket. James Talarico took the Democratic nomination, and the Republican contest pushed John Cornyn and Ken Paxton into a runoff after neither cleared 50 percent.
What Texas Showed About Speed
The Texas result was a useful test case. Heading into primary day, Kalshi traders gave Talarico an 86 percent chance of winning the Democratic nomination, even as several polls suggested a tighter race. He won comfortably. On the Republican side, traders had priced in a Cornyn-Paxton runoff before a single ballot was counted.
This is the pattern that keeps drawing attention. During the 2024 cycle, market prices repeatedly tracked outcomes more closely than survey averages did, and researchers who study election forecasting have long found that markets tend to beat polling in most contests they examine. The reason is blunt. A trader who bets on wishful thinking loses real money, which forces a kind of honesty that anonymous surveys rarely produce.
How the Contracts Actually Work
The mechanics are simpler than they look. Every election contract is a yes-or-no question priced between $0.01 and $0.99, and that price is the market’s estimate of probability. A contract trading at 67 cents means traders collectively give that outcome a 67 percent chance. If the outcome happens, the contract pays out $1.00. If it does not, it settles at zero.
You do not have to wait for a race to be called, either. Positions can be sold at the going price at any time before a market closes. Someone who bought Cornyn at 60 cents and watched his odds climb to 81 cents could sell and lock in the gain without holding through the result. The same exit works in reverse, letting a trader cut a loss rather than ride a collapsing position down to zero.
Across Kalshi and Polymarket, roughly 150 distinct midterm contracts are live at any given time, covering Senate and House control, individual primaries, governor races, and narrower props like margins of victory. Trackers that pull this activity into one place make the picture easier to follow. One running tracker of primary election odds, maintained by DeFi Rate, aggregates live numbers from Kalshi and Polymarket and refreshes them every half hour, which is about as close to real time as political forecasting gets.
Reading the Current Numbers
As of late spring, the markets lean toward a divided Washington. The clearest signal sits in the House forecast, where Kalshi’s projected seat count drifted about nine seats toward Democrats between November and March. That movement matters more than any single snapshot, because a sustained trend usually reflects accumulating information rather than a one-day overreaction.
History adds weight to that read. The party holding the White House loses seats in most midterms, and Republicans are defending narrow majorities in both chambers. That structural headwind is a large part of why Democratic contracts have firmed in the House while the Senate stays closer to a toss-up.
Where the Odds Can Fool You
Not every market deserves the same trust. Senate and governor nominee contracts usually carry deep liquidity and tight spreads, so their prices are reasonably stable. House and downballot races are a different story. Many individual House contracts trade under $50,000, and some downballot markets sit below $10,000, which leaves prices jumpy and easy to misread.
Thin markets also absorb news strangely. When a texting scandal hit a Maine Senate candidate this spring, the odds barely moved, holding the favorite in a comfortable lead. Scandals do not always reshape a race the way headlines suggest, and the way scandals actually move elections is a subject worth studying on its own terms. A market with little money behind it can stay frozen simply because too few traders are paying attention, not because the news is irrelevant.
What Makes This Legal
A fair question hangs over all of this. Is betting on elections allowed? For the regulated platforms, yes. Kalshi operates as a CFTC-regulated exchange, and in September 2024 a federal court ruled that its election contracts do not count as gaming under the Commodity Exchange Act. An appeals court upheld that ruling, and the CFTC dropped its challenge in May 2025, settling the legality question that had shadowed the industry for years. The contracts are available to traders in all 50 states.
What to Watch Next
The calendar still has plenty of road left. Maine Democrats vote on June 9, New York holds its primary on June 23, and a cluster of states including Michigan and Arizona follow in August before everything resolves on November 3. Volume on each market tends to spike in the week before a primary, as polling, endorsements, and early-voting data start to land.
For anyone trying to make sense of a long midterm season, the value of these markets is less about the wager and more about the signal. Prices are not prophecy, and a thin market can mislead, but a well-traded contract folds thousands of informed guesses into a single number that updates by the minute. That is a tool worth keeping open in another tab, right next to the polls.
