A clear, beginner-friendly guide to U.S. election prediction markets—how they price odds on the presidency and Congress, which platforms exist, how accurate they really are, and what the explosive growth of this industry means for American politics.
What Are U.S. Election Prediction Markets?
U.S. election prediction markets are platforms where people trade contracts tied to political outcomes—who will win the presidency, which party will control Congress, whether a government shutdown happens, or whether a specific policy gets enacted.
Each contract pays a fixed amount (usually $1) if the political event happens and $0 if it doesn’t. The price of the contract in cents can be read as the market’s implied probability. A contract trading at 60¢ means the crowd sees roughly a 60% chance of that outcome.
This concept isn’t new. Three economists at the University of Iowa created the Iowa Electronic Markets in 1988 after watching polls badly miss the Michigan presidential primary. Their idea was simple: markets where people have real money at stake might produce better forecasts than opinion surveys. Nearly four decades later, that academic experiment has grown into a multi-billion-dollar industry.
The platforms dominating election prediction trading in 2026 include:
Kalshi — A CFTC-regulated Designated Contract Market that handles more than 60% of global prediction market volume. Kalshi generated $23.8 billion in trading volume in 2025, is valued at $11 billion, and is accessible directly through its app as well as through Robinhood, Coinbase, and other partners.
Polymarket — The world’s most recognized prediction market brand, now re-entering the U.S. through QCX, a regulated exchange it acquired for $112 million. Polymarket attracted over $3.3 billion in wagers on the 2024 presidential race alone and is currently valued at $9 billion after a $2 billion investment from the Intercontinental Exchange (the company behind the NYSE).
PredictIt — The academic-origin market that still operates under a CFTC no-action letter, with lower trading limits ($3,500 per contract, raised from $850 in July 2025) that produce a different—and, according to some research, more accurate—type of crowd signal.
Iowa Electronic Markets (IEM) — The original academic prediction market, still run by the University of Iowa, now serving more as a research benchmark than a high-volume trading venue.
These markets are sometimes called election betting markets, event contract markets, or political prediction markets. Whatever the label, they all work the same fundamental way: prices reflect the aggregated beliefs of traders who have real money at stake.
How Election Prediction Contracts Work
Most U.S. election prediction markets use binary contracts:
The question: “Will [Candidate X] win the 2028 U.S. presidential election?”
The mechanics: You can buy YES or NO shares, each worth $1 if correct and $0 if wrong at settlement. If YES trades at $0.24, the market is saying there’s about a 24% chance this candidate wins.
Paired outcomes: YES + NO prices always sum to approximately $1, reflecting the $1 maximum payout. If YES is $0.24, NO is approximately $0.76.
Profit and loss: Buy YES at $0.24 and the candidate wins → you profit $0.76 per share. The candidate loses → you lose your $0.24 per share. You can also sell early at any time before the event resolves, locking in gains or cutting losses as prices move.
Multi-outcome markets work the same way but with separate contracts for each potential winner. For example, the 2028 presidential market on Polymarket lists contracts for JD Vance, Gavin Newsom, Marco Rubio, AOC, and others. The prices across all outcomes sum to approximately 100%.
What makes these markets distinctive is that every price adjustment represents someone willing to put money behind their belief. A trader who thinks the 24% price on a candidate is too low will buy, pushing the price up. A trader who thinks it’s too high will sell. The result is a continuously updated, incentive-driven probability that reflects the collective knowledge of the market.
How Election Markets Are Created and Settled
1. Writing the Market Question
Platforms define a precise question and resolution rules. For example:
“Who will win the 2028 U.S. presidential election?” Resolution source: Associated Press, Fox News, and NBC. The market resolves once all three call the race for the same candidate.
The specification tells traders exactly which office and date are covered, which sources determine the winner, and how edge cases—recounts, withdrawals, court challenges—are handled.
The quality of these rules matters enormously. A Vanderbilt University study of the 2024 election found that vaguely defined “niche” contracts (like whether a candidate would say a specific word during a speech) were significantly less accurate than clearly defined outcome contracts.
2. Trading Before the Election
As news breaks—polls, debates, endorsements, fundraising numbers, economic data, policy announcements—traders adjust their positions. Prices move tick-by-tick, turning the market into a live forecast.
A surge in polls for a candidate pushes their contract price higher. A damaging scandal or weak debate performance triggers a sell-off. Major policy events like the State of the Union historically trigger massive volume spikes in legislative control contracts.
In 2024, Polymarket’s presidential market showed extensive price movement that frequently diverged from polling averages. Research from the National Bureau of Economic Research found that starting around October 18, 2024, Polymarket and polling data were predicting meaningfully different outcomes—with Polymarket proving closer to the actual result.
3. Election Night and Resolution
After voting ends, data arrives from networks and wire services. The platform applies its resolution rules—watching the specified sources until the conditions are met (for example, AP, Fox, and NBC all calling the race for the same candidate).
Once resolved, payouts occur automatically. On crypto platforms like Polymarket, resolution is encoded on-chain via an oracle system. On regulated U.S. venues like Kalshi, the exchange and its clearinghouse handle settlement under CFTC rules. Winning contracts pay $1. Losing contracts pay $0.
What’s Being Traded Right Now: The 2026 and 2028 Markets
Election prediction markets in February 2026 are more active than at any comparable point in a non-presidential election year. Here’s a snapshot of the major markets:
2028 Presidential Race
Over $298 million has been traded on Polymarket’s 2028 presidential winner market. Current front-runners include JD Vance at approximately 24% (the leading candidate and the favorite for the Republican nomination at 50%), Gavin Newsom at roughly 20% (with 30% odds for the Democratic nomination), Marco Rubio at approximately 8.5%, and AOC at about 5.9%. Party-level odds show Democrats at 55% and Republicans at 43%.
2026 Midterms
House control markets on Kalshi have attracted over $5.1 million in trading volume, with Democrats favored at 77–78% and Republicans at 21–22%. Senate control markets show Republicans favored at 60% and Democrats at 40%, with $1.6 million traded. The most likely overall outcome is divided government.
Policy and Political Event Markets
The 2026 election prediction market ecosystem extends well beyond who wins and loses. Active markets include the Fed chair nomination ($490 million in volume on Polymarket, with Kevin Warsh at 94%), Supreme Court tariff ruling (over $9 million wagered), government shutdown timing and duration, Greenland acquisition odds, Ukraine ceasefire timelines, and dozens of other policy-adjacent questions.
Kalshi has partnered with 270toWin for live Senate race odds, and both CNN and CNBC have integrated prediction market data into their political coverage—a significant step toward mainstream legitimacy.
How Accurate Are Election Prediction Markets?
This is the most debated question in the industry, and the answer depends on what you measure and how.
The 2024 Track Record
Prediction markets got the big call right. Polymarket had Trump leading Harris 58% to 42% on the eve of the 2024 election—a margin that proved far more accurate than most polls, which showed the race at roughly 50-50. A University of Cincinnati economist noted that prediction markets demonstrated a real-world use case beyond pure speculation.
At the state level, NBER research found that Polymarket provided more accurate forecasts for battleground states than polling averages, particularly in states that were polled less frequently (where polls tended to be “stale”).
The Vanderbilt Study: A More Complicated Picture
A landmark study by Vanderbilt University researchers Joshua Clinton and TzuFeng Huang, released in late 2025, analyzed more than 2,500 political prediction markets across Kalshi, PredictIt, Polymarket, and IEM during the final five weeks of the 2024 campaign—representing over $2 billion in transactions.
Their findings were more nuanced than the industry’s victory lap suggested. PredictIt correctly predicted outcomes 93% of the time, compared to 78% for Kalshi and 67% for Polymarket. However, the researchers noted that once you account for the types of markets each platform offers, the accuracy gap between Kalshi and Polymarket narrows considerably. Polymarket’s lower headline number was partly driven by its many “niche or low-information” contracts—speculative bets with little public data to guide traders.
More concerning, the study found limited evidence of market efficiency: prices for identical contracts diverged across exchanges, daily price changes were weakly correlated or negatively autocorrelated, and arbitrage opportunities actually peaked in the final two weeks before Election Day.
The Bottom Line on Accuracy
The research suggests that election prediction markets are genuinely useful forecasting tools for major, well-defined questions (who wins the presidency, which party controls the House) where there’s abundant public information and high trading volume. They are less reliable for niche, speculative, or low-information questions. And even for major questions, they are not perfectly efficient—prices can diverge across platforms, and large individual traders (like the famous French “whale” who bet $30 million on Trump in 2024 and won $85 million) can temporarily move prices.
One factor that affects accuracy is participant demographics. PredictIt’s trading limits ($3,500 per contract) create what researchers call “enforced diversity”—a crowd of many small bettors rather than a market dominated by whales. Polymarket’s crypto-only requirement means its users represent a narrower demographic (estimates suggest only 17–20% of Americans own cryptocurrency). Kalshi’s mainstream accessibility through Robinhood and Coinbase is broadening its user base but also bringing in less-experienced “retail” traders whose influx can create noise.
How Traders Use Election Prediction Markets
People participate in election markets for different reasons:
Forecasting and information. Analysts, journalists, campaign strategists, and everyday voters watch market odds as an alternative or complement to polls. CNN and CNBC now display prediction market odds alongside traditional polling data in their political coverage.
Speculation. Traders try to profit by being more accurate or faster than the crowd. In the 2024 election, one Polymarket trader netted $85 million from a large Trump position that proved correct.
Hedging. Someone whose income, investments, or business is exposed to a particular political outcome might use prediction markets to offset that risk. A farmer worried about tariff policy could buy contracts that pay if tariffs are expanded, for example.
A typical trader workflow: Pick a platform legal in your state. Deposit funds (USD on regulated exchanges like Kalshi; USDC stablecoins on Polymarket’s global platform). Choose an election market and read the resolution rules carefully. Buy or sell contracts based on your assessment of the true odds. Adjust positions as new information arrives, or hold through the election and settlement.
The Legal and Regulatory Landscape in 2026
The legal environment for election prediction markets has shifted dramatically since December 2025. The landscape now involves an unprecedented clash between federal and state authority.
Federal Level
The CFTC regulates event-contract exchanges like Kalshi as Designated Contract Markets under the Commodity Exchange Act. The Trump administration has been aggressively supportive of this federal framework. CFTC Chairman Brian Selig published a Wall Street Journal op-ed declaring that the agency would no longer tolerate states undermining its exclusive jurisdiction. On February 17, 2026, the CFTC filed an amicus brief backing Kalshi and Polymarket against state enforcement actions.
State-Level Challenges
More than 20 states have challenged prediction market operations, arguing that event contracts—especially on sports, but sometimes extending to political markets—are functionally gambling requiring state licensing. Nevada’s Gaming Control Board has led the charge, suing Kalshi, Polymarket, Crypto.com, and Robinhood. Massachusetts issued a preliminary injunction in January 2026. New York Attorney General Letitia James warned residents that prediction markets lack consumer protections.
The Ninth Circuit Showdown
Kalshi, Crypto.com, and Robinhood have all filed separate lawsuits against Nevada’s Gaming Control Board. The regulator is seeking to consolidate these into a single Ninth Circuit Court of Appeals decision, with oral arguments expected in April 2026. The outcome will likely set the legal framework for the entire industry.
Conflict of Interest Concerns
The Trump administration’s support has drawn scrutiny because Donald Trump Jr. is both a strategic advisor to Kalshi and an investor in Polymarket through his venture capital firm. Any favorable CFTC decisions could financially benefit the president’s family.
Insider Trading
The rapid growth of political prediction markets has raised serious concerns about insider trading by government officials. A Washington Post investigation in January 2026 found that more than $200 million is staked on political or government action markets. The most dramatic incident involved a Polymarket user who placed trades hours before any public announcement of the Venezuela/Maduro operation and netted $436,000. Representative Ritchie Torres has introduced legislation to ban federal employees and elected officials from trading on political outcomes they could influence.
Risks of Trading Election Prediction Markets
Even on regulated platforms, trading election markets carries real risk:
Market risk. Prices can move quickly and you can lose 100% of the money staked on a given contract. Election outcomes can surprise even the best-informed markets.
Legal risk. Access varies by state and is actively changing. If you live in Nevada, New Jersey, Massachusetts, or other states with active enforcement actions, your access to specific platforms may be restricted or could change without warning.
Information asymmetry. Some traders have better data, faster access to news, or more sophisticated models. The Vanderbilt study found that arbitrage opportunities persisted even in the most liquid election markets, suggesting that not all traders are equally well-informed.
Insider trading risk. Political prediction markets are uniquely vulnerable to people with non-public knowledge of government decisions. Unlike stock market insider trading, there is currently no comprehensive federal law specifically prohibiting officials from trading on outcomes they influence (though the Torres bill would change this).
Platform and rule risk. Ambiguous contract wording, recounts, legal challenges, or platform technical issues can create unexpected outcomes. The ISW/Polymarket controversy over Ukraine maps in late 2025—where a prediction market resolution hinged on a map that was later corrected—illustrates how even well-intentioned rules can produce disputed outcomes.
Demographic bias. Prediction market participants are not a representative sample of voters. Polymarket’s crypto-only access skews toward a younger, more male, more affluent demographic. This means market prices may reflect the views of a specific population rather than the electorate as a whole.
How to Read Election Odds Without Getting Lost
Price equals crowd probability, not destiny. A 24% chance on JD Vance winning the presidency means roughly one-in-four scenarios end that way. It’s a real possibility, not a guarantee—and not a dismissal.
Compare across platforms. Kalshi and Polymarket sometimes price the same outcome differently. When they diverge, it usually means one platform’s trader base has information or sentiment the other doesn’t. Aggregator sites like Oddpool and PredictionHunt make these comparisons easy.
Watch moves around catalysts, not just static levels. The biggest information comes from how prices change after major events: debates, primaries, policy announcements, court rulings, economic data releases, and endorsements. A 5-point swing after a debate tells you more than a static number on a quiet Tuesday.
Understand the accuracy limits. Prediction markets are strong at forecasting well-defined, high-information questions with high trading volume. They’re weaker on niche, speculative, or low-liquidity questions. And even for major races, they can be temporarily distorted by whale activity or demographic bias.
Remember the legal landscape is shifting. Platform availability varies by state. The Ninth Circuit ruling expected later in 2026 could dramatically change which platforms operate where. Always check your local laws before participating.
The Bottom Line
U.S. election prediction markets have evolved from a bar-napkin idea in Iowa City in 1988 to a multi-billion-dollar industry that now rivals traditional polling as a forecasting tool. They correctly called the 2024 presidential race more accurately than most polls, attracted over $3.3 billion in election-year volume, and are now integrated into CNN and CNBC political coverage.
But the industry’s explosive growth has also exposed real tensions: between federal and state regulators, between market accuracy and market efficiency, between democratized information and insider trading risk, and between the promise of crowd wisdom and the reality that prediction market crowds are not representative of the electorate.
For anyone watching U.S. politics in 2026—whether as a voter, an analyst, or a trader—election prediction markets are now an essential tool for understanding what the informed money thinks will happen next. Just remember: even the most accurate prediction market is still a market, not a guarantee.
This is not investment, trading, or legal advice. Odds change constantly and availability varies by state. Always check your local laws before participating in any prediction market.
