The most important thing on Polymarket is not the question. It is the little number beside it. Will a candidate win? Will a central bank move rates? Will a film open above a certain box-office total? A YES share trading at 63 cents reads as a 63 percent chance. It is compact enough for a headline, restless enough for a live broadcast and legible enough to start an argument in a group chat.
That number is the product. Everything else - the order book, stablecoin collateral, market makers, smart contracts and resolution process - exists to make it move and, eventually, make it settle. Polymarket, founded in New York by Shayne Coplan in 2020, has built a marketplace where people trade on the outcomes of real-world events. But calling it only a betting site misses its more interesting ambition. The company wants its prices to function as public information.
This is why Polymarket can look like three businesses at once. To a trader, it is an exchange. To a casual visitor, it is a news feed from a slightly more anxious timeline. To a publisher or financial-data customer, it is a stream of probabilities produced by people who have money at risk.
An argument with an order book
A Polymarket contract is pleasantly blunt. Traders buy outcome shares priced between zero and one dollar. YES and NO shares are fully collateralized as a pair. Once the event is resolved, a correct share pays one dollar and an incorrect share pays nothing. Before that moment, traders can sell, change their minds or provide liquidity to others.
Ask
A precise question gets a deadline and resolution rules.
Trade
Buyers and sellers post prices for competing outcomes.
Update
News changes conviction, orders and the implied probability.
Resolve
The winning share redeems for $1; the other settles at $0.
The company is careful to say that it does not set the odds. Supply and demand do. That distinction separates the model from a traditional sportsbook, where the operator posts a line and manages its exposure. On Polymarket, participants take opposite sides and the platform supplies the venue. The result resembles a financial market, although the underlying assets are discrete events rather than companies, currencies or barrels of oil.
“Markets seek truth.”Polymarket's four-word operating thesis
The slogan is sharper than the reality, of course. Markets do not possess truth; they aggregate incentives and available information. A thin market can be noisy. A wealthy participant can move a price. A badly phrased contract can turn a factual result into a semantic wrestling match. Even a deep market reflects what traders believe will satisfy the written rules, which is not always identical to what a reader thinks the question means.
The probability becomes the headline
Polymarket broke into mainstream view during the 2024 U.S. presidential election. More than $3 billion changed hands in its presidential-winner market, and the odds became a recurring character in political coverage. The site offered something polls could not: a number that repriced continuously, across time zones, whenever a debate landed, a court ruled or a candidate stumbled.
The election demonstrated a distribution trick worth stealing. News moves the market. The market creates a new number. That number then becomes news, drawing more readers and traders back to the market. Polymarket has since made that loop deliberate. X named it an official prediction-market partner. Dow Jones agreed to distribute its data across products and publications. Intercontinental Exchange, the owner of the New York Stock Exchange, agreed to invest up to $2 billion and become a global distributor of its event-driven data.
Sports adds another version of the same loop. Deals with UFC and Zuffa Boxing, Major League Baseball, OneFootball and the ATP Tour put probabilities next to events people are already watching. The ATP agreement goes further than a logo on a court: it combines market data with live tennis streams across a calendar of roughly 20,000 matches per season. The practical goal is obvious. A market is more useful, and more active, when it is beside the thing being decided.
A probability is a moving price, not a promise
The hard work begins after “Yes” and “No”
The interface hides a formidable stack. Polymarket's international product runs on blockchain infrastructure and uses a central limit order book. Outcome positions are tokenized, collateral has historically been denominated in USDC, and trades settle onchain. Public APIs expose markets, prices and order-book data, inviting developers to build terminals, bots, research tools and alternative interfaces.
The company acquired DeFi infrastructure startup Brahma in March 2026, a revealing purchase. Crypto made Polymarket globally composable, but wallets, bridges and token mechanics can make a mainstream product feel like tax preparation. Brahma's task is effectively to preserve the programmable rails while allowing ordinary users to notice them less.
In the United States, Polymarket now travels on a separate regulatory rail. After a 2022 settlement with the Commodity Futures Trading Commission over an unregistered event-contract platform, the global site blocked U.S. users. In 2025, Polymarket bought QCEX and its clearinghouse for a reported $112 million. QCX now operates as Polymarket US, a designated contract market under CFTC oversight. That acquisition bought something code could not manufacture: a regulated route back into the home market.
The market is only as clear as the sentence above it.
Every contract needs a deadline, an authoritative source and rules for awkward edge cases. Resolution is not back-office cleanup. It is core product design.
Trust remains the exposed nerve. In June 2026, a compromised third-party vendor injected malicious code into the website for some users, and funds were stolen. Polymarket said it contained the incident and would refund affected users in full. The episode illustrated a crypto paradox: a protocol can be non-custodial while its front end remains a powerful point of failure. Transparent settlement does not make every layer safe.
A marketplace with a media shadow
Polymarket's customer base is broader than its trader count. Retail users arrive with opinions about politics, sports, finance, technology and culture. Market makers arrive for spreads and rebates. Developers consume the APIs. Researchers treat the public ledger and order book as a laboratory. Publishers and exchanges want a data feed that says, in one clean number, what an informed crowd currently believes.
Its business model has been evolving with that audience. For years, many markets charged no trading fee, a useful subsidy for liquidity and growth. In 2026, Polymarket expanded taker fees across selected categories, with part of the proceeds funding maker rebates. It can also earn from data distribution and commercial partnerships. The company is not the house hoping a customer loses; its economic interest is in more trading, deeper books and odds worth redistributing.
Kalshi is the clearest competitor, especially in regulated U.S. event contracts. But the larger competitive set includes sportsbooks, polls, expert newsletters, financial terminals and every television panel prepared to answer, “What happens next?” Polymarket differs through its crypto-native global product, open data, cultural range and unusually strong media circulation. Kalshi has its own regulatory and distribution strengths. Neither company can claim accuracy as a permanent property. Each market must earn it through participation, liquidity and sound rules.
For a visitor who never places a trade, the useful move is simply to watch. A product manager can follow the odds of a regulatory decision. A campaign reporter can see whether a debate changed expectations before the next poll arrives. A business can monitor a market tied to an acquisition, a rate cut or an important court ruling. Traders can express a view, hedge event risk in limited circumstances or sell a position before the final result. None of those uses turns the displayed price into destiny. It is a snapshot of marginal buyers and sellers under specific rules, and the risk of loss is real. The best reading habit is to pair the headline probability with volume, liquidity, resolution language and the history of the price. A number without that context is wonderfully portable and occasionally misleading.
That is where Polymarket fits: not quite a news organization, not merely an exchange and not comfortably reducible to gambling. It is an information marketplace whose raw material is disagreement. The company takes a fuzzy public question and forces it through a narrow pipe: two sides, one price, one settlement.
The format is addictive because the future never stops supplying inventory. Elections end; the next election opens. A rate decision settles; next month's begins trading. A tennis point finishes; another serve is already in the air. Polymarket's biggest opportunity is to make the probability feel indispensable. Its biggest burden is that a number so easy to read can be much harder to trust.