What Is a Prediction Market?

A prediction market is an exchange where you trade contracts that pay out based on the outcome of a future event. The price of each contract is the market's estimate of how likely that outcome is. Buy a contract for 65 cents and the market is telling you it thinks the event has roughly a 65 percent chance of happening. If you are right, the contract settles at $1. If you are wrong, it settles at $0.

That is the whole idea: a price between 0 and 1 that doubles as a probability. Everything else is mechanics.

How a contract prices a probability

Most prediction-market contracts are binary. There is a Yes side and a No side, and the two prices always add up to $1. If Yes trades at $0.70, No trades at $0.30. The event either happens or it does not, so exactly one side pays out a dollar and the other pays nothing.

Concrete example. A contract asks: "Will the Fed cut rates at the next meeting?" If traders collectively believe the answer is 70 percent likely, Yes settles toward $0.70 and No toward $0.30. Buy 100 Yes shares at $0.70 ($70 in). If the Fed cuts, you collect $100, a $30 profit. If it does not, you lose the $70. Your No counterparty has the mirror-image bet.

The price moves the way any order-book price moves. Buyers who think Yes is underpriced bid it up; sellers who think it is overpriced push it down. The equilibrium price is a live, money-weighted estimate of the odds, updating in seconds as news lands. That is the feature pollsters and pundits cannot match: a number with real capital behind every basis point of it.

What you can trade

Markets exist for almost anything with a clear, verifiable resolution. The high-volume categories in 2026:

• Sports. The dominant category by far. Sports has made up roughly 80 percent of Kalshi's volume since mid-2024, and it drove much of the surge that pushed combined Kalshi and Polymarket volume to about $24 billion in a single month by April 2026.
• Politics and elections. The category that put prediction markets in headlines, and where the calibration debate is sharpest.
• Economics. Rate decisions, CPI prints, jobs numbers, GDP. Contracts resolve against official data releases.
• Crypto and culture. Price thresholds, product launches, award shows, and other events with an unambiguous outcome and date.

The hard requirement is a resolvable question with a fixed deadline and an objective source of truth. "Will BTC close above $150k on Dec 31?" works. "Will crypto have a good year?" does not.

Why traders use them

Three reasons, in rough order of importance.

First, calibrated probabilities. Because traders are paid for being right and lose money for being wrong, the price tends to track reality more honestly than a poll or a hot take. This is the wisdom-of-crowds argument. Worth a caveat, though: a 2026 study by Gomez-Cram, Guo, Jensen, and Kung found that accuracy comes from a small informed minority, roughly 3.5 percent of accounts, while the rest largely fund price discovery rather than create it. Political markets in particular tend to show prices compressed toward 50 percent. The crowd is smart on average, not on every contract.

Second, a tradable view on things the stock market does not price directly. You can take a position on an election, a rate decision, or a championship without constructing some awkward equity proxy.

Third, liquidity before the event resolves. You do not have to hold to settlement. If your contract moves from $0.40 to $0.65 because news broke in your favor, you can sell and book the gain immediately, the same way you would close any position.

Polymarket and Kalshi, and the US-legal question

The two venues that matter most run on different rails, and Polymarket in particular splits into two separate entities you should not confuse.

Kalshi is a CFTC-regulated exchange. You fund an account in US dollars and trade event contracts the way you would trade futures on any regulated venue. It is open to US persons and posted record monthly volume in early 2026, hitting roughly $14.8 billion in April.

Polymarket's international platform runs on the Polygon blockchain. You trade with USDC, it is non-custodial so you hold your own funds, and outcomes resolve through the UMA optimistic oracle, a permissionless system where anyone can propose a result and anyone can dispute it. This is the larger, crypto-native book, and it blocks US persons.

Polymarket US is a separate, CFTC-regulated entity, built on the QCEX exchange and clearinghouse Polymarket acquired in 2025. It has full KYC and AML, its own order book, liquidity, and balances, and it began rolling out to US traders in 2026. Volume there is smaller than the international book for now, and it launched sports-first. The key point: it is a distinct venue from the international Polymarket, not the same book with a US login.

The practical takeaway for a US trader: Kalshi and Polymarket US are the legal, regulated doors, and both are CFTC-supervised. The international Polymarket site is not for you.

From one bet to a strategy

A single contract is a bet. A repeatable edge is a strategy: a rule like "buy any Yes priced under $0.10 inside the final hour when the order book is thin," applied consistently with a risk cap. Prediction markets are unusually good ground for this because every contract has a clean numeric price, a hard resolution, and an order book you can read programmatically.

That is where automation earns its keep, and it is what Banger is built for. You write a strategy in Python, paper-trade it against the live Polymarket and Kalshi order books, then run it live 24/7 under a declarative risk envelope: per-trade cap, daily loss stop, max open positions, kill switch. Banger never custodies funds; you bring your own venue keys. The market gives you a clean probability. What you do with it is the actual job.

The one-line version

A prediction market turns a future event into a tradable price between 0 and 1, where the price is the odds. You buy the side you think is mispriced, and you get paid a dollar if you are right. The reason serious traders pay attention is simple: a number backed by real money tends to tell the truth more often than one that is not.