What is Polymarket? A Complete Guide to the Largest Prediction Market

Guide · 7 min read · published 2026-04-20 · updated 2026-07-25

Polymarket is the largest prediction market in the world. People buy and sell YES/NO shares on real-world questions - who wins an election, whether bitcoin closes above a price, which team lifts the trophy - and the price of a share is the crowd's estimate of how likely that outcome is. This guide covers how that works, what it costs, how questions get settled, and where the model breaks down.

The one-sentence version

A Polymarket share pays exactly $1.00 if the outcome happens and $0.00 if it does not, so the price you pay - 34¢, 71¢, 96¢ - is the market's probability, expressed in cents.

How a market actually works

Every question has two sides, YES and NO, and their prices always add up to $1.00. Suppose "Will BTC close above $100k on Dec 31?" trades at YES 67¢ / NO 33¢. The market is saying: roughly a 67% chance.

• Buy YES at 67¢. If it resolves YES you receive $1.00 per share - a profit of 33¢ on 67¢ risked, about +49%.
• Buy NO at 33¢. If it resolves NO you receive $1.00 - a profit of 67¢ on 33¢ risked, about +203%.
• Sell before resolution. You are never locked in. If the price moves from 67¢ to 82¢ you can take the 15¢ and walk away without waiting for the event.

That last point is what makes prediction markets feel more like trading than betting. A sportsbook ticket is frozen until the whistle; a market position is a token you can sell to anyone at any moment.

A worked example

You put $500 into YES at 40¢. That buys 1,250 shares. Three things can happen:

• It resolves YES. 1,250 × $1.00 = $1,250. Profit $750.
• It resolves NO. The shares are worth nothing. Loss $500.
• News moves the price to 55¢ and you sell. 1,250 × $0.55 = $687.50. Profit $187.50, banked weeks before the event.

Notice that the third path is the one most active traders take. You are trading the probability, not the outcome.

Why the price is the probability

If you believe the true chance is 80% and the market says 67%, buying YES has positive expected value: you are paying 67¢ for something worth 80¢ in expectation. Anyone who spots that gap buys, which pushes the price up until the gap closes. Multiply that by thousands of traders with real money at stake and you get a number that is hard to beat.

This is why prediction markets have a good record against polls and pundits, particularly in the final weeks before an event, when new information arrives fastest and the incentive to be right is sharpest. A poll asks people what they think. A market asks people to fund what they think - and takes money away when they are wrong.

The mechanism is not magic. Markets are wrong all the time, especially early, in thin books, and on questions where nobody has an information edge. But they are wrong in a way you can measure: the price is a public, timestamped, falsifiable forecast.

What you can trade

• Politics and elections - primaries, general elections, nominations, cabinet appointments, control of the House and Senate.
• Geopolitics - ceasefires, sanctions, treaties and conflict questions on Ukraine, the Middle East and China.
• Crypto - bitcoin and ethereum price targets, ETF decisions, protocol launches, plus short-dated contracts that settle within the hour.
• Sports - NFL, NBA, soccer, tennis, UFC and Formula 1, from season futures to single games.
• Macro - Fed rate decisions, CPI prints, recession calls, company earnings.
• Culture - award shows, box office, AI model releases, Time Person of the Year.

New questions appear within hours of a news story, which is the practical difference between Polymarket and a regulated venue where every contract needs approval first.

How markets get resolved

This is the part beginners skip and later regret. Each market has a written resolution rule, and settlement runs through an oracle rather than a company deciding what it feels the answer is. If the outcome is disputed, the dispute goes through a challenge process before the payout is finalised.

Two habits protect you. First, read the rules text before you trade, not after - the question title is a summary, the rules are the contract, and the two are not always as close as you expect. Second, look at the exact source and deadline named in the rules. "Will X happen by December 31?" resolves on what the named source reports by that date, not on what you saw on television.

Most losses that feel unfair trace back to a rule the trader never read. Every market page on Polynter shows the full resolution text alongside the odds for exactly this reason.

What it costs

There is no deposit fee, no withdrawal fee from Polymarket itself, no custody fee, and no gas on trades - orders are matched off-chain and settled by the exchange contracts, so you sign but do not pay for the transaction.

Trading fees apply to takers in some categories. The fee scales with uncertainty rather than being a flat percentage, so it peaks near 50¢ and shrinks to almost nothing near the edges of the price range. The full breakdown, including maker rebates, is in our fee guide.

The cost most people forget is the spread. If a market shows 61¢ bid and 65¢ ask, crossing that spread costs 4¢ - usually more than the fee. In thin markets it can be 10¢ or worse.

How to start

1. Connect a wallet. MetaMask, Coinbase Wallet, Rabby and WalletConnect all work. Your positions live in a Polymarket-derived wallet controlled by your signer.
2. Fund it. Trading collateral is pUSD, which wraps a dollar-equivalent stablecoin one-for-one. The bridge accepts transfers from major EVM chains, Solana and Bitcoin - see deposits and withdrawals.
3. Approve once. The first trade needs one approval batch so the exchange contracts can move collateral and outcome tokens. It is relayed, so it costs you nothing.
4. Research before you click. Check the order book depth, the resolution rules and the end date. Polynter shows all three on every market page.
5. Start small. Liquidity varies enormously - some markets hold millions in depth, others a few hundred dollars. Your first trade should be one you can afford to get wrong.

The risks nobody advertises

• Resolution risk. Ambiguous wording and disputed outcomes are the single biggest source of unpleasant surprises.
• Liquidity risk. Getting in is easy. Getting out of a thin market at a fair price is not.
• Self-custody risk. Nobody can freeze your position, and equally nobody can recover it if you lose access to your wallet.
• Time risk. Capital parked in a market resolving in eight months is capital that cannot do anything else.
• Availability. Access depends on where you are; check the platform's own terms rather than a blog post, including this one.

A short glossary

• Share / outcome token - the thing you own, worth $1.00 if correct.
• Maker - you post a limit order that rests on the book. Taker - you cross the spread and fill immediately.
• Spread - the gap between the best bid and the best ask.
• Merge - convert a matched YES and NO pair back into collateral without waiting for resolution.
• Redeem - burn winning shares after resolution to receive $1.00 each.
• Multi-outcome market - a question with several candidates, where each candidate is its own YES/NO market and the prices should sum to about 100%.

Where Polynter fits

Polynter is an independent analytics terminal built on top of Polymarket. It is not affiliated with the platform and it never holds your funds. It exists for the research half of the job: live odds on every market, the trader leaderboard with full position history for any wallet, a screener for near-certain outcomes, live holding-reward APYs, LP reward tracking and a public API.

If you are still finding your feet, read how to read prediction market odds next - it is the arithmetic that turns a price into a decision.

More Polymarket guides →