A trader opens Polymarket, scrolls to the market on whether a particular central bank will cut rates at its next meeting, and sees a single number sitting next to the Yes button: 73¢. That is the entire pitch. Seventy-three cents to buy a contract that pays one dollar if the answer is yes, nothing if it is no. The number is doing a lot of work, and most new readers stare at it for a second too long before realising what it actually means.

Reading Polymarket odds is the foundational skill. Every chart, every news take, every "the market thinks" claim in a tweet starts from the same cent price. Get the translation right and the rest of the platform starts to make sense; get it wrong and you will spend months mistaking volume for conviction, or treating a 60/40 market as a sure thing. This guide walks through what the price is, how to convert it, where the spread hides information, and the traps that catch nearly every newcomer at least once.

The price is the probability, almost

A Yes share priced at 73¢ on Polymarket implies the market thinks there is roughly a 73% chance of the outcome happening. That is the whole trick. The contract pays $1 if Yes resolves true and $0 if it does not, so the price someone is willing to pay for it, in a reasonably liquid market, is the market's collective estimate of that probability.

Why "almost"? Because the price reflects more than pure probability. It reflects fees, the time value of money locked up until resolution, and any imbalance between eager buyers and reluctant sellers. On a fast-moving political contract resolving next week, the gap between price and true probability is small. On a contract resolving in eighteen months, the price will sit slightly below the "true" probability because traders demand a small premium to tie up capital that long. Worth knowing the distinction exists, even if you do not need to compute it.

The deeper mechanics live in our explainer on what implied probability is and how prediction market prices translate to odds. The short version: divide the cent price by 100 and you have your probability.

Yes and No should add up to one dollar

Here is a check that catches a surprising number of mistakes. On any Polymarket binary contract, the Yes share and the No share should sum to roughly $1. If Yes is 73¢, No should be around 27¢. If they sum to 98¢ or $1.02, something is happening, usually a temporary liquidity gap that arbitrageurs will close within minutes.

Why does this matter for a reader? Because it gives you a second number. A market with Yes at 60¢ and No at 40¢ is a clean coin-flippy contract. A market with Yes at 60¢ and No at 38¢ has a 2¢ gap that tells you the order book is thin and the displayed prices are stale or wide. The same Yes price means very different things in those two situations.

Our guide to yes shares versus no shares walks through why they trade as separate instruments and what the relationship between them actually represents. The relationship is the thing. Reading one side without checking the other is reading half the market.

The order book hides the real story

A single quoted price is a summary. Behind it sits an order book, a ladder of bids and offers at different prices, and on Polymarket you can usually see the top of that book if you click into the market. The headline 73¢ might be the price of the last trade, or the midpoint between the best bid and best offer, or the best offer itself. Different surfaces show it differently.

Why this matters: if the best bid for Yes is 71¢ and the best offer is 75¢, the "price" of 73¢ is a polite fiction. You cannot actually trade at 73¢. You can sell at 71¢ or buy at 75¢, and the 4¢ gap is the spread you would pay to cross the book. On a thinly traded contract on a long-dated question, that spread can be 10¢ or more, which means the "market probability" has a 10-point error bar baked into it.

Look at the depth too. A market quoted at 73¢ with $200,000 sitting on the bid is very different from one quoted at 73¢ with $400 sitting on the bid. The second one moves on a single decent trade. Our piece on what liquidity is in prediction markets and why some contracts move on tiny volume explains why the same headline number can mean conviction in one market and accident in another.

Volume tells you how much to trust the price

Is a 73% probability backed by ten million dollars of trading or by eight hundred? Polymarket displays volume prominently on each market page, usually as a total since inception and a 24-hour figure. Both matter, and they matter in different ways.

Total volume tells you whether the market has ever attracted serious attention. A presidential election contract with $400 million of total volume has been argued over by thousands of traders for months. A contract on whether a specific cabinet reshuffle will happen by a certain date with $12,000 of total volume has been mostly ignored. The first price is a consensus. The second is a few traders' guess.

Twenty-four-hour volume tells you whether the market is currently active. A contract with $50 million of lifetime volume but only $300 in the last day is essentially stale. The price you see is whatever the last person paid before the topic stopped being interesting, which could have been hours ago. Treat low recent volume as a flag that the displayed probability lags reality.

The mid-market price is the one to quote

When journalists or analysts cite "Polymarket's odds" for something, the cleanest number to use is the mid-market price: halfway between the best bid and the best offer. That number is the platform's tightest estimate of the contract's fair value at this moment.

Last-trade price is fine if the market is busy. It becomes misleading the moment trading slows down, because the last print could be from an hour ago and the book around it has since moved. If you see a market quoted at 73¢ but the bid is 65¢ and the offer is 68¢, the real number is around 66¢ or 67¢. The 73¢ is just an old transaction that has not been overwritten on the headline.

This is the single most common mistake in news articles citing prediction market prices. A confident "the market gives this a 73% chance" can be six points off the actual current consensus, simply because the writer pulled the last-trade number rather than checking the book.

Time decay is real on long-dated markets

A contract that resolves in three years and pays $1 if a certain event happens will not trade at its "true" probability. It will trade meaningfully below. If you think the chance of the event is 50%, you might still see the contract priced at 42¢ or 44¢, because the eight to ten cents of difference is the cost of locking your capital up for three years when you could have earned interest elsewhere.

This is not a flaw in the market. It is the market correctly pricing in the opportunity cost of holding the position. But it matters for reading the number. A long-dated 35¢ contract does not imply a 35% probability; it implies something a bit higher, after you mentally back out the time value. Short-dated contracts, resolving within days or weeks, do not have this problem in any meaningful size, so for most news-driven markets the price-to-probability translation is clean.

The broader mechanics of how prediction markets actually generate these prices, from order books to the resolution payout structure, are covered in our deeper read on how prediction market odds work.

What the price does not tell you

A Polymarket price is a probability, weighted by capital. It is not a forecast in the academic sense, and it is not a poll. It does not tell you why traders believe what they believe. It does not tell you whether one large trader is dominating the book or whether thousands of small traders converged on the same answer. It does not tell you anything about the quality of the resolution criteria, which matter enormously: a market that looks like it is about "who wins the election" might actually resolve on a narrower technicality buried in the rules.

Always check the resolution criteria. Always check the volume. Always check the spread. The cent price is the headline, but the headline is rarely the whole article.

iPredicta tracks Polymarket contracts alongside Kalshi, Smarkets and the UK exchanges, surfacing the price moves, the spreads and the resolution language together so readers do not have to triangulate across five tabs. The goal is to make the underlying number legible, then explain what it is actually telling you.

Frequently asked questions

What does the cent price on Polymarket actually mean?

The cent price is the cost to buy one share of that outcome, and it implies the probability the market assigns to it. A Yes share priced at 64¢ implies the market thinks there is roughly a 64% chance the answer resolves Yes; the share pays $1 if it does and $0 if it does not. That translation is the foundational skill on Polymarket and on any binary prediction market. The only caveats are that the price also baked in a small adjustment for time value on long-dated contracts and reflects whatever fees and spreads exist in the book. For short-dated, liquid markets the cent-to-percent translation is clean enough to use directly.

Why do Yes and No prices sometimes not add up to exactly $1?

Yes and No should sum to roughly $1 on any Polymarket binary contract, but small gaps appear constantly because the order book is not always perfectly balanced. If Yes is bid at 60¢ and No is bid at 39¢, the 1¢ shortfall reflects a thin moment in the book that arbitrageurs typically close within minutes. If you see a persistent gap of several cents, it usually means the market is illiquid and the displayed prices are wide or stale. Always check both sides. A clean sum near $1 with a tight spread tells you the market is functioning; a sum well below $1 with a wide spread tells you to trust the quoted probability less.

Is the last-trade price the same as the current market probability?

Not always, and on quiet markets the difference can be large. The last-trade price is the price someone actually paid most recently, which could have been minutes or hours ago. If the market has been quiet since then, the bid and offer may have drifted, and the true current consensus is the midpoint between the best bid and the best offer right now. On busy contracts with frequent trading, the last print and the mid-market price are usually within a cent of each other. On slow contracts, they can be five to ten cents apart, which is the gap between a stale headline and the market's actual current view. Use the mid-market price when you can.

Does high volume mean the price is more accurate?

Higher volume generally makes the price more trustworthy, because more capital has been deployed to argue about the answer. A presidential market with hundreds of millions in lifetime volume reflects a consensus among thousands of traders; a niche contract with a few thousand dollars of volume reflects a handful of guesses. But volume alone is not enough. A market can have huge lifetime volume yet have gone quiet in the last day, in which case the price has not been refreshed against current news. Check both the all-time volume and the 24-hour volume. The combination tells you how seriously the market has been argued over and whether the argument is still active.

Should I trust a Polymarket probability over a poll?

Often yes, with important caveats. Markets aggregate diverse information including polls, news, expert analysis and private knowledge, and the requirement to put real money behind a view filters out cheap talk. Multiple studies have shown prediction markets beating polls on election forecasting accuracy across many cycles. But markets can be wrong, especially when liquidity is thin, when the resolution criteria are ambiguous, or when a small number of large traders dominate. Polls, by contrast, give you a clean snapshot of stated intent at one moment. The two are complementary tools, not competitors. Our piece on how prediction markets compare with polls digs into when each one wins.