How to Read Prediction Market Odds: From Cents to Probabilities
Guide · 6 min read · published 2026-04-20 · updated 2026-07-25
Every Polymarket price is a probability in disguise. A YES share at 42¢ means the market thinks there is roughly a 42% chance the event happens. Once you can move fluently between cents, percentages, payout ratios and sportsbook odds, you can compare any forecast to any other - and spot the ones that do not add up.
1. Cents are percent
Because YES and NO always sum to $1.00, you read the price directly as a probability. No conversion table, no vig to strip out:
YES @ 0.42- the market implies about a 42% chance.YES @ 0.93- about 93%.YES @ 0.005- half a percent, and still tradable as a cheap tail-risk hedge.
This is the underrated feature of prediction markets. A sportsbook quoting -180 has a house margin baked in, so the implied probability it advertises is not the probability it actually believes. A two-sided market that sums to $1.00 has no margin to strip.
2. Payout per share
Winning shares pay exactly $1.00, so your profit per share is 1 − price and your risk is the price itself:
- Buy YES at 67¢ - risk 67¢ to make 33¢, a 0.49x payoff.
- Buy NO at 12¢ - risk 12¢ to make 88¢, a 7.3x payoff.
- Buy YES at 4¢ - risk 4¢ to make 96¢, a 24x payoff.
The break-even probability is the price. At 12¢ you need the event to happen more than 12% of the time to profit over many repetitions. That framing matters more than it sounds: a 24x payoff is not a good trade if the true chance is 2%.
3. Turning a price into an annual return
For near-certain markets the interesting number is not the profit but the profit per unit of time. Buy NO at 99.5¢ on a market resolving in 30 days and you make 0.5¢ on 99.5¢ staked - about 0.5% - which annualises to roughly 6%.
The same 0.5¢ edge over 3 days is about 65% annualised; over 300 days it is about 0.6%. Identical price, wildly different trade. This is exactly why our 99%+ screener ranks by annualised return rather than by discount, and why you should never evaluate a near-certain position without looking at the end date.
On eligible markets you can stack holding rewards on top of the price discount - see the rewards guide for how those two returns combine.
4. Converting to sportsbook odds
To compare a market price against a bookmaker, convert to the format you already know:
- Decimal odds =
1 / price. YES at 0.40 becomes 2.50. - Fractional odds =
(1 − price) / price. YES at 0.40 becomes 3/2. - American odds, for a favourite (price above 0.50) =
−100 × price / (1 − price). YES at 0.80 becomes −400. For an underdog =100 × (1 − price) / price. YES at 0.25 becomes +300.
Going the other way, remember to remove the book's margin before you call a sportsbook number a probability. Add the implied probabilities of both sides of a book line: if they sum to 105%, the extra 5% is the house edge, not information. Market prices sum to 100% by construction, which is what makes the comparison meaningful in the first place.
5. The mid-price lies when the book is thin
A market displaying 50¢ might fill ten shares at 50¢ and the next thousand at 62¢. The number on the card is the top of the book, not the price you will get.
Before sizing up, look at three things on the market page: the bid-ask spread, the size resting at each level, and the 24-hour volume. A tight spread with real size behind it means the displayed price is trustworthy. A tight spread with $40 of depth means nothing at all.
Liquidity versus volume
- High liquidity, high volume - tight spreads, prices you can act on, the healthiest state.
- High liquidity, no volume - someone is quoting but nobody is trading. Often a market where the outcome is settled in practice but the rules leave doubt.
- Low liquidity - your trade is the market. Whatever you pay becomes the new price, and the exit will be worse.
6. Multi-outcome markets and the sum test
For a question like "Who wins the 2028 nomination?", each candidate is a separate YES/NO market. Add up every candidate's price. In a well-priced market the total lands close to 100%.
When it sums to 106%, the field is collectively overpriced and shorting the whole board has theoretical edge. When it sums to 94%, buying the whole board does. In practice fees, spreads and the capital tied up until resolution usually eat the gap - but a large deviation is a reliable sign that something is stale, most often a long tail of forgotten candidates still quoted at 2¢ each.
The same test flags the opposite problem: a favourite at 61% in a field summing to 88% is really priced at about 69% once you normalise, and comparing the raw 61% to a poll would mislead you.
7. Reading the price over time
A single price is a snapshot; the shape of the curve is the story. Three patterns are worth recognising:
- The step. A vertical move on news. The market has repriced and, most of the time, it has repriced correctly and fast.
- The drift. A slow grind in one direction with no headline attached - usually informed money accumulating, and worth investigating.
- The convergence. As the end date approaches, prices push toward 0 or 1. This is not information, it is time running out. Do not mistake it for momentum.
8. Comparing a market to a poll
Polls and markets answer different questions. A poll estimates what a sample says today. A market estimates what will actually happen after every remaining event between now and resolution - turnout, scandals, weather, dropouts.
So a gap between a poll at 52% and a market at 61% is not automatically an error. It usually means traders are pricing something the poll does not measure. Treat a large gap as a prompt to find out what that something is, not as free money.
A short cheat sheet
- Price = probability. 42¢ means 42%.
- Profit per share =
1 − price. Break-even probability = price. - Annualised return =
(1 − price) / price × 365 / days remaining. - Decimal odds =
1 / price. - Multi-outcome prices should sum to about 100%.
- Always check depth before trusting the displayed price.
Where the numbers already are
Polynter does this arithmetic for you: annualised return on every near-certain market, best bid and best ask side by side, depth and historical price curves on each market page, and the trader leaderboard so you can see how the wallets with a track record are positioned before you commit. If you want the underlying mechanics first, start with what Polymarket is.