How to read prediction market prices as probabilities
A prediction market price is the market's implied probability. A Yes contract trading at 63¢ implies roughly a 63% chance the event happens, because the contract pays $1 if Yes is right and $0 if it is wrong.
Once money is involved, two details change the number: which price on the screen you read, and the fee, which raises the probability you need to break even.
Why a price works as a probability
Kalshi's own explainer puts it plainly: "Each contract on Kalshi settles to $1 if your answer is right, and $0 otherwise." If an event has a 40% chance of happening, a Yes contract is worth 40¢ on average, which is why Kalshi says its prices "have a very natural interpretation as probabilities."
The conversion is division by the $1 payout:
implied probability = price ÷ $1 payout
A 7¢ price implies 7%, and 93¢ implies 93%.
The number is a price traders agree on, not a measured fact. Research on Kalshi data finds prices informative but tilted at the extremes, with long shots overpriced. Our guide on whether prediction markets are accurate covers the evidence.
Which price should you read?
Every market has more than one price on screen at any moment.
| Price | What it is | Use it for |
|---|---|---|
| Best bid | Highest resting order to buy | What you could sell for right now |
| Best ask | Lowest resting order to sell | What you would pay to buy right now |
| Midpoint | (bid + ask) ÷ 2 | Best single probability estimate when the spread is tight |
| Last trade | Price of the most recent fill | Can be stale in a quiet market |
Polymarket's help center says the probability it displays is "the midpoint of the bid-ask spread," unless the spread is wider than $0.10, in which case it shows the last traded price instead.
Kalshi's API returns only bids, because the two sides mirror each other. Its documentation states: "A YES BID at price X is equivalent to a NO ASK at price ($1.00 - X)." In Kalshi's own example, the best Yes bid is 42¢ and the best No bid is 56¢:
- Yes ask = $1.00 − $0.56 = 44¢
- Yes midpoint = (42¢ + 44¢) ÷ 2 = 43¢, an implied probability of about 43%
If you want a probability estimate, read the midpoint. If you want to know what a trade will cost, read the ask. A wide spread is a warning that the midpoint is a loose estimate.
Why Yes and No can add up to more than $1
When a Yes buyer and a No buyer match, their money totals exactly $1. Polymarket's documentation describes a Yes order at 60¢ and a No order at 40¢ matching because "$0.60 + $0.40 = $1.00."
The prices you can buy at right now are the two asks, and those usually add up to a bit more than $1. Using Kalshi's example book:
- Yes ask 44¢ + No ask 58¢ = $1.02
- Yes bid 42¢ + No bid 56¢ = $0.98
The extra 2¢ is the spread. It plays the same role as a sportsbook's overround, except it goes to the traders whose orders you hit rather than to a house. Our prediction markets vs sportsbooks guide compares the two costs.
To strip the spread out of the asks, divide by their sum: 44 ÷ 102 ≈ 43.1%, which lands next to the 43% midpoint.
On a single order book the asks cannot sum to less than $1, because each Yes bid is someone's No ask. Across two platforms, prices for the same event can differ enough that they do. Our price-gap data tracks how often that happens. Buying both sides across venues is not risk-free: the two markets may use different resolution rules or sources, and fees can erase the gap.
Converting a price to American and decimal odds
If you think in sportsbook odds, these formulas convert a price p (in dollars) before fees:
- Decimal odds = 1 ÷ p
- American odds, p below 0.50 = +100 × (1 − p) ÷ p
- American odds, p above 0.50 = −100 × p ÷ (1 − p)
- At exactly 0.50, the odds are even (+100)
| Price | Implied probability | Decimal odds | American odds |
|---|---|---|---|
| 10¢ | 10% | 10.00 | +900 |
| 20¢ | 20% | 5.00 | +400 |
| 40¢ | 40% | 2.50 | +150 |
| 50¢ | 50% | 2.00 | +100 |
| 60¢ | 60% | 1.67 | −150 |
| 75¢ | 75% | 1.33 | −300 |
| 90¢ | 90% | 1.11 | −900 |
A quick check at 40¢: $40 buys 100 contracts. If Yes wins you collect $100, a $60 profit on $40, which is exactly what +150 pays.
Going the other way, a sportsbook price of −110 converts to 110 ÷ 210 = 52.38%, and +150 converts to 100 ÷ 250 = 40%.
How fees change your break-even probability
The price is not your full cost. Kalshi's fee schedule (last updated and effective July 7, 2026) sets the taker fee at:
round up(0.07 × C × P × (1 − P))
where C is the number of contracts and P is the price in dollars. Some series use a different multiplier; see our Kalshi fees guide, or run your own numbers in the fee calculator.
Your break-even probability is your total cost divided by the payout:
break-even = (C × P + fee) ÷ C
Worked example: 100 Yes contracts at 40¢ on Kalshi, as a taker
- Contract cost: 100 × $0.40 = $40.00
- Fee: 0.07 × 100 × 0.40 × 0.60 = $1.68
- Total cost: $41.68
- Payout if Yes: $100.00
- Break-even probability: 41.68 ÷ 100 = 41.68%
- Effective American odds: +100 × 58.32 ÷ 41.68 ≈ +140, down from +150 before fees
Buying at 40¢ pays off over time only if the true chance is above about 41.7%, not 40%.
| Price | Kalshi taker fee per 100 | Break-even probability | Odds before fee | Odds after fee |
|---|---|---|---|---|
| 10¢ | $0.63 | 10.63% | +900 | +841 |
| 25¢ | $1.3125 | 26.31% | +300 | +280 |
| 40¢ | $1.68 | 41.68% | +150 | +140 |
| 50¢ | $1.75 | 51.75% | +100 | −107 |
| 75¢ | $1.3125 | 76.31% | −300 | −322 |
| 90¢ | $0.63 | 90.63% | −900 | −967 |
The fee adds the most probability points near 50¢. At the extremes it adds fewer points but takes a bigger share of your possible profit: at 90¢, the $0.63 fee eats 6.3% of the $10 you stand to win.
Selling before settlement is a second trade with its own fee. Holding to settlement costs nothing extra, because Kalshi's schedule says "There is no settlement fee."
Polymarket US (QCX LLC) uses the same formula shape. Its CFTC filing set the taker coefficient at 0.0695 effective September 17, 2026, so break-even points there sit a hair below Kalshi's. See Polymarket fees for exceptions such as combos.
Common mistakes when reading prices
- Treating the last trade as the current price. In a thin market the last fill may be hours old. Check the bid and ask.
- Comparing a midpoint on one platform with an ask on another. Compare like with like, ideally the ask plus fee. Our live comparison shows both platforms side by side.
- Ignoring fees on favorites. A 90¢ contract needs better than 90.63% to break even as a taker on Kalshi.
This guide is general information, not investment advice. Every contract you buy can expire worthless, so you can lose the full amount you pay.
FAQ
What does a 70 cent price mean on Kalshi or Polymarket?
It implies roughly a 70% chance the event happens, because the contract pays $1 if Yes wins and $0 if it loses. It is the traders' current estimate, not a guarantee.
How do I convert a prediction market price to American odds?
Below 50¢, use +100 × (1 − p) ÷ p; above 50¢, use −100 × p ÷ (1 − p), with p in dollars. A 40¢ price is +150 and a 75¢ price is −300 before fees.
Why do the Yes and No prices add up to more than $1?
The prices you can buy at are the asks, and the gap between bid and ask (the spread) pushes their sum above $1. In Kalshi's documented example, a 44¢ Yes ask plus a 58¢ No ask totals $1.02.
Should I use the last traded price or the midpoint?
Use the midpoint of the best bid and ask when the spread is tight. Polymarket itself switches to the last traded price only when the spread is wider than $0.10.
How do fees change the probability I need to win?
Add the fee to your cost per contract. On Kalshi, a taker buying at 40¢ pays about 0.0168 per contract in fees, so the break-even probability rises from 40% to 41.68%.