How prediction market prices work
A contract trading at 62 cents implies roughly a 62% chance. Here is where that number comes from and what it leaves out.
Kivorent News desk, , 2 min read
What to know
- A yes contract pays $1 if the event happens and nothing if it does not.
- The price in cents is roughly the market's estimate of the probability.
- Fees, the gap between buy and sell prices, and thin trading all pull the price away from a clean probability.
Prediction markets let people trade on whether something will happen: an election result, an interest rate decision, a game. The prices get quoted as probabilities, and that reading is close to right for a specific reason.
The contract
Most prediction markets use a simple contract. A yes share pays $1 if the event happens and $0 if it does not. A no share pays the opposite.
If a yes share costs 62 cents, you risk 62 cents to make 38. That bet is worth taking only if you think the event is more than 62% likely. Someone who thinks it is less likely will sell to you or buy no. Where those views balance is the price, which is why it reads as a probability.
The arithmetic
| You buy | Price | If yes | If no |
|---|---|---|---|
| 100 yes shares | $62 | Receive $100, a gain of $38 | Receive $0, a loss of $62 |
| 100 no shares | $38 | Receive $0, a loss of $38 | Receive $100, a gain of $62 |
Yes and no prices add up to about $1. When they do not, the difference is trading costs.
You do not have to wait for the result
Shares can be sold at any time before the event is settled. If you buy at 62 cents and news moves the price to 75, you can sell and keep 13 cents a share. Many traders never hold to the end. They trade the changes.
Why the price is not exactly the probability
- The spread. There is a price to buy and a lower price to sell. A market might be 61 to bid and 63 to buy. The probability sits somewhere in that gap.
- Fees. A fee on trades or winnings means a 62-cent price can reflect a slightly different belief.
- Thin markets. With few traders, one large order can move the price a long way.
- Time. Money tied up until next year earns nothing meanwhile, so far-off favorites tend to trade a little below their true odds.
- Longshots. Contracts priced at a few cents are often priced too high, because people like cheap bets with big payoffs.
How accurate are they?
The test is calibration: of all the events priced near 70%, do about 70% happen? Large, heavily traded markets have generally done well on that test. Small markets do worse.
A 62% favorite loses 38 times in 100. One wrong call does not show a market was broken, and one right call does not show it was wise.
Before you trade
- Read the settlement rules. Each contract has a written definition of what counts as yes. Disputes come from the fine print.
- Check who regulates the platform and whether you are allowed to use it where you live.
- Count the costs. Fees and the spread come out of every trade.
- Size your positions. A 90-cent contract still pays zero one time in ten. See the math of risking 1%.
The buy and sell prices behind every quote live in an order book. We explain it in how an order book works.
This article is general information. It is not investment advice, and trading involves the risk of losing money.
Spotted an error? Tell us and we will fix it.