Prediction-market contracts represent possible outcomes such as an election result, weather threshold or price event. Trading prices can be interpreted as the market's current view of each outcome, subject to the contract rules and available liquidity.
Automated strategies can compare their own estimated probabilities with market prices and place orders when they see a gap. Results still depend on fees, fills, market depth, resolution rules and whether the model's probabilities are well calibrated.
ELI5
A prediction market is a market where people trade contracts about what will happen in the future. A contract pays according to a clearly defined outcome.
For example, a weather contract might pay if a city reaches a stated temperature. An AI model can estimate the chance and trade when the price looks wrong, but profit also depends on getting an order filled and on the market's settlement rules.
How do prediction-market prices relate to probability?
Under common contract designs, a price can approximate the market's implied probability, but fees, liquidity and trader behavior can distort it.
What risks affect prediction-market strategies?
Model error, thin liquidity, missed fills, contract wording, settlement decisions, fees and changing information can all affect results.
