On prediction markets the key is to judge how predictable the event itself is. There are four conditions that determine this.
-
Understanding the drivers. If the structure of the event is clear, you can assign a probability. If the factors are fuzzy, forecasting turns into guessing.
-
Availability of data. The more data and repeatable patterns, the better. If data is scarce, the market becomes a psychological game.
-
Similarity of the future to the past. Typical events are predictable. Unique events are almost not.
-
Impact of the forecast on the outcome. Some markets are self-influencing: expectations change reality.
For Polymarket this matters a lot: sometimes the market predicts the event, and sometimes the event is shaped by the market.
— S. Praevis