What a prediction market actually is
Event contracts settle at 100 or 0. That single fact explains the pricing, the strategy, and most of the mistakes newcomers make.
A prediction market is a venue where you buy and sell contracts tied to whether a specific thing happens. Will this index close above a level on Friday. Will this candidate win. Will the central bank hold rates in September.
The mechanic underneath is simpler than it sounds, and everything else follows from it.
Settlement is binary
Each contract resolves to one of two values: 100 if the event happens, 0 if it does not. There is no partial credit and no gradual decay toward an answer. On resolution day the contract is worth everything or nothing.
That is the whole design. A contract trading at 63 is one the market currently values at roughly a 63 percent chance of resolving YES.
Price is a probability, not a price
This is the part that takes a week to internalise if you come from equities or futures.
When a stock trades at 63 dollars, that number is an opinion about value. When an event contract trades at 63, that number is an opinion about likelihood. The contract is not cheap or expensive in the ordinary sense — it is correctly or incorrectly priced relative to how often that outcome would occur if the situation ran a thousand times.
So the question you are answering is never “is this going up”. It is:
Does the market’s implied probability match mine, and by enough to be worth taking the other side?
If a contract sits at 63 and you have done the work and believe the true figure is nearer 75, that gap is the entire trade. If you think it is 64, there is no trade — you agree with the market and you would be paying costs for the privilege of agreeing.
Where the edge actually comes from
Three places, in rough order of how reliably they pay:
Information the market has not absorbed yet. A filing, a poll, a roster change, a scheduling note. Prediction markets react quickly on headline events and slowly on obscure ones. The obscure ones are where an attentive person competes.
Correctly reading a resolution rule. Contracts resolve against specific written criteria, and a surprising number of disputes come from people trading the headline rather than the rule. “Will the bill pass” and “will the bill pass by 30 September” are different contracts with different fair values. Reading the resolution text is not administrative overhead; it is analysis.
Structural mispricing at the tails. Contracts trading very near 0 or very near 100 are systematically difficult to price, partly because the potential gain on the cheap side attracts buyers who are not doing probability at all.
The mistake almost everyone makes first
Treating a strongly held opinion as a large position.
Confidence and probability are different quantities. You can be completely convinced an outcome is likely and still be describing a 70 percent event — and 70 percent events fail three times in ten, in a row sometimes. Sizing as though conviction equals certainty is the fastest way to be right about the world and wrong about your account.
The practical version: decide your probability before you look at the price. Write it down. Then compare. If you find your estimate drifting toward whatever the screen says, you are not forecasting, you are reading.
Why this is a trading skill
People sometimes assume this is a softer discipline than trading an order book. It is not. You are pricing an asset, managing exposure, and being scored by an unforgiving mechanism at settlement. What differs is the input: instead of reading order flow, you are reading the world, and then converting a view into a number you are willing to defend with capital.
That conversion — from opinion to defensible probability to correctly sized position — is the entire craft.