A security that pays out if the outcome of a random variable falls within a specified interval. Enables prediction markets for continuous variables by allowing traders to express arbitrary endpoint precision.
Cluster: Mechanism Design
A security that pays out if the outcome of a random variable falls within a specified interval. Enables prediction markets for continuous variables by allowing traders to express arbitrary endpoint precision.
Referenced in 2 articles
A thread by Igor (@justigor) arguing that the primitive can be complex while the trading surface stays simple: complexity belongs in the mechanism, and the thing the trader actually touches should let them express a view for a price they like. Walks through a World Cup campaign built on fantasy points as the underlying, where one market supported four different interaction types — dynamic over/unders, range tickets, precision payoffs, and head-to-head duels like Messi vs Kane — and where third parties (Polysights, Trepa) built their own surfaces on the same core. Extends the logic to crypto, where range views on ETH are options-adjacent beliefs with the cognitive tax removed, and to new underlyings like GPU-hour rates, arguing that anything that resolves to a number can trade. Lesson: don't simplify the thing being traded — simplify how people express what they think about it.
Designs the first computationally efficient prediction market for continuous outcome distributions using interval securities. Achieves logarithmic-time operations by decomposing the LMSR into a balanced binary tree, and extends to parallel submarkets at multiple resolutions with constant bounded loss.