ForesightFlow: An Information Leakage Score Framework for Prediction Markets

Decentralized prediction markets such as Polymarket aggregate dispersed beliefs into continuously updated price signals, but their on-chain transparency and pseudonymous participation also create an unusually fertile environment for informed trading on material non-public information. Recent empirical work has documented hundreds of millions of dollars in anomalous profits on Polymarket between 2024 and 2026; existing detection approaches are almost exclusively post-hoc and offer no actionable signal during the window when informed flow is moving prices.

We propose ForesightFlow, an information-theoretic framework for quantifying informed flow on prediction markets. We introduce the Information Leakage Score (ILS), a label generator that quantifies how much of a market's terminal information move was priced in before the corresponding public news event, and show that ILS admits a clean interpretation in terms of the Murphy decomposition of the Brier score. We then specify the score's resolution-typology and operational scope conditions: ILS is interpretable only for event-resolved markets with substantive uncertainty at T_open, with article-derived rather than proxy-anchored news timestamps, and under an anchor-sensitivity robustness check. A pilot empirical study supports each scope condition: a resolution-anchored proxy for T_news does not separate event-resolved markets from a matched control population; high ILS on high-consensus markets reflects a formula edge effect; and the proxy-based ILS distribution is not robust to anchor choice. A proof-of-concept article-derived T_news recovery on the Epstein-files Barak market suggests that proxy quality is not the binding constraint and that wallet-level features rather than ILS alone are required to identify informed flow.

An audit of the publicly documented Polymarket insider-trading record reveals a structural finding: documented cases are systematically deadline-resolved ("Will event X occur by date Y?"), falling outside the original ILS scope. We accordingly extend the score to a deadline-ILS variant anchored to the underlying event timestamp, with a per-category exponential-hazard baseline for the time-to-event distribution. The extension closes the gap between the original methodology and the population in which informed trading has been empirically documented. An end-to-end empirical evaluation of the deadline-ILS extension on the 2026 U.S.–Iran conflict cluster, including hazard-rate fits by category, a single-case ILS^dl computation, cross-market wallet analysis, and the methodological refinements that the evaluation surfaces, is reported in a companion paper. We release the ForesightFlow Insider Cases (FFIC) inventory, the full resolution-typology classification of the 911,237-market corpus, and all system code openly at https://github.com/ForesightFlow.