Microstructure & Algorithmic Arbitrage on Prediction Markets
Published on April 20, 2026 • By Léo Lombardini
5 min read
Share:
Research on algorithmic participation and market making on Polymarket. Theoretical framework and empirical application.
Executive Summary
This document develops a rigorous quantitative framework for algorithmic participation in decentralized prediction markets (DPM), using Polymarket as the primary empirical environment. We analyze order-book dynamics and statistical arbitrage opportunities.
1. Market Context
Decentralized prediction markets offer a unique structure where information is aggregated asynchronously. The study focuses on the efficiency of these markets in the face of geopolitical and macroeconomic events.
2. Market Making Strategies
We model an inventory-adjusted market-making strategy, aimed at providing liquidity while minimizing directional risk (Delta Neutral).
Adapted Avellaneda-Stoikov model: Adjustment of spreads as a function of the contract’s implied volatility.
Risk Management: Strict limits on position size and hedging via other instruments where available.
3. Arbitrage and Microstructure
Analysis of latency and the impact of gas fees (Polygon) on the execution of high-frequency strategies. The study demonstrates the presence of exploitable correlations between prediction markets and classic derivatives markets.
Resources
The full paper is available for download below for an in-depth study of the mathematical formulas and backtest results.
Passionate about market analysis and statistical modeling, Léo oversees the strategic allocation of the model portfolio and the development of Horacle Capital's quantitative frameworks, as well as writing weekly articles.
