The Anatomy of a Decentralized Prediction Market: Microstructure Evidence from the Polymarket Order Book

Paper: arXiv 2604.24366

Authors: Philipp D. Dubach

Abstract

We study the microstructure of Polymarket, the largest on-chain prediction market, using a continuous tick-level archive of the public order-book feed (30 billion events over 52 days) joined to the authoritative on-chain trade record. On a pre-registered stratified panel of 600 markets we report eight stylized facts: a longshot spread premium; a depth profile closer to uniform than to top-of-book; a null block-clock alignment effect; broad maker-wallet diversity with a concentrated tail; category-conditional effective-spread differences; a sub-50 ms median archive-ingestion delay with a multi-second tail; a self-counterparty wash share with median 1% and a 22% upper tail (well below Cong et al. 2023’s 25-70% for unregulated crypto venues – a sanity bound, not an apples-to-apples reference); and a cross-sectional depth profile explained by market duration, price level, and volume, with no residual time-to-close effect. The paper also contributes a measurement result: trade direction inferred from Polymarket’s public order-book feed agrees with on-chain ground truth on only ~59% of buckets (panel mean 0.615, 95% CI [0.58, 0.65]), well below the ~80% Lee-Ready accuracy on Nasdaq. The effective half-spread changes sign between feed- and on-chain trade directions on 67%/50% of markets across two 7-day windows; Kyle’s lambda on 60%/43%. Microstructure work on Polymarket therefore needs to source trade direction from on-chain OrderFilled events; we release a replication package that performs the join.

Complexity vs Empirical Score

• Math Complexity: 4.5/10
• Empirical Rigor: 9.5/10
• Quadrant: Street Traders — practical and empirical, lighter on theory

Why this score: The paper relies on an immense dataset (30 billion events) and a rigorous on-chain/off-chain data join to evaluate market microstructure, focusing on empirical findings over abstract theoretical derivations. While it employs standard econometric measures like Kyle’s lambda and spread decompositions, the core contribution is a robust, data-first forensic analysis of micro-structure reality.

Research Flowchart

  flowchart TD
    A[<b>Goal:</b> Analyze Polymarket Microstructure] --> B[<b>Data:</b> Tick-level Order-book Feed + On-chain Trade Records]
    B --> C[<b>Process:</b> Join Feed to On-chain 'OrderFilled' Events]
    C --> D[<b>Analysis:</b> Stratified Panel of 600 Markets <br/>(30B events over 52 days)]
    D --> E{<b>Findings</b>}
    E --> F[<b>Stylized Facts:</b> Depth profile, longshot premium, wash trading <1%]
    E --> G[<b>Measurement:</b> Fee-based trade direction accuracy is low (~60%)]
    G --> H[<b>Outcome:</b> Recommended On-chain sourcing & Replication Package]

Cited References (18)

1. SoK: Market Microstructure for Decentralized Prediction Markets (DePMs) — → 2510.15612
2. Crypto Wash Trading (2019) — link
3. High Frequency Trading and Price Discovery (2014) — link
4. Do prediction markets produce well-calibrated probability forecasts? (2013) — link
5. Market Liquidity: Theory, Evidence, and Policy (2013) — link
6. LOGARITHMIC MARKETS CORING RULES FOR MODULAR COMBINATORIAL INFORMATION AGGREGATION (2012) — link
7. Empirical Market Microstructure: The Institutions, Economics, and Econometrics of Securities Trading (2007)
8. Interpreting the Predictions of Prediction Markets (2006) — link

1 of 18 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2518 linked papers.