Open datasets released alongside ForesightFlow research. All datasets are published under CC-BY-4.0 unless otherwise noted.
Full reasoning traces for five LLM multi-agent coordination configurations evaluated on 100 resolved Polymarket binary prediction markets. Includes the Murphy decomposition leaderboard from the companion paper.
A curated validation set mapping eight publicly documented episodes of suspected informed trading on Polymarket to concrete on-chain market identifiers. Cases span 2023–2026 and cover 24 individual markets across military and geopolitical actions, corporate proprietary disclosures, and regulatory decisions.
Fill-side behavioral clusters, feature tiers, and per-market microstructure signatures from 13.36M OrderFilled events on Polymarket CTFExchange (43,116 markets, 77,203 addresses, 2026-04-21 to 2026-04-27). k-means k=5 archetypes, 6 feature tiers, Spearman bilateral analysis with BH-FDR correction. Bundle 3 of the PMXT family; companion to Paper 4 of the Event-Linked Perpetuals research programme.
Per-(engine/mechanic, leverage, class) summary statistics from two counterfactual simulation experiments on 13,000+ resolved Polymarket binary-event markets. E2 compares three margin-engine variants on survivability and drawdown; E3 compares four resolution-zone mechanics (R0–R3) on final-hour liquidation rates, bad-debt frequency, and average PnL. Bundle 2 of the PMXT family; companion to Paper 1 of the Event-Linked Perpetuals research programme.
Per-market stylized-fact measurements (SF1–SF9) for 13,314 resolved Polymarket binary-event markets over 2026-04-21 to 2026-04-27. Empirical foundation for Paper 1 of the Event-Linked Perpetuals research programme.
Deadline Information Leakage Scores (ILS^dl) for 88 Polymarket markets across military/geopolitical, regulatory, and corporate categories. Includes LLM-recovered event dates (T_event), bootstrap confidence intervals, and the full 2,375-market attrition chain. Snapshot: 2020–2026.
Per-category maximum-likelihood exponential hazard fits for the time-to-event distribution on Polymarket deadline-resolved contracts. Headline: military_geopolitics λ=0.241 events/day, 95% CI [0.143, 0.365], half-life 2.9 days (KS p=0.425, n=18, adequate). Baseline survival function for the deadline-ILS framework.
Population-scale Information Leakage Score (ILS) computations for 4,801 resolved Polymarket markets. Anchored at t_resolve − 24h proxy for 99.9% of records (4,796 markets); 5 markets carry a genuinely event-anchored timestamp. Includes multi-window ILS variants (30min, 2h, 6h, 24h, 7d), top-wallet concentration (HHI), and scope-condition flags. For event-anchored analysis on a smaller, deeper sample, use polymarket-deadline-ils-v3.
A three-class classification of 911,237 Polymarket markets by resolution mechanism (deadline_resolved, event_resolved, unclassifiable). Snapshot cutoff 2026-04-27. Released as a methodological resource for ILS-based informed-trading research.
Public-event and news-arrival timestamps for 2,052 resolved Polymarket markets, recovered across three methodological tiers: UMA Oracle proposer evidence (Tier 1, n=12, confidence 0.95), GDELT proxy (Tier 2, n=1,993, confidence 0.60), and LLM-assisted multi-source verification (Tier 3, n=47, confidence 0.80–0.90). Snapshot: 2022-12–2026-04.