Information Leakage at Population Scale: An Evaluation of the Polymarket Insider-Relevant Subpopulation

We carry the deadline-resolved Information Leakage Score (ILS^dl) framework from a single-case proof of concept to a population-scale evaluation across 12,708 Polymarket markets in three insider-relevant categories spanning October 2020 through April 2026. We frame the paper as a scope-discovery study: scaling reveals that the framework's effective domain is materially narrower than initial framing suggested, and the principal obstacle is not score computation but resolution semantics.

We report four findings. First, only 88 of 12,708 candidate markets (0.7%) yield computable ILS^dl values, and only 1 of 32 markets in the ForesightFlow Insider Cases (FFIC) inventory is in scope; 14 of 32 FFIC markets are flagged unclassifiable due to genuine resolution-criterion ambiguity ("strike", "custody", "out as Supreme Leader"). Second, only 12 of the 88 computed markets (13.6%) satisfy anchor-sensitivity, and an independent-second-pass T_event validation reaches 57.8% exact-date agreement, below the ≥ 90% ex-ante criterion. Third, raw ILS^dl medians are negative across all six (sub-bucket × period) cells, but a hazard-decay baseline correction yields a heterogeneous result: regulatory_formal post-2024 shifts to near-zero (−0.21 → −0.02), while regulatory_announcement post-2024 retains a confidence interval entirely below zero (−0.84, [−1.45, −0.24]). Fourth, the constant-hazard exponential is rejected on the pooled post-2024 cell in favor of Weibull, but a per-subcategory check confirms that the pooled-cell preference reflects category mixture rather than within-cell duration dependence.

The substantive implication is that detection of informed flow on the insider-relevant subpopulation requires methodological refinement on the resolution-typology axis and on the score-baseline axis, not only on the score-computation axis where prior work concentrated. We release dataset polymarket-deadline-ils-v3 alongside this paper at https://github.com/ForesightFlow/datasets (CC-BY-4.0).