Comment Text:
COMMENT ON ADVANCE NOTICE OF PROPOSED RULEMAKING ON PREDICTION MARKETS
Submitted by Gianlorenzo Ranieri, Founder, Convexly
The complete comment with citations is attached as cftc-comment-ranieri.txt. This inline summary covers identity, the structural opportunity, the recommendation, and the summary; the full text expands these sections and includes academic and industry citations.
1. IDENTITY AND DISCLOSURE
I am the founder of Convexly Research, a prediction-market analytics company that publishes open methodology for measuring trader skill, structural inference on market prices, and behavioral detection of market abuse. Convexly maintains six pre-registered methodology papers on AsPredicted (#287368, #287436, #287442, #287714, #287983, and #288046), reproducible from public data bundles at https://www.convexly.app/research. Convexly is bootstrapped, has no external investors, and is not affiliated with any prediction-market platform or government entity.
I am separately an active-duty United States Army intelligence officer. This comment reflects my work as Convexly's founder and is not affiliated with, nor representative of, the views of the U.S. Department of Defense or any government agency.
Convexly is a for-profit venture. I am developing a commercial product (the Coherent Markets Engine, or CME) that implements the methodology I describe below. The recommendation is not unique to Convexly; it would benefit any vendor or platform that meets the methodology bar.
2. THE STRUCTURAL OPPORTUNITY
Prediction markets present a transparency advantage no other financial market provides: outcome contracts are settled against verifiable real-world events whose prices, if priced consistently, must obey known probability laws (additivity, inclusion-exclusion, conditional Bayes consistency). Yet on every major U.S.-accessible prediction market, individual contracts are priced independently and cross-market constraints are not enforced. The result is a measurable, real-time stream of structural inconsistencies. Saguillo, Ghafouri, Kiffer, and Suarez-Tangil (2025), at IMDEA Networks, analyzed 86 million Polymarket bid records over April 2024 to April 2025 and estimated approximately 39.6 million dollars of realized arbitrage from probability-axiom violations during that window (arXiv:2508.03474, AFT 2025).
3. RECOMMENDED TRANSPARENCY STANDARD
I recommend the Commission consider, as part of any framework it ultimately adopts:
A. COHERENCE-FEASIBILITY METADATA. Encourage prediction-market platforms operating under Commission supervision to publish, alongside last-traded prices on related contracts, a coherence-feasibility tension indicator. For each defined cluster of related markets ("coherent cluster"), the platform would publish: (1) the cluster definition; (2) the current coherence-feasibility projection (the closest constraint-satisfying probability vector); (3) a tension indicator (in L2 or KL-divergence units); and (4) a statistical significance test indicating whether the observed tension is consistent with bid-ask noise or constitutes a meaningful coherence violation.
B. PUBLIC AUDITABILITY. The coherence projection methodology, including constraint set definitions and any approximations or solver tolerances, should be publicly documented in sufficient detail to allow third-party reproduction.
C. STATISTICAL RIGOR IN DETECTION THRESHOLDS. Any compliance threshold based on coherence-feasibility tension should be paired with: (1) a pre-registered methodology for the projection; (2) a formal statistical test with explicit type-I error rate; (3) a multiple-testing correction (Benjamini-Hochberg FDR is standard) if many clusters are tested simultaneously; (4) documented bounds on solver tolerance and numerical stability.
The Pennock-Lahaie-Kroer LCMM lineage (Csiszár 1975 on I-projection geometry; Kroer, Dudík, Lahaie, and Pennock 2016 on arbitrage-free combinatorial market making) provides the canonical theoretical foundation. Convexly and other commercial vendors have implemented working variants now in active deployment on public market data.
4. SUMMARY
The Commission has a unique opportunity, given current oversight of Polymarket and Kalshi, to establish coherence-feasibility transparency as a foundational standard. The mathematics is well-established, operational implementation is tractable, and the commercial vendor ecosystem is actively developing the necessary tooling. Convexly is committed to publishing its CME methodology under open documentation and to making the underlying mathematical framework available for third-party verification. We welcome the opportunity to engage with Commission staff on technical implementation specifics.
Respectfully submitted,
Gianlorenzo Ranieri
Founder, Convexly Research
[email protected]
https://www.convexly.app
