Starling Quantitative Logo
STARLING QUANTITATIVE QUANTITATIVE RESEARCH NOTES
← RETURN TO DESK
METHODOLOGY MONOGRAPHS // DESK PUBLICATIONS

QUANTITATIVE
Research Notes.

Formal mathematical derivations and production risk specifications from Starling Quantitative—covering sharp benchmark probability extraction, favorite-longshot microstructure, constrained Fractional Kelly sizing, and out-of-sample Brier decomposition.

REGULATORY MANDATE: Starling Quantitative is a proprietary internal research and capital allocation desk of Biwada. Research notes are published strictly for mathematical and engineering discourse; we do not solicit outside retail capital, manage third-party funds, or offer investment advice.
RESEARCH NOTE // RN-01 MARKET MICROSTRUCTURE
SUM(PI_I^K) = 1 // POWER & SHIN DASHED: MULTIPLICATIVE (IGNORES FLB) RAW IMPLIED PROBABILITY →

Extracting True Implied Probabilities from Vigged Markets: Additive, Multiplicative, Power, and Shin Devigging Compared

Why raw bookmaker and CLOB quotes embed asymmetric favorite-longshot bias, how Newton-Raphson root-finding solves for k > 1 in Power devigging, how Shin’s microstructure parameter z estimates informed order flow, and how asymmetric tick rounding preserves net spread across Kalshi and Polymarket.

OCTOBER 2026 • 16 MIN READ READ PAPER →
RESEARCH NOTE // RN-02 RISK & CALIBRATION
C = 0.25 QUARTER-KELLY LOCK f* (NOISY PEAK) f > 2f* DECAY

Fractional Kelly Sizing Under Parameter Uncertainty & Out-of-Sample Brier Decomposition

Why plugging noisy edge estimates into full Kelly systematically overbets, analytical drawdown tail bounds for quarter-Kelly (c = 0.25), portfolio covariance constraints, and the Murphy (1973) decomposition of Brier Score into reliability, resolution, and uncertainty.

OCTOBER 2026 • 18 MIN READ READ PAPER →