Clinical Trial Outcome Predictor

Thesis: Phase III success rates vary predictably by indication, trial design, sample size, primary endpoint, and sponsor history. Many retail biotech traders treat all trials as coin-flips; a proper prior does not. Data source: ClinicalTrials.gov v2 API + FDA.gov — both free, structured, commercially usable (US government works). Methodology: Bayesian success-rate model conditioned on therapeutic area, trial-design features, endpoint type, and sponsor track record. Calibrated on historical approvals/failures from 2010 onward. Academic evidence: DiMasi et al. (2003, 2016), *Clinical Approval Success Rates for Investigational Drugs.* Well-documented base rates by phase and indication. Biotech-focused hedge funds use variants of this approach. Status: Pending. Model not yet trained.