Model Documentation

Methodology

L1 logistic regression with a historical phase × indication base-rate prior, post-hoc isotonic calibration, and propensity-matched causal adjustments for FDA designation effects.

Data

Live, point-in-time-safe collection at scoring time — no look-ahead from information unavailable on the catalyst date.

SourceData
ClinicalTrials.govTrial phases, enrollment pace, primary endpoints, p-values
SEC EDGARXBRL financials, Form 4 insider trades, 8-K filings
openFDADesignations, AdComm votes, CRL history, prior approvals
CTO Dataset14,700 Phase 3 trial outcome labels (Gao et al. 2024)
PubMedPublished Phase 3 trial results for endpoint verification

Model

Each drug starts from a historical phase × indication base rate (BIO/IQVIA). Factors across four signal categories shift the probability up or down in log-odds space; an isotonic map fit on a rolling 2018+ window corrects systematic overconfidence at the high end.

Category# Factors
Clinical12
Regulatory11
Financial3
Non-linear interactions (SHAP)3

Validation

Trained on ~1,000 historical public NDA/BLA events through 2020, evaluated out-of-time on ~500 events from 2021-present.

MetricValue
AUC-ROC (out-of-time, n=512)0.841
    Small-cap subset (mega-pharma excluded)0.81
Brier score0.117
Expected Calibration Error2.1%
Accuracy (optimal threshold)83%

The headline AUC is buoyed by mega-pharma events (PFE, JNJ, NVS) that are trivially predictable from sponsor track record. The small-and-mid-cap subset — where the prediction is hard and the use-case lives — is 0.81.

Benchmarks

All models trained and evaluated on the same out-of-time public test set.

ModelAUC-ROC
ApprovalAlpha0.841
Lo et al. 2019 method (reproduced)0.807
Phase × indication base rate only0.667

Designations Are Not a Free Lift

A propensity-matched analysis on the training data: after controlling for base rate, trial results, sponsor history, endpoint type, and mechanism of action, Priority Review is the only FDA designation with a positive causal effect. Breakthrough Therapy, Fast Track, and Orphan Drug are markers of clinical difficulty — FDA grants them to drugs whose path is inherently harder. The model uses the causal-adjusted coefficients, not the naive correlations.

Dossier Outputs

Beyond the headline probability, every scoring emits five structured outputs. Deterministic post-hoc transforms of the model output — no LLMs, no extra training.

OutputWhat it is
Top driversPer-factor pp impact on this specific prediction.
ComparablesFive most-similar historical events (cosine similarity weighted by L1 coefficient magnitude) with realised outcomes.
SensitivityWhat the prediction becomes if any one factor flips toward the opposite class's typical value.
SubscoresClinical / Regulatory / Sponsor decomposition. Describes which feature bucket pulls probability down, not which CRL category will occur.
Probability historySparkline of prior scorings for the same drug as new data lands (8-Ks, AdComm decisions, dilution events).

Disclosures

References

Built by Sean Koth, finance student at Fordham University's Gabelli School of Business.

LinkedIn ↗