The Biotech Valuation Puzzle: Why Pipeline Labels Predict Less than Pipeline Activity

<div> Standard risk-adjusted net present value frameworks predict that categorical pipeline characteristics — clinical development stage, therapeutic indication, and drug modality — should meaningfully differentiate biotech company valuations. Stage-specific success probabilities, indication-specific market sizes, and modality-specific development costs all enter valuation models as first-order parameters. I test this prediction using 4,117 biotech financing transactions spanning 2008 to 2026, sourced from SEC EDGAR filings and cross-referenced with ClinicalTrials.gov for independent clinical taxonomy assignment. The categorical labels collectively explain less than 5% of pre-money valuation variance beyond deal structure controls. However, continuous measures of the same underlying pipeline construct — the number of active clinical trials, the diversity of development phases represented, and maximum trial enrollment — explain substantially more, outperforming the categorical labels by a factor of 1.4 to 1.6 in partial R². On a matched sample of 1,630 transactions, pipeline activity measures achieve a cross-validated R² of 0.153, compared with 0.115 for the categorical labels. When both are included, the combined model reaches 0.172; activity contributes an incremental 0.057 beyond labels, while labels contribute only 0.019 beyond activity. The asymmetry is the puzzle: the pipeline predicts valuations, but its continuous dimensions carry substantially more information than its categorical classifications. A comprehensive screening of 65 candidate variables contextualizes these findings, revealing that calendar metadata — the day of the week a deal closes — achieves comparable explanatory power to therapeutic indication, a variable with direct theoretical relevance to drug economics. Twenty-five placebo variables confirm the methodology's specificity. These findings suggest that supplementing the comparable transaction framework's categorical filters with continuous pipeline activity measures, available at no cost from public clinical trial registries, could improve valuation precision. </div>