Catalyst Family

Catalyst families are categories that group individual catalyst types by their structural similarity — both in terms of the underlying market mechanism and the typical price behavior they produce. Rather than treating every catalyst type as unique, the TMS scoring engine organizes catalysts into families and applies family-level historical pattern libraries when scoring new events. This allows the engine to leverage broader pattern data even when a specific catalyst sub-type has limited historical examples.

Primary Catalyst Families

Regulatory family. FDA decisions, EMA approvals, DEA scheduling decisions, and other government regulatory actions. Common characteristics: binary outcomes (approve/reject), pre-announced PDUFA dates, confined primarily to pharmaceutical and biotech sectors, often high-magnitude moves (20–100%+).

Earnings family. Quarterly EPS results, revenue reports, full-year guidance, pre-announcements, and earnings warnings. Common characteristics: scheduled events (earnings calendar), magnitude determined by surprise size relative to consensus, applies to all public companies, post-earnings drift phenomenon.

Corporate action family. M&A announcements, going-private transactions, spin-offs, divestitures, and special dividends. Common characteristics: company-initiated, fundamentally changes ownership structure or asset composition, target stocks show anchored repricing behavior.

Insider/institutional family. Insider buying clusters, 13D activist filings, significant institutional position disclosures. Common characteristics: informed buyer signal, typically mid-duration catalyst effect (days to weeks), strongest signal when multiple insiders act simultaneously.

Sentiment family. Analyst upgrades/downgrades, short squeeze triggers, social media momentum, and unusual options activity without primary catalyst. Common characteristics: derived from secondary signals rather than primary corporate announcements, shorter reaction window, lower reliability than primary catalyst families.

How Catalyst Families Improve TMS Accuracy

By applying family-level historical distributions when scoring individual catalysts, the TMS engine achieves better calibration on rare or novel catalyst sub-types. A new type of regulatory decision (say, a DEA scheduling action on a novel drug class with limited precedent) is scored using the broader regulatory family patterns — FDA decisions, DEA precedents, and comparable binary regulatory events — rather than attempting to match against a thin history of identical events. This family-based approach is one of the key architectural decisions that reduces false positives in the scoring engine.