Insider Buying — Clusters & Single-Insider Conviction Buys
Open-market insider buying: multi-insider clusters (≥2 distinct insiders in a rolling window) AND high-conviction single-insider buys, filtered to buy-dominant activity (Cohen/Malloy/Pomorski, 2012). C-suite participation and a catalyst-impact score are attached to each setup.
How this signal works
What you're looking at. Two or more insiders at the same company bought shares on the open market within a rolling window. We filtered the cluster to real open-market purchases (excluding tax withholding and routine option exercises), counted CEO and CFO participation, and attached a 1-10 catalyst-impact score.
How it has behaved. Clustered open-market buys have outperformed the market over 6 months (Cohen/Malloy/Pomorski, 2012), CEO/CFO buys more than director buys — a multi-week effect, not an intraday one. A setup is priced at the next session open after the cluster report, or at the live price when it lands during the session, and tracked from T+1 to T+60.
Where it has been weaker.
- All-director clusters with no C-suite buyer.
- Clusters totalling under $200k.
- Clusters in the same week as a sell-side downgrade or guide-down.
- Impact score under 4 — the buying has no clear catalyst behind it.
Further reading.
- Cohen, Malloy & Pomorski (2012) — separates "routine" vs "opportunistic" insider trades. The opportunistic ones predict returns; the routine ones don't.
- Lakonishok & Lee (2001) — insider buying predicts returns, selling doesn't. Small firms show the strongest effect.
- Jeng, Metrick & Zeckhauser (2003) — insider purchases earn ~11% annual abnormal return over 6 months. Sales earn nothing.
- Seyhun (1986) — the foundational paper. Top executives outperform other insiders.
- Marin & Olivier (2008) — the absence of insider buying before a crash is itself a signal. Pay attention to tickers that stopped.
No insider clusters detected today.