Filings, Insiders & OwnershipExtended setExperimentalNew

Form 144 Filer Cluster

Updated dailyData needs: lowlong only
paper
2012
Source
Cohen, L., Malloy, C., Pomorski, L. (2012). Decoding Inside Information. Journal of Finance 67(3): 1009-1043 (paper_url DOI). Signal construction (distinct Form 144 filer count within a 30-day window) additionally informed by Henderson, Jagolinzer, Muller (2015), working paper on strategic Form 144 / Rule 10b5-1 sale disclosure.
Read the paper →

In plain terms

When 3 or more insiders file Form 144 (intent-to-sell notices) within a single month, the cited papers predicted the stock underperforms for 2-6 months. Our testing found the opposite on the current universe: these clusters happen at high-momentum names that keep rising, so the strategy buys on the cluster instead of shorting it.

How it works

Papers posit that when MULTIPLE distinct insiders file Form 144 (intent-to-sell notice) within a short window (3+ in 30 days), the clustering reflects coordinated information about over-valuation, predicting underperformance over 60-180 days. INTENTIONAL DEVIATION (template v4, 2026-06-08): on this universe the short thesis fails; pooled IS/OOS tests on two independent liquid baskets show the SHORT pool robustly negative (basket A IS -1.43 / OOS -2.68; basket B IS -1.90 / OOS -1.64), so the shipped implementation trades the INVERSE: clustered Rule 10b5-1 sell notices occur at high-momentum growth names after a run-up, and those names continue (momentum continuation), so the code goes LONG T+1 on the cluster. Paper citation is retained for the signal construction (distinct-filer count), not the sign.

No live results for this strategy yet. Charts appear once it has earned a top spot on at least one stock, either on its own or as part of a blend of several strategies.
Loading substrate evidence…

Data dependencies

  • Daily prices

    Adjusted-close OHLCV for every US-listed ticker; primary price feed.

  • SEC form 144 filings

    SEC Form 144 planned-sale notices, parsed from EDGAR structured filings.

Expected edge

Reported return
-3-5% over 90d (papers' SHORT direction; shipped implementation intentionally trades the inverse LONG)
Tested over
T+1 to T+180d

Papers' original claim: -3-5% over 90d SHORT on >=3 distinct filers within 30d (2003-2012 sample). Shipped (flipped LONG): pooled IS +1.43..+1.90 / OOS +1.64..+2.68 Sharpe across two independent baskets; mean forward-60d after fire +10.9% vs +8.1% same-names all-days baseline.

Related families

Explore Form 144 Filer Cluster on alphactor.ai

See which tickers this family is currently firing on, with live signals and rankings.

For informational and educational purposes only. Not financial advice. Learn more