1.0 is a normal window for that sector and dataset. 2.4 means filings arrived at 2.4x the usual rate. Cells with too thin a baseline are left blank rather than faked. Counted by the date each filing became public, so the time machine never shows what nobody could have seen.
WaveMan joins seven official US government sources that no other free site holds together: SEC Form 4 insider transactions, Form 13F institutional holdings, Schedule 13D/G 5%+ stakes, Form 144 planned sales filed before the trade, XBRL fundamentals, STOCK Act congressional disclosures, CFPB consumer complaints and NRC nuclear reactor output. Every figure carries the date it was filed, so you can ask what was known on a past date rather than what we know now. Free, updated daily, and the methodology is published, including a backtest showing the naive insider-buying signal does not beat the market.
Start here: insider buying now · Congress trades · activist stakes · planned sales · browse all companies · does it work?
What happened after each person's own past purchases, 90 days, versus the Russell 2000, entering only after the filing was public. Adjusted is shrunk toward the 42.7% base rate in proportion to how thin the evidence is, so a hot streak of five cannot outrank a record of ninety. Cos. matters: 76% of records cover a single company, which makes them that company's trajectory more than portable skill. A historical association, not a prediction.
Members must report trades within 45 days, so this is a disclosure record rather than a live signal, the lag column shows how long each trade stayed private. Senate filings are excluded: that system refuses automated access, so coverage is the House only.