How much moved before the filing was public, and after?
An insider has two business days to file a Form 4; a fund has 45 days after the quarter ends to file its 13F. Whatever the stock did inside that lag, no reader could have caught. This page measures both sides of the filing date against SPY, on the same method as the congressional disclosure decay study.
This measurement is rebuilt once a day from the stored filings and price history, and the first build has not finished yet. Nothing is shown rather than a partial number.
How this is measured
What are the two windows?
Inside the lag runs from the as-of date (the Form 4 transaction date, or the quarter end a 13F reports) to the first session on or after the filing date: the move that happened before anyone outside could see the filing. After disclosure runs from that filing-date close for the next 30 days: the only window a reader could have traded. Both are the stock's return minus SPY's over the identical sessions.
Which filings are counted?
Purchases only: open-market buys (code P) on Form 4, and new or added positions on 13F. A sale or a trim says nothing about a forward return a reader could capture. Form 4s come from the newest 24 Form 4s of the last 180 days for each company we cover; 13F positions come from the funds we track, against each fund's prior 13F, dated by the filing they came from.
Why are some buckets blank?
A figure prints only from 25 distinct windows. Several purchases of one stock that share an entry and an exit are one window, counted once. Below the floor an average is noise, so the bucket is withheld and says so. An interval prints only from 20 disclosure weeks, because windows filed in the same week share most of their days.
Why does the 13F study have no intervals?
Funds file 13Fs in the same few weeks after each quarter, so one quarter of filings spans only a handful of weeks. The figures print once they clear the floor; the interval waits until the weeks do.
Does a positive number mean I should follow these filings?
No. This is an observed record on a stated sample, not a forecast and not a strategy. It ignores position size, assumes a fill at a daily close and counts no costs or taxes, and nothing here is a recommendation about any security or any filer.
Keep going: insider trades · fund 13Fs · congressional disclosure decay · what StockTools does not do