Method, mostly.
Notes on how this data behaves when you look at it closely — including the measurements that came back against us. No market calls, no newsletter, no "5 stocks to watch".
13F filings: what they show, what they hide, and the CUSIP problem
23.4 million institutional positions from 9,985 managers — filed 45 days late, long-only, and by CUSIP rather than ticker. All three facts change how you read them.
Read →Post-earnings drift: we measured it on 40 quarters. Here is the number.
+0.64% excess return, t = 2.83. Small, real, and much smaller than the versions you will be sold. Here is the method that produced it.
Read →How to read a Form 4, and which insider trades are worth anything
Insiders file within two business days. Most of what they file is noise — scheduled sales and option exercises. Here is how to tell the difference.
Read →45% of volume sold short is normal — reading short interest properly
The most misread number in retail research. Short volume is not short interest, and half of it is market makers doing their job.
Read →Survivorship bias: what it does to a backtest, with numbers
Between 10% and 23% of the positions in a historical run are companies that no longer exist. If your data does not contain them, your backtest never took their losses.
Read →What an AI can and cannot tell you about a stock
The useful part is not the verdict. It is that a model given real data has to quote it — and quoted numbers can be checked.
Read →Or skip the reading and open the data.
No card, no demo call, no sales email. The terminal is public — an account raises the limits, and the AI runs on credits.
Free plan for good. Pro is $69.99/mo, or $4.99 for a 7-day trial.See what each plan includes →