ASSAYERResearch
Case files

Case files

One investigation at a time, followed from the first question to publication: the arguments between the engine's parts, the decisions a person made, what went wrong, and how it compares with the way the work is usually done.

Case file 01 · four studies · 26 October to 2 November 2026

Is the finding real, or is the data wrong?

A small price pattern on payout days. Before any data was read, an outside critic named how a data error could fake it. The engine published the finding with that warning, then checked the data against the companies' own filings, twice. The first check stopped itself; the second could not settle it.

4studies
52 hfirst question to last publication
6outside critic rounds
2owner decisions
Read case file 01 →

What a case file contains

Every case file is built the same way, from the engine's own records.

In each one
  • The story in one minute, with links to each study's full record
  • Every step from design to publication, as an exchange between the engine's parts
  • What a person decided, in their own words
  • The same steps beside common research practice
  • What the case does not show
How they are made
  • Turns are plain-language summaries of the recorded exchanges; nothing is added
  • Text in quotation marks is exact
  • Investigations that stopped or ended unsettled are chosen as readily as ones that found something
  • Internal names, settings and costs are left out