About
This site is an independent education and reference library on market manipulation. It explains how each manipulation technique works and maintains a structured, machine-readable index of enforcement actions brought by financial regulators, each linked to its primary source.
What this site is
Two things, held together by one join.
The first is an explanation of how market manipulation works — technique by technique, in plain English, for an intelligent reader who is not a specialist. Each technique page defines the technique in a way that stands on its own, walks through the mechanic step by step, shows the arithmetic with a worked example, cites the statute it breaks with a link to the primary text, explains how surveillance actually catches it, and states the red flags an outsider can observe.
The second is a library of enforcement actions, compiled from the primary releases of financial regulators and stored as structured records. Each record carries the agency, the dates, the procedural status, the respondents, the monetary relief, and a link back to the regulator’s own document.
The join is the technique tag. Every case is tagged with the techniques it involves, which is what lets a technique page report the penalties that technique has actually attracted — computed from our own records, at build time, rather than quoted from anywhere.
Why it exists
Because the two halves are usually kept apart, and both suffer for it.
Explanations of market manipulation tend to be either legal commentary written for practitioners or journalism written around a single case. Neither gives a reader the thing they usually want: how does this actually work, mechanically, with numbers, and what has happened to people who did it.
Enforcement data, meanwhile, exists as thousands of individual press releases in a dozen formats across a dozen regulator websites. It is public, and it is very nearly unusable in aggregate. Making it a clean, structured, downloadable dataset is a small amount of engineering and a large amount of value, and nobody was doing it for this subject.
How it is built
The site is static. Every page is a file, generated at build time from content in the repository. There is no database and no server-side code.
The case library is compiled by a set of ingest scripts that read regulator indexes, fetch the release pages and filings, extract structured fields, and classify each action against a technique taxonomy using keyword rules. Everything fetched is cached and never re-requested. Requests are serialised per host with a delay and identified by a user agent carrying a contact address.
Structured fields are extracted mechanically. The summaries and narratives on case pages are composed from those extracted fields, not from the regulator’s prose — no release text is copied onto this site. That constraint is enforced in the pipeline rather than left to discipline.
The full method, source by source, is on the sources page, including the sources we cannot reach and why.
What this site does not do
It does not measure how much market manipulation there is. It measures enforcement. Those are different populations, separated by an unknown and unknowable detection rate. Every chart on this site carries that caveat, and it is worth restating: a rise in the number of actions in a year tells you about regulators, not about markets.
It does not give advice. Nothing here is legal advice or investment advice, and the terms say so at greater length.
It does not accuse anyone of anything. Where an enforcement action has been filed and not resolved, the conduct is alleged. That is how it is described throughout, and the editorial policy explains the rules the site follows.
It does not publish personal information. Pages about individuals are limited to legal facts drawn from public filings, each with a source. No addresses, no contact details, no photographs, no speculation about anyone’s private life.
Corrections
If something here is wrong, we want to know, and we log the fix in public. The corrections page explains how to tell us and what happens next. Where a case has resolved in a respondent’s favour, that correction is the most important kind and gets priority.
Reuse
Our writing, our classification and our compiled data are published under CC BY 4.0 — free to reuse with attribution and a link. The underlying US federal releases are works of the United States government and are not subject to copyright at all.
The machine-readable exports are at /cases.json, /llms.txt and /llms-full.txt, with per-chart CSVs under /data/.