About

Quant Data is an independent research project run by one person. It exists because learning to trade quantitatively is mostly a problem of not being able to check anything, and that is a solvable problem.

The idea

If you are learning markets on your own, you can get charts, opinions and courses easily. What you cannot easily get is the thing that would actually teach you: a measured answer. Is today a trend day or does it just feel like one? Does the rule your course repeats hold up over sixteen years? Does volume confirm price, and by how many percentage points?

Quant Data builds those answers into a handful of endpoints and free pages, and shapes them so an AI agent can fetch them mid-conversation. You ask a question in plain language, your agent pulls the number, and you get to compare your own read against something that was measured. That loop is the product.

Why the failures are published

Nearly every number here could have been made bigger. Fit the model on all the data instead of holding years back, pick the best of eleven variants, quote the 20-bar version of a claim instead of the 5-bar one, and 52% becomes 80% without a single lie being told out loud.

So the failures are published in the same place as the wins: the chart patterns that turned out to be a coin flip, the markets where the edge inverts, the pre-registered test on the Asian session that came back NO_GO. A win rate you cannot see the denominator of is worth nothing, and the only way to show you the denominator is to show you everything that was tested. If we ever stop publishing failures, stop trusting the successes.

It also means you can calibrate what you buy. The measured edges here are between 51% and 56%. That is what a genuine edge looks like in one of the most liquid markets on earth. Anyone showing you 90% is measuring something narrow, or is not measuring.

Affiliation, plainly

Quant Data is not affiliated with, endorsed by, or connected to Al Brooks or Brooks Trading Course, David Weis or the Weis Wave software, Thomas Bulkowski, any exchange, any broker, any data vendor, or any trading educator.

Their names appear here because their published ideas are what is being tested. The Brooks price- action tradition supplies the day-type vocabulary; the Wyckoff and Weis tradition supplies the volume-wave events; Bulkowski's chart-pattern catalogue supplied the detectors that did not work. Testing someone's published claims is commentary and research, and it cuts both ways — sometimes the measurement supports them and sometimes it does not. If you want to learn any of these methods properly, buy the authors' own books and courses. We are not a substitute for them.

Contact

One address for everything: quantdata@agentmail.to — key requests, renewals, bug reports, corrections, and arguments about the statistics. A real person reads it.

Want a key?

Tell us in one sentence what you are building or studying. We reply with a Stripe link, and the key follows once the first month is paid — usually within a day.

Email quantdata@agentmail.to Read the methodology

No signup form, no card on this site. Keys are issued by hand, one at a time. Max Pain pages stay free and need no key at all.