PatternKilled

~90 chart-pattern hypotheses, tested all at once

0 of ~90

patterns that survived multiple-testing correction

In plain English

We tested about ninety chart patterns at once; after correcting for the fact that a few always look good by chance, not one held up.

What was claimed

Somewhere in ninety classic chart patterns there must be one that actually makes money.

The bar it had to clear

stay significant after correcting for testing ~90 patterns at once

What the data said

0 of ~90 survived

Why it's dead

Test ninety patterns and a few will look significant by chance. After correcting for multiple testing, none of ~90 candidate chart patterns survived. This is the honest denominator behind every other entry here.

The detail

~90 pattern hypotheses screened together. With a Benjamini-Hochberg correction for multiple testing applied (the standard statistical adjustment for the fact that testing many ideas at once guarantees a few false winners), zero survived. The handful that looked significant raw were false positives from the size of the search itself.

Kill date: 2026-03-12
Sample: ~90 hypotheses
Method: Documented kill
Verdict: multiple-testing wipeout