PatternKilled

Buy HYPE after an “under-trend-gap” signal

+2.11 bps

net return, but over just 8 trades, you can’t tell skill from luck

bps = basis points = a hundredth of a percent (100 bps = 1%).

In plain English

This pattern made a tiny profit, but over only eight trades, far too few to tell skill from luck.

What was claimed

A specific chart pattern, the 'under-trend-gap', reliably precedes a HYPE bounce, so buy when it appears.

The bar it had to clear

a real signal, not noise

What the data said

over just 8 trades, indistinguishable from random

Why it's dead

The direction is slightly positive but n=8 with p=0.923 is not a result, it's randomness with a positive label. A signal that fires 8 times in a 90-day window cannot accumulate enough evidence before conditions change.

The detail

n=8 trades. Mean net: +2.11 bps. p=0.923, not remotely significant. The signal fires too rarely to evaluate within a meaningful window. Even if the direction held, the combination of ultra-low frequency and zero statistical power makes this untestable in any reasonable timeframe.

Kill date: 2026-03-12
Sample: n=8
Method: Documented kill
Verdict: insufficient sample (n=8)