DirectionalKilled

A profitable friend's trend-following, systematised into a bot

92%

of live net P/L came from a single trade out of 44, strip it and the edge is gone

In plain English

We turned a profitable friend’s trend-following into a live bot on real capital. It made money for 26 days, but 92% of the profit was a single trade, and the three-month backtest of the same setup lost to just holding the coin.

What was claimed

Turn a profitable trader's approach into a bot and the profits come with it.

The bar it had to clear

beat simply holding, with no single trade carrying it

What the data said

92% of profit from one trade; lost to holding over 92 days

Why it's dead

The live net looks positive, until you realise one trade drove almost the entire result. The 92-day backtest of the same configuration reveals it's a buy-and-hold underperformer.

The detail

n=44 live trades on real capital over 26 days. A single trade (#1) accounted for 92.3% of all net P/L; the other 43 combined for the rest. 92-day backtest of the same configuration: it finished net-negative while simply holding finished net-positive, and lost to unlevered HODL (just holding the coin, no leverage) across all five BTC pullbacks in the window.

Kill date: 2026-05-09
Sample: n=44 live + 92d backtest
Method: Documented kill
Verdict: single-trade concentration