For about a year the strategy waited for a pattern before entering. We read candles, looked for signs the move was running out of steam, and took a certain amount of pride in being good at it.
Then we measured whether the pattern contributed anything, by running the same strategy with the pattern requirement removed and the entry decided by a schedule instead.
It had stopped mattering. Whatever the pattern once added was no longer detectable.
Entries now run on a fixed schedule, in a small number of slots each day, and the strategy does not look at candle colour at all.
The part worth recording is not the result but the resistance to it. Pattern reading is the most enjoyable part of the work. It feels like skill, it produces a story for every trade, and giving it up feels like giving up craft for a timetable.
The market is indifferent to how elegantly we justify our timing. That sentence cost a year.