For about a year the strategy scaled in. Price moves against the position, we add lower, we wait for the turn. It is the most natural thing in the world to do and it feels like conviction.
We measured it properly: the same nine years, the same signal, the same total exposure, the only difference being whether the position was taken at once or built up as the market moved against us. Single entry won, and not narrowly. The comparison is in Letter 001, where it belongs with its date.
A pyramid adds size precisely when the position is wrong.
That sentence is the whole result. Every other property follows from it. The times you are most heavily committed are, by construction, the times the market has most strongly disagreed with you. When the trade comes back you make more, which is the memorable part. When it does not, you lose on a position several times larger than the one you originally chose, which is the part that gets remembered as bad luck.
The equity curve of a pyramid is pleasant almost all of the time. That is not a defence. It is the mechanism by which it survives long enough to do damage.
Because the wins are vivid and the losses are attributable.
Every recovery from a deep average-down is a story with a hero in it. Every failure has an obvious external cause: an unusual move, a news event, a market that "was not itself". The structure produces its own excuses, and they are all true individually.
The only way past that is to compare at identical exposure, which is the comparison almost nobody makes, because it is not the comparison the strategy invites you to make.
There is no flag. There is no configuration value that turns averaging down back on, and no branch in the engine that would execute it.
That was deliberate. A dormant capability is a capability, and the moment when scaling in looks most attractive is exactly the moment when a flag becomes easy to justify. Deleting it means the argument has to be had in a text editor, in daylight, and not at the point of maximum discomfort.
One consequence is worth stating: the strategy now takes its full intended size at entry and lives with the drawdown that follows. That is less comfortable week to week. It is the trade we made knowingly.