The trade not taken
· CRAFT · markdown source
An account records the trades you took. It has no field for the ones you did not.
That reads as bookkeeping and it is really a hole in how the craft gets learned. The usual story of skill is a loop: you act, the market answers, you adjust. The answer only comes back for actions that left a mark. Decline a setup and nothing happens — no fill, no line in the statement, no number at the end of the month that is different because of what you decided. The loop does not run. It is not that you learn slowly from the trades you skip. Left alone, you do not learn from them at all.
What fills the gap
Something always fills it, and what fills it is memory, which is a poor instrument here for a specific reason: the two kinds of skipped trade are not equally memorable.
A setup you passed on that then ran for three days without you is unforgettable. You watched it. It had a price attached and you can still name the price. A setup you passed on that would have stopped you out is not a memory at all — nothing happened, so there was nothing to store. Both were the same decision, made the same way, on the same evidence. One is available for recall and the other has no existence whatsoever.
So the informal record of your own restraint is assembled almost entirely from the occasions when restraint was expensive. Anyone reviewing that sample honestly would conclude they are too cautious. It is the only conclusion the sample supports, and the sample is not real.
The taken half is not clean either
The other half is graded, but badly. A trade that made money gets filed as a good decision. A trade that lost gets filed as a mistake. In an environment where a real edge might be wrong four times in ten, that grading is close to noise. Repeat it for long enough and what you have taught yourself is the shape of the last few months’ luck.
Put both halves together. Half your decisions are graded by a coin. The other half are not graded at all, and are remembered through a filter that keeps only the painful cases. That is the raw material most of us are calling experience.
Writing it down at the time
There is one fix and it is unglamorous. Write the decision down before the outcome exists: what you saw, what you expected, what would have to be true for you to be wrong, and — the part everyone skips — the same note for the setups you turned down, and why.
The point is not the archive. The point is the timing. A reason recorded before the result cannot be adjusted by the result. Written afterwards, the same sentence is a reconstruction, and reconstruction runs downhill toward whatever happened. You will not catch yourself doing it. Nobody does, which is why it is worth engineering around rather than resolving to avoid.
After a few months of it the declines become a sample instead of a grievance. Some of them turn out to be good decisions with bad outcomes, which is the whole category an account statement has no way to represent. A few turn out to be the same avoidable hesitation eleven times over, which is a thing you can work on and would never have found by feel.
What comes out of this is not flattering, and it is not meant to be read for comfort. It is the only version of your own history that was not written by someone who already knew the ending.