Reviewing a period of trades

Why a closed period

A period still in progress cannot be reviewed, only monitored, and the two produce different writing. A running week invites conclusions about trades whose outcome is not yet known; a closed one has a fixed set of facts in it.

It also makes the exercise repeatable. Reviews of comparable periods can be read in sequence, and a note from six weeks ago is only useful if it describes the same kind of window as this one.

Which trades get a second look

Memory selects badly — it offers the largest loss and the most recent trade, which are rarely the informative ones. A record can select on properties instead:

• Trades with no notes, no setup tag or no stop recorded. These are the entries that make every statistic less reliable, and they are cheapest to fix while the period is recent.
• Trades whose MFE was far above their result — what was reached and given back.
• Trades tagged with a mistake, grouped by which mistake, rather than read one at a time.
• Outliers in size: anything much larger than the book's own median position, whether it won or lost.
• The trades taken immediately after a loss, as a group.

What gets written down

The output of a review is prose, not a score. A number that summarised the period is already in the reports; what is not stored anywhere is the reading of it — what the period looked like, what was repeated, and what specifically to watch in the next one.

Writing it against the period rather than against a trade matters, because the observations that matter across a week are rarely about a single trade. A note attached to one entry is found again only by someone reading that entry.

Editing the record while reviewing it

A review pass is usually also a correction pass: filling in a missing tag, adding the lesson a trade produced, marking a mistake that was obvious afterwards. This is the last point at which the context is still recoverable, and a book corrected weeks later is a book of reconstructions.