How to review a week without getting lost in metrics

Check the quality of the record before interpreting it

Start by finding missing trades, duplicate fills, inconsistent fees, incomplete tags and notes written too late. A clean total built from incomplete records can still tell the wrong story.

Do not reconstruct feelings days later as though they were precise observations. If you want emotional context, record emotions close to the decision. During the review, mark missing context as missing instead of filling the gap with hindsight.

Summarise the week with five useful indicators

Use a compact dashboard: number of trades and sessions, result in R and realised risk-reward ratio, percentage of trades executed according to the plan, and frequency of the behaviour you are trying to change.

These are not the only valid metrics. They cover activity, outcome, exposure, execution and behaviour without repeating the same information under different names. Add another measure only when it answers a defined question.

Separate market context, outcome and execution

Group trades by market, time, setup and condition. Then compare execution within each group. A difficult environment, a losing outcome and poor execution are three different findings.

Score execution quality using observable rules: valid setup, planned entry, respected risk, coherent management and justified exit. This prevents a profitable deviation from hiding behind P&L and a well-executed loss from being treated as a mistake.

Compare with a baseline, not with an ideal week

A weekly value has meaning when you compare it with your normal range. Put this week beside the previous four or another relevant baseline. Ask whether the change is large, repeated and linked to comparable conditions.

If you take only a few trades, one week may not provide enough comparable observations. Keep the routine, but delay the conclusion until a fortnight or month contains a useful sample. The calendar should serve the method, not force a verdict every Friday.

Find the story behind the numbers

Select three cases: one well-executed trade, one clear deviation and one decision that still raises questions. Reconstructing them adds the detail that averages remove.

Look at sequences as well as isolated trades. Compare post-loss trades, decisions after missing an entry and behaviour after a strong winner. A stable weekly total can hide a real improvement in patience and a deterioration in exits.

Choose one discovery instead of collecting five

Prioritise the finding with the best combination of frequency, impact and ability to intervene. A costly exception may require immediate protection, while a moderate behaviour repeated three times a week may be the clearer improvement target.

Write the evidence neutrally: “Risk exceeded the limit in four of six trades after a loss.” Avoid diagnoses such as “I lack discipline”. Data should describe a decision, not define the person who made it.

Write a weekly conclusion you can test

Use four parts: what you observed, where it appeared, what impact it had and what you will do. For example: “I entered late in four of nine setup A trades. Three followed a missed signal and reduced average reward-to-risk. Next week I will not chase entries more than 0.3R beyond the planned level.”

Keep observations separate from changes to the trading plan. A single week can justify a protective barrier or a new hypothesis; it rarely proves that a strategy parameter must be redesigned.

Close with one measurable mission

Define a trigger, a behaviour and a period: “For the next five sessions, after a loss I will wait fifteen minutes and keep risk below 0.8%.” The next review begins by checking whether that mission was followed.

One priority is a method of focus, not a universal law. If two risks require immediate containment, protect both. But avoid changing several unrelated variables at once, because you will not know which adjustment produced the result.

A repeatable 60-minute weekly review template

Reserve one hour with the market closed: 15 minutes to clean the record, 10 to summarise indicators, 15 to compare contexts, 10 to review the previous mission and 10 to choose the next action.

If an important question needs more data, save it as a hypothesis. Do not mix periods or search for the statistic that confirms your first impression. A time limit makes the routine sustainable; it does not force an answer.

Connect weekly action with monthly perspective

Use the week to adjust nearby behaviour and the month to test persistence. At month-end, compare four reviews: did the same deviation return, did mission adherence improve, and did the intervention create a new cost?

The trading journal keeps each review, its evidence and its mission connected to the next period. The record becomes a trace of what you observed, what you changed and whether the change held, rather than a collection of isolated reports.

Reserve strategic changes and backtest conclusions for a sample that can support them. The week corrects the gesture; the month checks whether that gesture is changing the way you trade.