Trade Journalling and Performance Review

The limitations of recollection

Recollection of trading results is subject to systematic distortion. Trades that produced large profits are remembered in detail and attributed to skill; trades that produced losses are remembered less clearly and attributed to circumstance. Departures from the plan that happened to be profitable are not recorded as departures at all. The net effect is that a trader's mental account of performance is more favourable than the actual record and, more importantly, contains no reliable information about which behaviours are producing the losses.

A journal corrects this by recording each trade at the time it is taken, before its outcome is known, and by recording the outcome subsequently. Over a sufficient number of trades, the journal reveals patterns that are invisible in any single trade: a particular session in which losses concentrate, a setup that is taken frequently and rarely succeeds, a consistent tendency to exit winning positions before the target.

Information to be recorded

The transactional details of each trade are recorded first: the instrument, the direction, the position size, the entry price, the stop-loss level and the take-profit level. The MetaTrader 5 account history provides these details and can be exported as a basis for the journal. These data are necessary but insufficient, because they describe what was done and not why.

The material of greater value is the reasoning. The journal records the setup that justified the trade, with reference to the rule in the plan that it satisfied; the conditions that would have caused the trader to decline it; and the trader's state at the time, including fatigue, recent results and any sense of urgency. A screenshot of the chart at the moment of entry, with the levels marked, records the basis of the decision more accurately than a written description and is recommended for every trade.

Separating decision quality from outcome

A trading decision and its outcome are distinct. A decision that follows the plan can produce a loss, because the method has a win rate below 100 per cent and losses are an expected part of its operation. A decision that violates the plan can produce a profit, because the market can move in the intended direction for reasons unconnected with the trade's justification. A trader who evaluates trades solely by their financial result will, over time, learn to repeat profitable errors and to abandon correct decisions that happened to lose.

The journal therefore includes a separate assessment of each trade against the plan, recorded independently of the profit or loss. Over several hundred trades this column is more informative than the profit column, because it identifies whether the method is being applied. The four possible combinations of process and outcome are set out below; the third is the most dangerous, because it rewards a departure from the plan and encourages its repetition.

• Plan followed and trade profitable: the intended operation of the method.
• Plan followed and trade unprofitable: an expected cost of applying a method with a win rate below 100 per cent.
• Plan violated and trade profitable: the most hazardous outcome, since it reinforces the violation.
• Plan violated and trade unprofitable: an error whose cost has at least been made visible.

Periodic review for recurring patterns

The journal is reviewed weekly. The object of the review is not the assessment of individual trades, whose outcomes are largely determined by chance, but the identification of patterns across the sample. The relevant questions concern concentration: whether losses cluster in a particular session, a particular instrument, a particular setup or a particular state; whether winning positions are consistently closed before their targets; whether position sizes were increased after wins or losses.

Each identified pattern is recorded, and a specific corrective rule is added to the trading plan at the next monthly review. One pattern identified and corrected per month represents a substantial rate of improvement, and is more effective than a broad and unfocused intention to trade better. The review should be conducted at a fixed time, away from the trading session, and its findings recorded in writing.

Performance metrics

Several metrics are calculated from the journal over a sample of trades. The win rate is the proportion of trades closed at a profit. The average win and the average loss are the mean profit on winning trades and the mean loss on losing trades respectively. The expectancy is the average result per trade, calculated as (win rate × average win) − (loss rate × average loss), and must be positive for the method to be profitable over time. The maximum drawdown is the largest decline in equity from a peak to a subsequent low.

These metrics are meaningful only over a sufficient sample, and conclusions drawn from fewer than several dozen trades are unreliable. They are also calculated separately for trades that followed the plan and trades that did not, since the latter do not describe the method. A trader who tracks these figures over time has an objective measure of whether the method and its application are improving, independent of the fluctuations of any single week.