Rule Adherence in a Trading Journal: Score Process, Not P&L

A profitable trade is not automatically a good trade. An impulsive entry can win, while a fully on-plan setup can stop out. If a journal labels decisions by outcome, random reinforcement teaches the trader to repeat lucky mistakes and abandon sound rules after normal losses.

Rule adherence measures whether a decision matched the written process using information available at the time. It creates a process score that can be reviewed separately from P&L, then connected to results across a meaningful sample.

What is rule adherence in trading?

Rule adherence is the proportion or degree to which a trade follows predeclared entry, risk, management, and exit rules. The rules must be observable enough that the trader—or another reviewer—can reach the same score from the record.

• Entry: the approved setup, confirmation, session, and timing were present.
• Risk: the stop, position size, and total exposure stayed within limits.
• Management: partials, trailing rules, and no-add conditions were followed.
• Exit: the planned target, invalidation, time exit, or discretionary protocol was used.
• State: daily loss, news, fatigue, and no-trade restrictions were respected.
• Documentation: required plan, screenshot, or note was captured at the defined time.

A rule such as “take only good setups” cannot be scored. A rule such as “long only after the 15-minute close above the opening range, before 11:00, with the stop below the trigger low” can. If a score repeatedly causes debate, improve the written rule before trying to improve the percentage.

Start with a minimum viable scorecard

The best scorecard is short enough to complete on every trade. Begin with four to six rules that protect the strategy’s edge or account risk. Do not score cosmetic preferences beside critical risk controls as though they have equal importance.

1. Write one observable pass condition for each critical rule.
2. State when evidence must be captured: before entry, during management, or after exit.
3. Define whether a violation makes the whole trade off-plan.
4. Test the scorecard on ten historical examples for ambiguity.
5. Freeze the wording for a review cycle before comparing scores.

A complete trading plan template may contain many details. The adherence scorecard is its measurable control panel, not a duplicate of every sentence.

Binary model: pass or fail

The simplest model marks each rule 1 for followed and 0 for violated. A trade-level percentage is the number of passed rules divided by the number of applicable rules.

Binary scoring is fast and repeatable. It works well for clear constraints such as “risk no more than 0.5%,” “no entry after 11:00,” or “never widen the stop.” Its weakness is severity: missing a screenshot and doubling risk both score zero unless a critical-fail rule distinguishes them.

Add critical fails without hiding the component score

Define a small number of safety violations that make the trade off-plan regardless of the percentage: exceeding maximum risk, trading after a daily stop, removing the protective stop, or taking an unapproved setup. Preserve the numerical score as detail, but classify the trade off-plan.

Graded model: measure partial execution

Some decisions are not naturally binary. An exit protocol may allow discretion within a defined range, or documentation may be complete, partial, or absent. A graded model can assign 0, 0.5, or 1 to each rule, provided each level has an observable definition.

Suppose entry quality has weight 3 and scores 1; risk has weight 4 and scores 1; management has weight 2 and scores 0.5; documentation has weight 1 and scores 0. The weighted score is (3 + 4 + 1 + 0) ÷ 10 = 80%. An unweighted average would be 62.5%, showing how weights materially change the story.

• Use binary scoring when rules are objective and speed matters.
• Use graded scoring only where partial compliance has a real definition.
• Use weights sparingly and document why one rule matters more.
• Keep critical risk limits as hard fails even in a graded model.
• Do not tune weights to make a disappointing month look disciplined.

Choose one model and make it auditable

More detail does not guarantee more truth. A 20-question score with subjective 1–10 ratings creates false precision and review fatigue. Two reviewers should be able to score a saved chart and plan with minimal disagreement.

1. Give each rule a short identifier and plain-language pass condition.
2. Attach evidence such as planned levels, timestamps, size, or screenshots.
3. Use N/A only when the rule genuinely does not apply.
4. Record the score before opening the final P&L when practical.
5. Audit a few trades monthly for consistency between the evidence and score.

Separate trade score from period score

A daily or weekly score should aggregate applicable rule decisions, not merely average trade percentages when trades contain different numbers of applicable rules. The clean formula sums passed rule units across all trades and divides by all applicable rule units.

Also report the share of trades classified fully on-plan. A week can have 92% component adherence but only 60% fully on-plan trades if small violations are spread across many entries. Both views are useful: the component score measures frequency, while the on-plan share measures complete execution.

Compare on-plan and off-plan expectancy

Once trades are classified, calculate performance separately. Trading expectancy in R allows trades of different cash size to remain comparable, while the R-multiple guide explains how to preserve the initial risk denominator.

Consider 50 trades. Forty on-plan trades average +0.25R and contribute +10R. Ten off-plan trades average −0.60R and contribute −6R. The full book earns +4R, only +0.08R per trade. The strategy may have a viable on-plan edge while deviations consume 60% of its gross contribution.

Now reverse the short-term outcome: five off-plan trades happen to average +0.8R. That does not validate the violations. The sample is tiny, the trades may carry uncontrolled tail risk, and profitable rule breaking erodes the ability to distinguish strategy from improvisation. Continue to classify them off-plan.

• Report count beside expectancy so a three-trade subgroup is not treated as established.
• Compare average win, average loss, and tail losses—not expectancy alone.
• Keep strategy and market regime reasonably stable inside each comparison.
• Include costs and slippage in realized R.
• Separate critical violations from minor documentation misses where useful.

Do not change a strategy because off-plan trades lost

An off-plan loss is weak evidence about the planned strategy because the strategy was not executed. If a breakout rule requires a close above resistance but the trader anticipates the close and loses, changing the breakout stop or target in response uses contaminated evidence.

This does not mean ignoring the loss. Record its full R impact, review the trigger, and install a process control. The losing-trade journal guide helps separate market information from execution information without excusing either.

Opportunity quality and execution quality are different

Trade-only adherence can look perfect while the trader skips valid setups. It can also punish restraint if it scores only entries and treats every no-trade as a missed opportunity. A complete review begins with qualified opportunities: moments that met the plan whether or not a position was opened.

Opportunity score: did the planned situation occur?

An opportunity log records every qualified setup using a fixed scan universe and review window. This denominator allows you to identify valid trades taken, valid trades skipped, and invalid trades taken. Without it, a journal sees commissions but not omission.

Execution score: what did you do with the opportunity?

Execution evaluates entry, risk, management, and exit after a valid opportunity appears. An excellent setup can receive poor execution; a mediocre but valid setup can receive excellent execution. Keep setup quality and execution quality in separate fields.

Example: during a fixed week, 20 valid opportunities occur. Sixteen are taken on-plan, two are skipped without a plan-based reason, and two are intentionally skipped because a daily risk limit is active. Opportunity capture is 16 ÷ 18 = 88.9% if risk-limit exclusions were predeclared as non-actionable. If two additional invalid setups were traded, false-action rate is 2 ÷ 18 = 11.1%.

• Valid and taken on-plan: correct opportunity recognition and execution.
• Valid but skipped: possible hesitation, capacity limit, or legitimate no-trade rule.
• Invalid but taken: selection failure or impulse.
• Valid and taken off-plan: opportunity was real, execution deviated.
• Invalid and skipped: correct restraint, usually logged only in aggregate.

Build a compact mistake taxonomy

The adherence score says whether a rule failed. A mistake category says what kind of control should prevent recurrence. Keep this layer small and causal; do not rebuild the entire setup and context taxonomy explained in the trade-tags analytics guide.

• Selection: entered without the required setup or context.
• Timing: entered early, late, or outside the approved session.
• Risk: oversize, correlated exposure, missing stop, or exceeded daily limit.
• Management: unauthorized add, moved stop, or unplanned partial.
• Exit: fear-based early exit, ignored invalidation, or missed time rule.
• Process state: revenge, FOMO, fatigue, distraction, or checklist bypass.
• Documentation: missing evidence that prevents reliable review.

Use one primary mistake category and, when necessary, one specific behavior. “Risk → moved-stop” is more useful than applying six emotional labels to the same trade. Setup tags answer what you intended to trade; mistake categories answer where the process failed; notes preserve the unique context.

Measure the cost of each mistake carefully

The simplest cost is the realized R of trades carrying a mistake category. That is an impact total, not necessarily the causal cost: an on-plan version of the same trade might also have lost. A stronger estimate compares realized execution with a predeclared compliant alternative when that alternative can be reconstructed without hindsight.

For example, a stop should have closed a trade at −1R but was widened and the position closed at −2.2R. The excess execution loss is approximately −1.2R before any fill differences. In contrast, an unplanned entry that loses −1R has no guaranteed counterfactual profit; report the full impact but do not claim the mistake “cost” a winner you cannot observe.

Consistent stop handling is especially important. The stop-discipline guide shows why redefining risk after entry corrupts both adherence and the planned-versus-realized risk-reward comparison.

A worked weekly process review

Suppose a trader takes 24 trades with 120 applicable binary rule checks. They pass 108 checks, for 90% component adherence. Eighteen trades are fully on-plan and six contain at least one violation, so fully on-plan share is 75%.

• 18 on-plan trades: +4.5R total, +0.25R expectancy.
• 6 off-plan trades: −3.0R total, −0.50R expectancy.
• All 24 trades: +1.5R total, +0.0625R expectancy.
• Three timing violations: −0.8R total.
• Two risk violations: −2.0R total, including one moved stop.
• One documentation violation: −0.2R outcome, but no proven causal cost.

The correct response is not to alter the entry setup because the all-trade expectancy looks weak. The on-plan sample remains positive but small; the immediate control is the high-impact risk violation. The next week’s commitment might be “stop orders are entered with the position and can only move toward reduced risk.”

Use the right review cadence

After each trade: score while memory is fresh

Complete the rule checks, attach evidence, and classify any deviation. Do not redesign the strategy or write a long emotional essay. The goal is accurate capture.

Daily: enforce safety limits

Review critical failures, daily loss status, unlogged opportunities, and whether trading should stop. Daily review protects the account; it is not enough data for strategy conclusions.

Weekly: choose one process control

Run the weekly trade review: component adherence, fully on-plan share, on-plan versus off-plan R, opportunity capture, and the highest-impact recurring mistake. Commit to one observable control for the next week.

Monthly or per sample: evaluate patterns

Compare rolling adherence and expectancy by stable strategy version. Audit score consistency, merge duplicate mistake categories, and decide whether a rule is unclear, impractical, or repeatedly ignored. Strategy changes belong here only when the on-plan evidence supports them.

Do not overreact to a high or low score

A 100% adherence week can lose money because outcomes vary. A 60% adherence week can make money because violations got lucky. Process and outcome should become related over a sufficiently broad sample, but they can diverge sharply in short periods.

Sample size also applies to the on-plan/off-plan comparison. Ten off-plan losses do not precisely estimate future off-plan expectancy, and 20 on-plan winners do not prove the strategy. Use the framework in how many trades to test a strategy, keep rule versions stable, and show uncertainty beside subgroup results.

• Do not lower the pass threshold after a bad week.
• Do not add rules solely because one trade lost.
• Do not remove a valid rule because breaking it produced a winner.
• Do not split every mistake by session and setup until samples disappear.
• Do not increase size because recent adherence is high without edge and risk evidence.

When the score reveals a rule problem

Repeated non-adherence is not always a motivation problem. The rule may be ambiguous, impossible to execute at market speed, dependent on unavailable data, or incompatible with the trader’s schedule. A process review should distinguish unwillingness from poor design.

1. Identify the rule with repeated failures across several review periods.
2. Inspect evidence and ask whether independent reviewers agree on pass or fail.
3. Determine whether the failure comes from clarity, capability, environment, or impulse.
4. Design the smallest control: alert, order template, checklist gate, size cap, or rule rewrite.
5. Version any material rule change and evaluate future trades separately.

A rule should change because it is undefined, unsafe, impractical, or contradicted by adequate on-plan evidence—not because following it happened to lose last Tuesday.

A practical journal template

• Strategy and version: which fixed rule set applied?
• Qualified opportunity: yes, no, or excluded by a predeclared constraint?
• Entry, risk, management, exit, and state checks: pass, partial, fail, or N/A.
• Critical failure: yes or no, with the violated control.
• Overall classification: on-plan or off-plan.
• Primary mistake category and one specific behavior if needed.
• Initial R, realized R after costs, and planned versus realized exit.
• Evidence: chart, order details, note, and relevant timestamp.
• Next control: only when the review identifies a repeated actionable pattern.

Keep process fields alongside the trade rather than in a disconnected spreadsheet. That makes it possible to filter all on-plan trades for expectancy, inspect every risk violation, and revisit the chart without reconstructing context from memory.

How Traderizz supports rule-adherence review

Traderizz keeps trades, strategies, R-multiples, tags, screenshots, notes, and diary context together. Use a stable on-plan/off-plan field or controlled process labels, then filter each group and compare count, total R, average R, win rate, and expectancy.

During review, pair process labels with the journal workflow for losing trades and the compact taxonomy from trade tags and setup analytics. The objective is not a flattering discipline score. It is a reliable feedback loop that protects a valid edge from execution noise and keeps normal strategy losses from triggering random changes.