Log review
Reading back through hypothesis, process, emotion, and outcome.
System Review and Rule Iteration
A journal full of entries is only raw material. Turning it into an actual improvement means confirming a real, repeating pattern before writing a specific rule change — never the other way around.
- Describe the four-stage refinement cycle: log review, pattern identification, rule update, and re-test.
- Explain why a pattern needs to repeat across multiple entries before acting on it.
- Write a specific, checkable rule update rather than a vague resolution.
- Re-test a rule update against the trades that revealed the pattern.
- Avoid overreacting to a single trade's outcome.
Three losses, then you change entry, stop, size, and session. Can you still learn anything?
Rule iteration should begin with an identified process problem, make a limited versioned change, and gather new evidence instead of rebuilding the system around the latest P&L. Use trade attribution before changing a rule.
Quest 49 · Review & IterateWhat should be fixed first?
Diagnose 16 simulated trades, then change exactly one variable
Do not hunt for the worst-looking trade. Review the sample: whether the entry rule was valid, size complied, execution deviated, and the final R. The dataset deliberately contains execution deviation as the clearest recurring controllable problem.
#RuleSizeExecR
1✓✓✓-1R
2✓✓×+0.8R
3✓✓✓-0.6R
4✓×✓+1.4R
5✓✓×-1R
6✓✓✓+0.5R
7✓✓×-0.9R
8✓✓✓+1.8R
9✓✓✓-0.4R
10✓✓×-1R
11✓✓✓+1.1R
12✓×✓-0.7R
13×✓✓+0.6R
14✓✓×-1R
15✓✓✓+1.5R
16✓✓×-0.5R
Rule design1
Execution6
Sizing2
Sample noise3
If several variables change at once, an improved next sample cannot tell you which modification mattered.
Four Stages Turn a Journal Into an Actual System Update
A trade journal, covered in an earlier module, is only useful if it's actually reviewed. That review follows four stages: reading back through entries, identifying a pattern that repeats rather than a single event, writing a specific rule update, and re-testing that update against the trades that revealed the issue.
Pattern identification
Confirming a deviation repeats, rather than acting on one instance.
Walk Through the Four Stages of Turning a Review Into a Rule Update
Switch between the four stages and read what each one is actually meant to accomplish.
A
Log Review
Reading back through trade journal entries — hypothesis, process, emotion, and outcome — across a meaningful sample. This step surfaces raw observations before any conclusions are drawn.
B
Pattern Identification
Looking for a deviation or mistake that repeats across multiple entries, not a single isolated trade. A pattern needs to show up more than once to be distinguished from ordinary variance.
C
Rule Update
Writing a specific, checkable change to the entry rule, exit rule, or execution checklist based on the identified pattern. The update should be specific enough to actually change behavior, not just a vague resolution to be more careful.
D
Re-Test
Checking whether the updated rule would have changed the outcome of the specific trades that revealed the pattern. This step confirms the update actually addresses the pattern before it's adopted going forward.
One Trade Is a Data Point — a Pattern Needs Repetition
Rewriting a rule after a single losing trade risks reacting to ordinary variance rather than a genuine, fixable issue — the same distinction covered when reviewing win rate and drawdown in earlier modules. A rule change is warranted once the same deviation shows up across multiple, separate entries, not after any single outcome.
Was Each Refinement Done Properly?
Pick a case and judge whether the refinement process was followed correctly.
A
Overreacting to one trade
After a single losing trade, a trader immediately rewrites their entire entry rule without checking whether other trades show the same issue. This overreacts to a single data point rather than confirming a repeating pattern first.
B
A specific, evidence-based update
A trader notices the same execution mistake — entering before their rule's condition was fully met — across five separate trade journal entries, and writes a specific new checklist item to catch it. This reflects a properly identified, repeating pattern addressed with a specific, checkable rule update.
C
A vague resolution
After noticing a mistake, a trader resolves to just be more disciplined next time, without changing any specific rule or checklist. This is not a specific, checkable change and is unlikely to actually alter future behavior.
Four Checks Before You Rewrite a Rule
1
Sample reviewed
How many journal entries were actually reviewed?
2
Pattern confirmed
Does the deviation repeat across multiple entries, not just one?
3
Specific update
Is the rule change specific and checkable, not a vague resolution?
4
Re-tested
Would the update have changed the outcome of the trades that revealed it?
Revise One Rule for One Documented Reason
Log review, pattern identification, rule update, and re-test.
A single trade is a data point, not a confirmed pattern.
A checkable rule change is what actually alters future behavior.
- 1Record the observationWrite the simulated fact and its source before assigning a meaning.
- 2State the boundaryName what the current evidence cannot establish and what must still be checked.
- 3Keep a counterexampleRecord what later evidence would require the explanation to be revised.
Put Your Understanding to the Test
Submit your answers to see detailed explanations.
Question 1 of 3
After a single losing trade, a trader immediately rewrites their entire entry rule. What is the issue?
Question 2 of 3
A trader notices the same execution mistake across five separate journal entries and writes a specific new checklist item to catch it. What does this represent?
Question 3 of 3
A review turns up four entries where the trader jumped in before the confirmation bar closed. Three candidate updates get drafted. Which one gives the next entry a condition that can be checked before the order goes in?
Stuck? Ask Mira to Break It Down
Describe the pattern you've noticed across your journal entries, and Mira can help you draft a specific, checkable rule update — it won't review your journal for you.
Checking sign-in status...
Put this concept into practice
A Review Isn't 'Lost, So Change the Strategy'
Understand that a review has to first distinguish an execution gap, normal variance, and a strategy/environment problem, without changing the rules right away because of a handful of wins or losses; changing a strategy needs evidence, and only a limited number of variables should change at once.