How to Review Your Trades
To review your trades, compare each one against the rules you wrote before taking it, and grade the process separately from the result. A trade that followed the plan and lost is a good trade; one that broke the plan and won is the more dangerous entry in the record.
A review compares what you did against what you said you would do. Everything useful about it depends on the second half of that sentence existing before the trades did.
Before you start
A complete record of closed trades, including the ones you would rather skip. A record with holes reviews your memory rather than your method, and the holes are never random.
The plan or rules the trades were supposed to follow, written down beforehand. Without it there is nothing to compare against and the review becomes a discussion of results.
A fixed time in the week for it, because a review that needs deciding never happens. Sunday evening, Friday afternoon — the specific slot matters less than it being the same one every week.
The steps
1. Pull every closed trade from the period
Take them from the broker statement rather than from your notes. Any trade in the statement and not in your journal is the first finding of the review.
2. Mark each one as followed the plan or did not
A binary judgement against the written rules. Entry taken at the stated signal, stop placed where the rules say, size from the formula. Followed, or not.
3. Grade the outcome separately
Won or lost, in units of risk. You now have four categories, and the two that matter most are the rule-following loss and the rule-breaking win.
4. Count the rule breaks and write the number down
Four breaks out of eleven trades is a measurement. “I was undisciplined this week” is not, and it cannot be compared to next week.
5. Find the single most expensive pattern
Not a list of everything that went wrong. The one recurring behaviour that cost the most — moving stops, entering early, sizing up after a win.
6. Change one thing and write down what it is
One rule, one adjustment, applied for the next period. Changing three things at once means the next review cannot attribute any improvement to any of them.
7. Compare this period’s break count to the last one
The trend in rule breaks is the review’s real output. A falling count means the process is being followed more often, which is the only thing you directly control.
How to tell it worked
The trade count in your journal matches the broker statement, with 0 trades missing. That is the precondition for everything else.
Every trade has a process grade and an outcome grade, and they disagree on at least some of them. If they always agree, the process grade is being assigned from the result.
Exactly 1 change is being carried into the next period. More than one and nothing is attributable.
And the rule-break count sits next to the previous 4 weeks, so the direction is visible rather than remembered.
Why the split matters so much
A rule-breaking win teaches the wrong lesson and pays you for learning it. It is the entry most likely to be repeated, and the outcome grade rewards exactly the behaviour the process grade condemns.
On this site’s shared series, direction runs average 2.01 bars with a longest run of 11 across 286
runs. Short-run outcomes are noisy enough that grading only on results grades mostly noise. The
figures are in research/series-measurements.json.
Which is why the process count is the more stable signal. Rule breaks are directly observable and do not need a sample size to interpret; results need hundreds of trades before they mean anything.
What to record for each trade
Four fields do most of the work and none of them is the profit. The setup name, the rule that justified the entry, the stop level as placed, and the reason for the exit.
Add one line written at the time about what you expected. Written afterwards it becomes a description of what happened, which is useless — written before, it is the only record of what you actually believed.
Screenshot the chart at entry. A month later the reasoning is unrecoverable from prices alone, and the picture is what makes the review a comparison rather than a reconstruction.
And record the trades you decided not to take. A process that produces good passes is working as much as one that produces good entries, and nothing else in the record captures them.
The original data
Of the 24,971 unique videos in research/search-study-corpus.jsonl, 29 mention trading journals in
the title, at a median of 24,870 views across 27 channels — and 45% of those titles are
instruction-shaped. Trading psychology appears in 46 instruction-shaped titles at 14,632 and trading
plans in 23 titles at 10,042. The counts come from site/corpus_count.py and site/rank_howto.py.
29 videos at a 24,870 median. The audience per video is high and the coverage is thin, which is the signature this site finds on every subject that is operational rather than analytical.
The answer to the question on that chart is that a profitable week is the most important one to review. Rule breaks that were rewarded are the ones that get repeated, and a winning period is exactly when nobody looks. The review’s job is to find them while they are still cheap — which is before the same behaviour meets a market that does not reward it.
When it fails
The failure is a review that grades outcomes and calls it process, and it feels rigorous while doing it. Winning trades are marked as good, losing trades as mistakes, and the conclusions drawn are about the market rather than about behaviour. Over a few months this trains you to abandon rules after losses and to repeat whatever produced the last win — which is the opposite of what a review exists to do, arrived at through a process that looked disciplined throughout.
The second failure is reviewing only bad weeks. The rewarded breaks are in the good ones.
A third is changing several rules at once. Nothing is attributable afterwards.
A fourth is having no written plan to compare against. Then there is only the result.
A fifth is skipping trades in the journal. The gaps are systematically the informative ones.
And a sixth is reviewing without a schedule. A review that requires a decision does not survive a bad month.
Related
Trade review covers what the process is structurally and how it feeds the plan. Trading journal is the record it reads from. And trading plan is the document the trades are being graded against.
The grading split is the part that took longest to accept. A trade that followed every rule and lost money used to feel like a failure, and it is the entry I most want to see in the record — because it means the process ran and the result was the market’s. The rule-breaking winner is the one that quietly teaches the wrong lesson.
— Michael Whitman
This page is educational, not financial advice. Test every idea on your own charts before risking money.