To review a month of trades, first reconcile your trade log against the broker statement, then measure the month in R rather than dollars, split results by setup and by whether you followed your rules, check whether the sample is big enough to mean anything, and finish with one written change for next month. Plan on about an hour.
A monthly review is where a trading journal earns its keep. Logging records what happened; the review decides what, if anything, to do about it. Done badly, it becomes a monthly excuse to rewrite a strategy after a bad run. Done well, it produces one small, testable adjustment and a clear list of what to leave alone.
Why review monthly at all?
A day is too little data and a year is too long to wait. A month gives you roughly 21 trading sessions — enough for patterns in your behaviour to show, and a natural moment to do the admin that otherwise slides: downloading statements, filing receipts, checking fees. Treat it as a fixed appointment, not something you do after a painful week.
Step 1: Reconcile before you analyse
Check that your log matches reality. Compare the number of trades, the net result and the total commissions and fees against the broker statement. Missing trades are usually the ones you did not want to log, which makes them the most informative. If the two do not match, fix the log first; every number after this depends on it.
Step 2: Measure the month in R, not dollars
R is the amount you planned to risk on a trade. Converting every result into multiples of R strips out position size and account growth, so you are measuring decisions instead of deposits. Record five numbers:
- Net R for the month.
- Number of trades.
- Average win and average loss, in R.
- Expectancy per trade, in R: average win times win rate, minus average loss times loss rate.
- Largest single loss, in R. Anything worse than −1R means a stop was moved, skipped or slipped badly, and it deserves its own note.
Step 3: Split the month three ways
Totals hide everything useful. Sort the same trades by setup, by time of day, and by whether the trade followed your written rules. The third split is usually the most revealing. An illustrative example:
| Group | Trades | Net R |
|---|---|---|
| Followed the plan | 24 | +6.5R |
| Broke a rule | 9 | −7.0R |
| Month | 33 | −0.5R |
That month looks like a strategy problem and is actually a discipline problem. The fix is not a new setup; it is fewer of the nine trades that broke the rules.
Step 4: Grade the process separately from the outcome
Every trade falls into one of four boxes: good process and a win, good process and a loss, bad process and a win, bad process and a loss. Good-process losses are the cost of doing business and need no fixing. Bad-process wins are the dangerous ones, because they reward the habit you are trying to break. Count how many of each you had; the aim over time is more of the first two and fewer of the last two, regardless of the P&L.
Step 5: Check whether the sample means anything
This is the step most reviews skip, and it is why traders abandon working setups after one bad month. Twenty trades is a small sample, and small samples are noisy.
A standard way to measure that noise is a confidence interval for a proportion. Using the Wilson method described in the NIST/SEMATECH e-Handbook of Statistical Methods, a 60% win rate over 20 trades is consistent, at 95% confidence, with a true win rate anywhere from about 39% to 78%. Over 100 trades, the same 60% narrows to roughly 50% to 69%.
In practice: one month rarely proves a setup works, and rarely proves it has stopped working. Judge setups on a rolling run of trades rather than a calendar month, and read the month mainly for behaviour, which does show up quickly. When the evidence against a setup does build up over enough trades, our guide on when to retire a trading setup covers the decision.
Step 6: Study the three best and three worst trades
Pull up the chart for each. For the winners, ask whether they were planned or lucky. For the losers, ask whether the loss was the plan working as designed — a clean stop at −1R — or a decision you would not repeat. Write one sentence per trade. Six sentences are enough; reviewing all 30 charts in detail turns into a search for patterns in noise.
Step 7: Write one change and a keep list
Finish with a short note of two parts:
- One change for next month, specific enough to test. “No entries in the first five minutes” can be checked; “be more patient” cannot.
- A keep list of what is working and must not be touched, so a bad week does not talk you out of it.
One change at a time is what makes next month’s review readable. Change five things and you cannot tell which one helped. Then file the month’s statements and receipts while they are open — the admin half of the review takes ten minutes.
Frequently Asked Questions
How long should a monthly trading review take?
About an hour for most part-time traders. Reconciling the log takes ten to fifteen minutes, the numbers and splits another twenty, and studying the three best and three worst trades plus writing the change the rest. If it regularly takes much longer, you are probably reviewing every chart instead of the ones that matter.
What should I do after a losing month?
Separate process from outcome before changing anything. If most of the losses came from trades that broke your rules, the fix is discipline, not strategy. If you followed the plan and still lost, check the sample size: a month of trades is usually too few to prove a setup has stopped working.
How many trades do I need before a review means anything?
For behaviour, such as rule-breaking or trading at bad times of day, even 20 trades can show a clear pattern. For judging whether a setup has an edge, you need far more: a 60% win rate over 20 trades is statistically consistent with anything from about 39% to 78%, so treat monthly results as a hint and judge setups over a longer run.
Should I change my strategy after a bad month?
Usually not on the strength of one month alone. Make at most one specific, testable change, keep everything that is working on a written keep list, and judge the setup over enough trades to be meaningful. Changing several rules at once makes it impossible to tell next month which change mattered.
Bottom line
A good monthly review reconciles first, measures in R, splits results by setup, time and rule-following, grades process apart from outcome, and respects the sample size — a 60% win rate over 20 trades could honestly be anything from 39% to 78%. It ends with one testable change and a keep list, not a new strategy. Built into a routine, it is one of the habits that separates a trader who improves from one who just trades more; it sits with the other foundations in how to start day trading, and our FAQ explains how the room fits alongside your own review.