For the whole of 2021 I kept a trading journal. Two hundred and fourteen trades, every entry filled in — entry, exit, pair, pips. And it changed absolutely nothing, because I never once read it back. My account drifted sideways for eleven months while I collected data like a librarian who never opens the books. The turn came in January 2022, when a mentor asked me one question: "What did your last twenty losers have in common?" I had no answer, despite having all the data. That weekend I built my first real post trade analysis review system — not a journal but a review machine with a weekly interrogation routine. Within ninety days my win rate on A+ setups went from 41% to 58%, and the year finished +34% instead of flat. The journal was the raw material. The review system was the refinery.
A post trade analysis review system is the difference between recording history and learning from it. This post trade analysis review system is the exact framework I run every Sunday, with the scoring math behind it, and the mistakes that nearly made me quit reviewing altogether.
Ninety days of reviews turned my red quarter green
Q1 2022 started badly: January finished -$3,200 on a $60,000 account. I started the review system on February 1st. The first review exposed something ugly — 68% of my losers were trades I had labeled "B setup, took it anyway." Not bad analysis. Bad discipline. By March I had a rule: B setups get half size or nothing, decided before the week starts. Here is the quarter in numbers.
| Month | Trades | A+ setup win rate | Net P&L |
|---|---|---|---|
| January (no review) | 31 | 41% | -$3,200 |
| February (review on) | 24 | 52% | +$1,850 |
| March (review on) | 22 | 58% | +$4,300 |
Notice the trade count falling: the review system did not make me smarter, it made me pickier. Twenty-two trades in March beat thirty-one in January because seven B-setup donations disappeared. Fewer, better trades — that is what a post trade analysis review system actually optimizes.
The anatomy of a review that actually changes behavior
Most review advice is "look at your losers and learn." Useless. A working post trade analysis review system has four mechanical parts: a graded trade log (every trade gets A+, A, B, or C at entry, before the outcome), a weekly aggregation that groups results by grade and by mistake tag, a single rule change per week (never more — I tried three at once in 2022 and followed none), and a monthly audit that checks whether the rule change moved the numbers. The mistake tags are the engine: "early entry," "widened stop," "news gamble," "revenge," "FOMO." After six weeks, my tag cloud showed "early entry" on 40% of losers. One tag, one fix — wait for the candle close — and the leak shrank by half.
| Review component | What it captures | Cadence |
|---|---|---|
| Graded trade log | Setup grade assigned before outcome known | Every trade |
| Mistake tags | Which discipline breach, if any, occurred | Every trade |
| Weekly aggregation | Win rate and P&L sliced by grade and tag | Sunday, 45 min |
| Rule change | Exactly one process fix for next week | Weekly |
| Monthly audit | Did the rule changes move the metrics? | First Sunday monthly |
The grade-before-outcome rule is the part most traders skip, and in any post trade analysis review system it is the part that matters most. Grading after you know the result just measures your ability to rationalize. My notes on confirmation bias in analysis were born in this exact trap — I kept "discovering" that my winners were all A+ setups.
How I reviewed trades for a year and learned nothing
My 2021 journaling failure deserves its own autopsy, because the mistakes are the ones I see in every trader's post trade analysis review system. Mistake one: I logged outcomes, not decisions — "lost 40 pips" tells you nothing about whether the entry followed the plan. Mistake two: no tags, so patterns were invisible; I sensed I was "trading badly" but could not name a single concrete leak. Mistake three: I reviewed monthly instead of weekly, by which time the emotional memory of each trade had evaporated and every loser looked "unavoidable." The killer mistake was number four: I changed five things at once after every bad month, so when results improved I never knew which change did it. A post trade analysis review system that changes everything at once changes nothing. One rule per week. That constraint felt painfully slow in February 2022 and looks like genius in hindsight.
Logging a trade into the signal calculator for review
The DarkWolf MT5 Signal Calculator is my pre-trade checkpoint, and it doubles as review evidence: every planned trade gets run through it first, and the screenshot goes into the log. At review time, I compare the calculator's planned risk against the actual realized risk — the gap between those two numbers is pure discipline measurement. A trade planned at $300 risk that actually lost $520 means I moved the stop or added size mid-trade, and the tag writes itself.
| Input Field | What it means | Sample value |
|---|---|---|
| Planned entry / stop | Levels from your pre-trade analysis | 1.0850 / 1.0815 |
| Planned risk ($) | Calculator output before the trade | $300 |
| Setup grade | Your honest grade, assigned pre-entry | A |
| Actual exit / P&L | What really happened, filled at review | 1.0890 / +$400 |
| Mistake tag | Discipline breach, if any | none |
This planned-vs-actual comparison is the highest-signal column in my entire post trade analysis review system — the heartbeat of the whole method. In 2023 it caught a slow drift I would never have felt: my average realized risk had crept from 1.0% to 1.35% over four months through tiny stop-widening. The account was still profitable, which is exactly why the drift was dangerous — profits were hiding a discipline leak.
The Sunday review ritual, step by step
Sunday, 6 PM, markets closed. Forty-five minutes, timer on — the weekly heartbeat of my post trade analysis review system. Step one (10 min): export the week's trades and fill planned-vs-actual for each. Step two (10 min): slice by grade — if B and C setups are more than 30% of volume, next week's rule writes itself. Step three (10 min): count mistake tags and find the top offender. Step four (10 min): write exactly one rule change for next week, phrased as a physical action ("I will set a phone alarm for candle closes" beats "I will be more patient"). Step five (5 min): check last week's rule — did I follow it? Score it 0–100% adherence. I keep a running adherence average, and it predicts my monthly P&L better than my win rate does. Weird but true: in months my adherence topped 85%, I was profitable eleven out of twelve times. Win rate alone predicted nothing.
Scoring your execution: the numbers behind the review
Feelings lie; the review system needs formulas. I score every week on three numbers: plan adherence (trades taken exactly per plan divided by total trades), risk discipline (planned dollar risk vs. actual), and grade efficiency (P&L per setup grade). The composite is simple enough to compute in a spreadsheet while the coffee brews.
| Variable | Meaning | Example value |
|---|---|---|
| Adherence % | Trades executed exactly per plan ÷ all trades | 80% |
| Risk Discipline % | 100 − (avg risk overshoot × 100) | 92% |
| A-grade Share % | A/A+ trades ÷ all trades | 65% |
| Execution Score | (80×0.5)+(92×0.3)+(65×0.2) | 80.6 |
My target is a score above 80. Below 70 for two weeks running, and I cut size in half until it recovers — the system policing the trader, not the market. Investopedia's trading journal guide covers the logging side well; what I add here is the scoring layer that turns logs into decisions.
Three reviewed trades, three calculator re-runs
Every quarter, as part of the post trade analysis review system, I pick three old trades and re-run them through the signal calculator with fresh eyes, asking: would my current system take this trade, at this size? The answers are humbling. These three are from my 2024 journal.
| Trade re-run | Calculator re-check | Review verdict |
|---|---|---|
| USDJPY long, Mar 2024, +$1,100 | Planned $400 risk, actual $390 | A grade, textbook — keep doing this |
| GBPUSD short, May 2024, -$620 | Planned $400 risk, actual $620 | Widened stop mid-trade; tag: widened stop |
| XAUUSD long, Jul 2024, +$900 | Planned $400 risk, actual $410 | Winner, but B grade — entry was early |
The gold trade is the interesting one: it made money and still got flagged. Any post trade analysis review system that only interrogates losers teaches you that winning excuses sloppiness. The B-grade winner got the same "early entry" tag as any loser, and my early-entry rate fell from 22% to 9% the next quarter. Also check costs while you review — my commission vs. spread breakdown lives next to my journal because a "profitable" scalping month once turned out to be -$180 after costs.
Post-trade review questions worth answering
How is a post trade analysis review system different from a trading journal?
A journal records; a post trade analysis review system interrogates. My 2021 journal had 214 logged trades and changed nothing because I never aggregated, tagged, or scored anything. The review system adds graded entries, mistake tags, weekly aggregation, one rule change per week, and a monthly audit. The journal is the database — the post trade analysis review system is the query that extracts decisions from it.
How much time does a proper weekly review take?
Forty-five minutes, timer on, every Sunday. Ten minutes on planned-vs-actual, ten on grade slicing, ten on mistake tags, ten writing one rule change, five scoring last week's rule. I tried two-hour deep reviews in 2022 and quit within a month — the ritual survived because it is short enough to never skip. Consistency beats depth here.
What should I do with winning trades during review?
Grade them exactly like losers. My July 2024 gold winner made $900 and still earned an "early entry" tag because the entry broke my rules. Winners that break rules are the most dangerous trades you take — they teach your brain that sloppiness pays. Tag them, count them, fix them like any other leak.
How many rule changes should I make per week?
Exactly one. I tried three simultaneous changes in 2022 and followed none of them; with one change I could attribute every improvement to a specific fix. Phrase it as a physical action, not an attitude — "set a phone alarm for candle closes" works, "be more patient" does not. Slow feels wrong and compounds beautifully.
What metrics actually predict my future performance?
In my data, plan adherence predicts monthly P&L better than win rate — months above 85% adherence were profitable eleven out of twelve times. Track adherence, risk discipline (planned vs. actual dollar risk), and the share of A-grade setups. Win rate and total pips are lagging noise by comparison; they tell you what happened, not whether your process is intact. If you trade under UK regulation, the FCA publishes useful material on understanding the risks you are actually running.