Filling in a journal is the first half of the process. The second half — analysis — is where the actual improvement happens. Most traders who keep journals never get to this step. They accumulate entries, review individual trades reactively, but never extract the systematic patterns that are sitting in their data. This guide explains how to analyse 100+ trade entries to find those patterns.
When to Start Analysis
| Statistic | Formula | Healthy | What an unhealthy value means |
|---|---|---|---|
| Win rate | wins ÷ total trades × 100 | any value can be fine | On its own it says nothing: pair it with the average winner |
| Average winner | sum of positive R ÷ number of wins | above 1.0R | Below 1.0R means winners are being cut before the target |
| Average loser | sum of negative R ÷ number of losses | close to 1.0R | Well above 1.0R means stop-losses are being moved or ignored |
| Expectancy | (win rate × avg winner) − (loss rate × avg loser) | positive | Negative means no number of good trades fixes it — the system must change |
| Maximum drawdown in R | worst peak-to-trough run, counted in R | under 10R | At 1% risk, 10R is a 10% account drawdown; at 2% it is 20% |
Meaningful pattern analysis requires enough trades to produce statistically relevant signal rather than noise. The minimum thresholds:
- 20 trades: Can identify which pairs or setups are consistently losing — enough to stop trading them
- 50 trades: Can calculate reliable win rate, average R, and expectancy with some confidence
- 100 trades: Full statistical picture — time-of-day patterns, emotional state correlations, setup-type performance
Do not wait until 100 trades to start reviewing. Weekly reviews from the beginning build the habit. But treat the 50-trade and 100-trade marks as formal analysis checkpoints.
The Five Core Statistics to Calculate
| Statistic | Formula | Healthy | What an unhealthy value means |
|---|---|---|---|
| Win rate | wins ÷ total trades × 100 | any value can be fine | On its own it says nothing: pair it with the average winner |
| Average winner | sum of positive R ÷ number of wins | above 1.0R | Below 1.0R means winners are being cut before the target |
| Average loser | sum of negative R ÷ number of losses | close to 1.0R | Well above 1.0R means stop-losses are being moved or ignored |
| Expectancy | (win rate × avg winner) − (loss rate × avg loser) | positive | Negative means no number of good trades fixes it — the system must change |
| Maximum drawdown in R | worst peak-to-trough run, counted in R | under 10R | At 1% risk, 10R is a 10% account drawdown; at 2% it is 20% |
1. Win Rate
Wins ÷ Total trades × 100. Calculate separately for different categories: by pair, by session, by setup type, by emotional state before entry.
2. Average Winner (in R multiples)
Sum of all positive R multiples ÷ number of wins. This should exceed 1.0R for any viable system at normal win rates.
3. Average Loser (in R multiples)
Sum of all negative R multiples ÷ number of losses. Should be close to 1.0R if stop-losses are being respected. Significantly above 1.0R indicates stop-loss moving.
4. Expectancy
(Win rate × average winner R) − (loss rate × average loser R). This is the single most important number in your analysis.
Worked example: 48% win rate, average winner 2.1R, average loser 1.1R:
= (0.48 × 2.1) − (0.52 × 1.1) = 1.008 − 0.572 = +0.436R per trade
At R500 risk per trade (1% of R50,000), this is R218 average profit per trade. Over 50 trades per month, that is R10,900/month — a 21.8% monthly return. These numbers are illustrative but show how positive expectancy compounds.
5. Maximum Drawdown in R
The worst peak-to-trough sequence in R terms. If your worst losing streak was 8 consecutive losses, your drawdown was 8R. At 1% risk per trade, that is 8% account drawdown — recoverable. At 3% risk, it is 24% — a serious problem.
Google Sheets Analysis Template
The following columns capture everything needed for systematic analysis:
| Column | Description |
|---|---|
| Date | YYYY-MM-DD |
| Time | Entry time (SAST) |
| Pair | USD/ZAR, EUR/ZAR, etc. |
| Direction | Long / Short |
| Entry | Actual entry price |
| Stop | Stop-loss price |
| Target | Take-profit price |
| Exit | Actual exit price |
| Size (lots) | Position size |
| R at risk (ZAR) | Actual ZAR amount risked |
| Outcome R | Result in R multiples (+1.5R, −1R, etc.) |
| Outcome ZAR | Actual ZAR P&L |
| Emotion (1-10) | Pre-trade emotional state |
| Setup type | Your named setup (e.g., "EMA bounce", "breakout") |
| Rules followed | Yes/No |
| Lesson | One sentence |
With these columns, the following analyses become one-click pivot tables:
- Win rate by pair
- Average R by time of day
- Performance by emotional state rating
- Rules-followed vs rules-violated comparison
- Best and worst setup types
The Three Most Revealing Analyses
| Win rate | Winner 1.0R | Winner 1.5R | Winner 2.1R | Winner 3.0R |
|---|---|---|---|---|
| 35% | −0.365R · −R9 125 | −0.190R · −R4 750 | +0.020R · +R500 | +0.335R · +R8 375 |
| 40% | −0.260R · −R6 500 | −0.060R · −R1 500 | +0.180R · +R4 500 | +0.540R · +R13 500 |
| 45% | −0.155R · −R3 875 | +0.070R · +R1 750 | +0.340R · +R8 500 | +0.745R · +R18 625 |
| 48% | −0.092R · −R2 300 | +0.148R · +R3 700 | +0.436R · +R10 900 | +0.868R · +R21 700 |
| 50% | −0.050R · −R1 250 | +0.200R · +R5 000 | +0.500R · +R12 500 | +0.950R · +R23 750 |
| 55% | +0.055R · +R1 375 | +0.330R · +R8 250 | +0.660R · +R16 500 | +1.155R · +R28 875 |
| 60% | +0.160R · +R4 000 | +0.460R · +R11 500 | +0.820R · +R20 500 | +1.360R · +R34 000 |
Analysis 1: Emotional state vs outcome
Group your trades by emotional state rating (1–3, 4–6, 7–8, 9–10) and calculate win rate and average expectancy for each group. The most common finding: trades entered at emotional state 4–7 (steady, neutral) outperform trades entered at 8–10 (overconfident or desperate) by a margin that surprises most traders.
Analysis 2: Rules-followed vs rules-violated
Calculate expectancy separately for trades where you followed your rules and trades where you deviated. In virtually every case, rule-following trades are more profitable than rule-violations — often by a ratio of 2:1 or better. This analysis, more than any other, makes the case for executing a process rather than improvising.
Analysis 3: Setup type performance
Which of your named setups actually has positive expectancy? Most traders are surprised to find that 2–3 setups carry their performance while 3–4 others drag it down. The response is to stop trading the losing setups — not to try to improve them immediately, but to concentrate on what is already working.
Red Flags That Require Immediate Action
| What the journal shows | What it indicates | Action |
|---|---|---|
| Average loser above 1.3R | stop-losses are being moved after entry | Stop trading until you have ten entries in a row where the stop held |
| Win rate high, expectancy negative | winners are cut, losers are held | Set the target before entry and do not touch it |
| Worst results cluster on one emotional word | state, not strategy, is the variable | Add a rule that bars entry in that state |
| Rationale field often empty | trades are being taken without a thesis | No rationale, no trade — write it first or skip the setup |
| Drawdown beyond 10R | size is too large for the system's natural streaks | Halve the risk per trade and keep the same rules |
Stop live trading and investigate if any of these appear in your analysis:
- Average loser significantly above 1.0R (e.g., 1.8R): You are moving stop-losses after entry or holding trades past planned stops. This pattern is more dangerous than a low win rate.
- Win rate above 60% but negative expectancy: Your average winner is much smaller than your average loser. Classic disposition effect — fix the exit process before worrying about entries.
- Consistent underperformance at specific times: If trades between 22:00 and 02:00 SAST consistently lose while other times are profitable, the solution is simple — stop trading those hours.
- Rules-violated trades showing positive results: This is the most dangerous finding. If rule violations are being reinforced by occasional wins, the pattern will continue until a catastrophic loss eventually arrives.
Moving from Analysis to Improvement
The analysis reveals problems. Fixing them requires choosing one item and addressing it specifically:
Week 1–4: Fix the one biggest issue identified (usually: stop-loss placement or not moving stops). Nothing else changes.
Week 5–8: Assess whether the fix worked. Calculate the same statistics over the new 20-trade sample. Move to the next issue only if the first one is resolved.
Trying to fix five things simultaneously produces regression to old habits under pressure. One specific, observable behaviour at a time is how the data translates into lasting improvement.
This is general information only, not financial advice. Trading forex and CFDs carries a high level of risk and losses can exceed your initial deposit.
Apply what you've learned with an FSCA-regulated ZAR account. Open FxPro account →
