📈Analyzing Results
Understanding Your Backtest & Optimization Data
After running backtests and optimizations, you have mountains of data. This guide helps you interpret the numbers, identify what matters, and make informed decisions about your trading strategy.
Goal: Transform raw backtest data into actionable insights for strategy development and live trading decisions.
Key Performance Metrics Explained
Net Profit
Definition: Total profit/loss after all trades closed
Formula: Sum of all winning trades - Sum of all losing trades - Costs
What It Tells You:
Bottom line: Did EA make or lose money?
Absolute profitability
Interpretation:
Positive: Strategy profitable on tested data
Negative: Strategy lost money (not viable)
Zero/Near-Zero: Strategy barely breakeven (costs eating profits)
Limitations:
Doesn't account for risk taken
Doesn't show consistency
Ignores drawdown endured
Example:
Pro Tip: Net profit alone is misleading. Always consider it with drawdown and profit factor.
Profit Factor
Definition: Ratio of gross profit to gross loss
Formula: Gross Profit ÷ Gross Loss
Example Calculation:
Interpretation:
< 1.0
🛑 Losing
Strategy loses money overall
1.0 - 1.3
⚠️ Marginal
Barely profitable, risky
1.3 - 1.5
🟡 Acceptable
Okay but not great
1.5 - 2.0
✅ Good
Solid strategy
2.0 - 2.5
✅✅ Very Good
Strong performance
> 2.5
🤔 Suspicious?
Excellent OR over-fitted
BananaEA Target: Profit Factor > 1.5 (preferably 1.8-2.2)
Why It Matters:
Shows efficiency (how much you make per € risked)
More stable than raw profit
Less sensitive to single lucky/unlucky trades
Recommendation: Prefer Profit Factor 1.8 with €10K profit over Profit Factor 1.2 with €15K profit. Better efficiency = more robust.
Maximum Drawdown
Definition: Largest peak-to-trough decline in account balance
Types:
Absolute Drawdown: Drop from starting balance
Example: Start €10,000 → Drop to €8,500 = €1,500 absolute DD (15%)
Maximal Drawdown: Largest drop from any peak
Example: Peak €15,000 → Drop to €12,000 = €3,000 maximal DD (20%)
Relative Drawdown: Max DD as % of peak balance
Most commonly reported metric
Interpretation:
< 10%
✅✅ Excellent
Very safe, conservative
10% - 20%
✅ Good
Acceptable for most traders
20% - 30%
⚠️ High
Requires strong psychology
30% - 50%
🛑 Very High
Dangerous, stressful
> 50%
💀 Extreme
Unacceptable for live trading
BananaEA Target: Max Drawdown < 20% (ideally < 15%)
Why It Matters:
Shows worst-case loss scenario
Tests psychological tolerance
Indicates risk management quality
Predicts account survival probability
Drawdown Recovery:
Win Rate (% Winners)
Definition: Percentage of trades that closed with profit
Formula: (Number of Winning Trades ÷ Total Trades) × 100%
Example:
Interpretation:
< 30%
🛑 Very Low
Losing too often
30% - 40%
⚠️ Low
Needs large winners
40% - 60%
✅ Balanced
Healthy range
60% - 70%
✅ Good
Strong performance
70% - 80%
🤔 High
Check if realistic
> 80%
🚩 Suspicious
Likely over-fitted
BananaEA Target: Win Rate 45-65% (balanced)
Important Context:
High win rate ≠ Profitable: Can have many small wins, few large losses
Low win rate CAN be profitable: Few wins, but winners are MUCH larger than losers
Strategy Styles:
Never judge strategy by win rate alone! A 40% win rate with 3:1 reward-to-risk is MUCH better than 60% win rate with 1:2 reward-to-risk.
Average Trade
Definition: Average profit/loss per trade
Formula: Total Net Profit ÷ Total Number of Trades
Example:
Interpretation:
Negative
🛑 Losing
Strategy unprofitable
€0 - €20
⚠️ Low
Barely profitable, costs eat gains
€20 - €50
🟡 Acceptable
Okay for high-frequency
€50 - €100
✅ Good
Solid performance
> €100
✅✅ Excellent
Strong per-trade profit
Considerations:
Account Size Matters: €50/trade on €10K account (0.5%) is different from €50/trade on €100K (0.05%)
Trade Frequency: Low avg trade with many trades can be fine
Consistency: Check if average is consistent or skewed by few outliers
Expected Payoff
Definition: Mathematical expectation of profit per trade
Formula: (Win Rate × Avg Win) - (Loss Rate × Avg Loss)
Example:
Interpretation:
Positive: Strategy has positive expectancy (profitable long-term)
Negative: Strategy has negative expectancy (losing long-term)
Higher is Better: More expected profit per trade
Use Case:
Comparing different parameter sets
Assessing strategy viability
Calculating position sizing for Kelly Criterion
Sharpe Ratio (Advanced)
Definition: Risk-adjusted return (return per unit of volatility)
Formula: (Average Return - Risk-Free Rate) ÷ Standard Deviation of Returns
Interpretation:
< 0
🛑 Losing
Underperforms risk-free investment
0 - 0.5
⚠️ Poor
Barely beats risk-free rate
0.5 - 1.0
🟡 Acceptable
Adequate risk-adjusted return
1.0 - 2.0
✅ Good
Strong risk-adjusted performance
> 2.0
✅✅ Excellent
Outstanding risk-adjusted returns
BananaEA Target: Sharpe Ratio > 1.0 (ideally > 1.5)
Why It Matters:
Accounts for volatility (not just profit)
Compares strategies with different risk profiles
Preferred metric by institutional traders
Note: MT4 doesn't calculate Sharpe Ratio automatically. You'll need to export trade data and calculate in Excel or use specialized analysis tools.
Analyzing Equity Curves
What is an Equity Curve?
Equity Curve: Graph showing account balance + floating P/L over time
Components:
Balance Line (Grey): Closed trade balance only
Equity Line (Green/Red): Balance + floating P/L
Healthy Equity Curve Characteristics
1. Upward Trend ✅
2. Moderate Drawdowns ✅
3. Consistent Recovery ✅
Unhealthy Equity Curve Patterns
1. Vertical Spike 🚩
2. Deep Hole 🚩
3. Staircase Down 🚩
4. Flat Line ⚠️
5. Erratic Swings 🚩
Trade Distribution Analysis
Consecutive Wins/Losses
Maximum Consecutive Wins:
Shows longest winning streak
High number (10+) = system on a roll
Check if realistic or luck
Maximum Consecutive Losses:
Shows longest losing streak
CRITICAL for psychology: Can you handle 8-10 losses in a row?
High number (15+) = potential strategy problem
Example:
Reality Check: If backtest shows 12 consecutive losses, expect 15-20 in live trading. If you can't handle that psychologically, don't trade the strategy!
Largest Win vs. Largest Loss
Analysis:
Red Flags:
Largest Win 10x Average Win: One lucky trade, not repeatable
Largest Loss 10x Average Loss: Indicates risk management failure
Largest Loss > 3% Account: Single trade can seriously damage account
Comparing Optimization Results
Multi-Parameter Comparison
When comparing optimization passes, create comparison table:
Ranking Criteria:
A+: PF > 2.3, DD < 12%, WR 55-65%, Trades 150+
A: PF > 2.0, DD < 15%, WR 50-70%, Trades 100+
B: PF > 1.7, DD < 20%, WR 45-75%, Trades 75+
C: PF > 1.4, DD < 25%, Trades 50+
F: PF < 1.4 or DD > 25%
Visual Comparison
Create Scatter Plots (Excel/Google Sheets):
Profit vs. Drawdown:
Profit Factor vs. Total Trades:
Statistical Significance
Sample Size Matters
Minimum Trade Requirement:
50+ trades: Bare minimum for statistics
100+ trades: Acceptable sample
200+ trades: Good confidence
500+ trades: High confidence
Example:
Never trust backtests with <30 trades! Sample size too small for statistical validity. Extend test period or reduce risk to generate more trades.
Red Flags in Backtest Results
Indicators of Over-Optimization
🚩 100% Win Rate - Impossible in real trading 🚩 Zero Drawdown - Look-ahead bias or no losing trades 🚩 Single Huge Winner - One trade = 80% of profit 🚩 Perfect Equity Curve - Smooth line up, no variance 🚩 Too-Good Metrics - PF > 5.0, Sharpe > 4.0 🚩 Unstable Across Periods - Q1: +€20K, Q2: -€15K, Q3: +€25K 🚩 Parameter Sensitivity - Tiny change (BuyBuf: 7.0 → 7.1) = huge performance difference
Indicators of Robust Strategy
✅ Consistent Performance - Similar results across years/quarters ✅ Moderate Metrics - PF 1.5-2.5, DD 10-20%, WR 45-65% ✅ Gradual Equity Curve - Smooth upward trend ✅ Parameter Stability - Small parameter changes = small performance changes ✅ Trade Diversity - Trades spread across different times/conditions ✅ Sufficient Sample - 100+ trades minimum ✅ Out-of-Sample Validation - Performs well on unseen data
Practical Decision Framework
Is This Strategy Worth Trading?
Minimum Requirements:
Bonus Points:
✅ Sharpe Ratio > 1.0
✅ Expected Payoff > €30/trade (on €10K account)
✅ Recovery Factor > 2.0 (Net Profit ÷ Max Drawdown)
✅ Works on multiple symbols
✅ Consistent across timeframes
Next Steps
Now that you can analyze results:
Validate Your Findings → Validation & Forward Testing
Learn Advanced Techniques → Advanced Optimization
Follow Professional Workflow → Best Practices & Tips
Pro Tip: Create an "Analysis Checklist" document. For every backtest/optimization, fill out the checklist with all key metrics. Over time, you'll build intuition for what "good" looks like for your trading style!
Related Resources
Running Optimizations - How to optimize parameters
Optimization Fundamentals - Theory and concepts
Best Practices & Tips - Professional workflow
Validation & Forward Testing - Confirming results
Last updated