How to Compare Live Results
vs Backtest Results for a Gold EA
Published 17 June 2026 · 11 min read
The comparison involves eight key metrics: win rate, average win/loss in pips, profit factor, max drawdown, avg trade duration, trades per month, and spread cost per trade. Pull these from your MT5 History tab and compare against the backtest report. Each metric has an "acceptable gap" range — live win rate down 8% or less is normal; profit factor down 0.3 or less is expected. The comparison becomes meaningful after 3 months of live trading.
Live vs Backtest Comparison Table
Select the range that matches your live result for each metric. Status column updates automatically.
Step 1: Export Your Live Trade History from MT5
Before you can compare anything, you need your live metrics. MT5 makes this straightforward:
- 1
Open MetaTrader 5 and go to the bottom panel
- 2
Click the "History" tab (next to Trade, Exposure, etc.)
- 3
Right-click anywhere in the History panel
- 4
Select "Report" from the context menu
- 5
Choose the date range (your entire live trading period)
- 6
Export as HTML — this opens in your browser with full statistics
- 7
Find: Total Trades, Profit Trades %, Profit Factor, Max Drawdown, Avg Win/Loss
Step 2: Find the Backtest Metrics to Compare Against
If you ran the backtest yourself, the MT5 Strategy Tester produces a full report. If you are using a vendor-provided backtest, you need the same set of metrics — not just screenshots of the equity curve.
In the Strategy Tester report, look for the "Strategy Results" tab which shows: Total Net Profit, Profit Factor, Expected Payoff, Max Balance Drawdown, Max Equity Drawdown %, Total Trades, Short Trades, Long Trades, and the Profit Trades / Loss Trades breakdown with averages.
The key figure for win rate is "Profit Trades (% of total)" — this is the equivalent of win rate. For average pip wins and losses, use the "Average Profit" and "Average Loss" values and convert from the account currency to pips using the standard $1/pip calculation for XAUUSD at 0.01 lots.
Acceptable Deviation vs. Alarming Deviation
The single most common mistake traders make when comparing live to backtest is panicking at any deviation. Some gap is not just expected — it is guaranteed. The relevant question is whether the gap is within the normal range.
Normal — no action needed
- ✓Win rate: live is 3–8% below backtest
- ✓Profit factor: live is 0.1–0.3 below backtest
- ✓Max drawdown: live is 20–50% higher than backtest
- ✓Avg win: live is 10–20% smaller than backtest
- ✓Trade count: live is ±25% of backtest monthly average
Alarming — investigate before continuing
- ✗Win rate: live is 15%+ below backtest
- ✗Profit factor: live is below 1.0 (losing money)
- ✗Max drawdown: live is 2× higher than backtest
- ✗Avg win significantly smaller OR avg loss significantly larger
- ✗Trade count: live is less than 50% of expected (EA not firing)
For context on max drawdown specifically: Goldie Razor V2.8.4 is designed with the expectation that live drawdown will typically run 30–40% above the backtest figure. This is normal and is accounted for in the risk model. A backtest showing 14% max drawdown should be read as "expect up to approximately 18–20% in live conditions" — not as a guarantee of 14%.
Why Deviation Happens: The Main Causes
When your live results deviate significantly from the backtest, there are four primary causes to investigate in order of likelihood:
Spread difference
The backtest spread versus your live broker spread. If the backtest used 10 points (1 pip) and your broker charges 20 points (2 pips), every trade starts at a 1-pip disadvantage. This alone can drop win rate by 3–5% and profit factor by 0.1–0.2. Check: go to MT5 Symbols, find XAUUSD, and look at the average spread during peak session hours.
Slippage on fills
The backtest assumes fills at the exact signal price. Live execution always involves some slippage — typically 0.5–2 pips on entries and 0.5–3 pips on stop loss exits. During news events, slippage can be 5–10 pips or more. This increases effective loss sizes and reduces effective win sizes systematically.
Different market conditions
If the backtest covered 2022–2024 and your live trading starts in a different volatility regime (higher or lower), the EA's performance characteristics will differ. This is not a flaw — it is expected. Monitoring the metrics over 3+ months gives a more reliable picture than the first few weeks.
Incorrect EA settings
The most correctable cause. If your live EA settings differ from the recommended configuration — different session hours, different lot size calculation, news filter disabled — the live behavior will differ from what the backtest describes. Double-check every parameter against the setup documentation.
Related Reading
Why the Gap Exists and What Is Normal
The structural reasons every EA underperforms its backtest in live trading — and how to set realistic expectations.
Getting Accurate Backtest Metrics to Compare
How to run a backtest that produces reliable baseline metrics to compare against your live results.
The Metrics to Track Live
The 7 live profitability metrics to monitor and what each one tells you about EA performance.
Making Sure Your Backtest Baseline Is Honest
Before comparing live to backtest, ensure your backtest figures are not inflated by unrealistic assumptions.
When Live Results Fall Outside Acceptable Range
Practical steps to take when your comparison reveals significant deviation in the EA's live performance.
Frequently Asked Questions
At minimum 50–100 trades, and ideally 3 full months of operation. With fewer than 50 trades, the sample size is too small — a short streak of wins or losses can skew every metric significantly. Three months is better because it captures different market conditions (not just the specific conditions present in the first few weeks). If you are getting 7–10 trades per day, 3 months gives you a statistically meaningful 600–900 trade sample.
In MT5: go to the History tab in the terminal, right-click on any trade, and select 'Report'. This opens a dialog where you can export to HTML or XML format. For the most useful analysis, set the date range to match your live trading period, and export all closed trades. The report includes win/loss statistics, drawdown, and average trade metrics. You can also open the History tab, copy all rows (Ctrl+A, then Ctrl+C), and paste into Excel for custom analysis.
A drop from 1.65 to 1.3 is within the expected degradation range. For a profit factor of 1.65, you would expect live performance of around 1.35–1.45 — so 1.3 is slightly below the ideal range but not alarming. Continue monitoring over 3 months. If the profit factor continues declining (reaching 1.1 or below), that is when to investigate whether market conditions have changed fundamentally or whether there is an execution issue with your broker or settings.
A 15% drop in win rate is a significant deviation and warrants investigation before continuing. First, check whether the drop is consistent across all trade types or concentrated in specific conditions (high-spread periods, news days, specific sessions). Second, verify your broker spread against the spread used in the backtest. Third, check whether the EA settings are exactly as recommended — incorrect parameters can significantly affect entry logic. Fourth, consider whether the live period coincided with an unusual market event. If the drop persists across 3+ months with no identifiable cause, the EA may have stopped working.
Yes — almost universally. The backtest models historical tick data, but live execution adds real-world factors: spread widening during news, slippage on stop loss fills, momentary broker outages. Additionally, the live period may simply coincide with a more volatile market phase than the backtest average. A live max drawdown up to 50% higher than the backtest figure is within normal expectation. For example, if the backtest shows 14% max drawdown, a live max drawdown up to 21% is not cause for concern.
Goldie Razor V2.8.4
M15 breakout + H4 EMA filter — built for XAUUSD on MT5