Before you risk real money on a trading system, you need evidence that the system has a positive expectancy. Not a feeling that it works. Not a few cherry-picked examples from the past. Systematic, objective evidence from a large sample of historical trades showing that the rules you have defined would have produced positive results over time. Backtesting provides that evidence. It is not a guarantee of future performance — markets change and no backtest predicts the future. But it is the only tool available for validating that your system has a genuine edge before risking capital on it.
What Backtesting Is
Backtesting is the systematic application of your trading rules to historical price data to see how the system would have performed. You apply your entry checklist to past charts, identify every trade the system would have generated, record the outcome of each, and then analyse the aggregate results. The goal is to determine whether your rules produce positive expectancy over a statistically meaningful sample.
Backtesting does not tell you that future trades will be profitable — it tells you that similar setups in similar conditions have historically been profitable. That is not a guarantee. But it is significantly better evidence than trading a system you have never tested.
Manual vs Automated Backtesting
Manual backtesting involves scrolling through historical charts, identifying setups according to your rules, recording each trade, and calculating the results by hand or in a spreadsheet. It is time-consuming but provides an invaluable benefit: deep familiarity with how your setups look across different market conditions. A trader who has manually backtested 200 trades of their system has seen the system fail in ranging markets, succeed in trending conditions, and every variation in between. That familiarity builds genuine pattern recognition.
Automated backtesting uses software — MT4 Strategy Tester, TradingView's Pine Script, or dedicated backtesting platforms — to apply rules automatically to historical data at high speed. It is faster and handles larger data sets but requires coding skills and is prone to curve fitting if the rules are not strictly defined in advance.
How to Backtest Manually
The manual backtesting process is structured and disciplined. Random chart-browsing looking for trades that worked is not backtesting — it is confirmation bias.
Step 1 — DEFINE RULES COMPLETELY Write your entry checklist, stop loss rule, and target rule before opening any charts. The rules must be complete — no mid-test adjustments. Step 2 — SELECT HISTORICAL DATA Choose a pair and a specific date range. Start at least 12 months ago. Move forward through the chart candle by candle — do not look ahead. Using a platform like TradingView: set the chart to your entry timeframe, set the date to your start point, scroll forward one candle at a time. Step 3 — APPLY YOUR CHECKLIST At each candle, run your entry checklist. Do all boxes check? No = move to next candle. Yes = record the trade setup. Step 4 — RECORD THE OUTCOME Apply your stop and target rules. Record: entry price, stop price, target price, outcome (win/loss/breakeven), R multiple. Step 5 — CONTINUE FOR 100+ TRADES 100 trades minimum. 200+ is better. Fewer than 100 produces statistically unreliable results.
What to Record
Date Pair Direction (Long/Short) Entry Price Stop Price Target Price Stop Distance (pips) Target Distance (pips) R:R Ratio Outcome (Win/Loss/Breakeven) R Multiple (+2R, -1R, +0R etc) Market Condition (Trend/Range/Transition) Notes (why setup was taken, what happened)
The market condition column is particularly important — it allows you to analyse whether your system performs differently in trending vs ranging conditions. If your backtest shows 65% win rate in trending markets and 30% in ranging markets, you know to add a trend filter to your entry criteria.
Interpreting Backtest Results
Four metrics matter most from a completed backtest.
WIN RATE: Total winning trades ÷ total trades. Context: is it above the minimum needed for profitability at your R:R? AVERAGE R:R: Average R multiple on winning trades. A 45% win rate at average 2.5R is a highly profitable system. MAXIMUM DRAWDOWN: Largest peak-to-trough loss during the test period. If max drawdown was 15 consecutive losing trades, ask: can I psychologically tolerate 15 losses in a row before it recovers? If not, adjust position sizing. PROFIT FACTOR: Total gross profit ÷ total gross loss. Above 1.5 = robust. Above 2.0 = excellent. Below 1.0 = unprofitable — do not trade live.
The most important question after backtesting is not "does this system make money?" — it is "can I follow this system through its worst drawdown period with my planned position size?" If your backtest shows a 20-trade losing streak at one point, that will happen again in live trading. If 1% risk means a 20% drawdown during that streak — can you continue following the rules? If the answer is no, reduce your risk percentage until the answer is yes.