You have a trading strategy, you believe it works, and you are about to pay several hundred euros for a prop firm challenge to test it in real conditions? That is exactly the wrong order of priority. Most firms require a profit target between 8% and 10% in Phase 1 with drawdown limited to 5% or 10% depending on the package: any unvalidated strategy is eliminated within days. Backtesting is the step that separates traders who pass their challenges from those who fund the firms. Here is the complete method to test your strategy before risking a single euro.
Why backtest before a prop firm challenge
A challenge is not a testing ground. Between the profit target, the maximum drawdown and often a time limit, your trading rules simply cannot be approximate. Backtesting answers three essential questions before you pay anything: does your strategy win on average, over how many consecutive trades can it lose, and what maximum drawdown does it endure. These three numbers determine whether you stand a statistical chance of passing.
Take a concrete example. If your backtest reveals that your strategy regularly absorbs streaks of 6 consecutive losses with 1% risk per trade, your typical drawdown reaches 6%. On a phase allowing a 5% maximum drawdown, you have already failed after a single normal losing streak. That is a discovery you want to make on a spreadsheet, not on the live account.
Backtesting completes your trading plan and builds on the foundations described in our beginner's guide to technical analysis: it turns intuitions into measurable rules.
Manual or automated backtesting: which should you choose?
Manual backtesting
You scroll through the chart candle by candle and record every entry and exit signal in a spreadsheet. It is the most accessible method: free, no code required, and above all it forces you to know your strategy by heart. It is also the slowest: expect one to two hours for 100 trades going back in time. For a beginner, it is the best school, and it naturally prepares you for the trading journal you will have to keep afterwards.
Automated backtesting
If your strategy is coded (Pine Script on TradingView, MQL on MetaTrader, Python), the backtest runs in a few minutes over years of historical data. The trap is believing that the code does the work: without having personally verified the entry rules on a hundred concrete cases, automation faithfully reproduces anything.
The 5-step backtesting method
Step 1: write your rules down
Entry conditions, exit conditions, stop loss, take profit, position size, assets traded, trading hours. If a rule cannot be written down in black and white, it cannot be backtested. Remove any ambiguity such as "I aim for strong moves": a backtest only tests what is precise.
Step 2: choose your data
Work on the exact asset and timeframe you will trade during the challenge, ideally with 2 to 3 years of history to cover different market regimes: trend, range, high volatility. On forex, check that your historical data includes realistic spreads.
Step 3: execute trade after trade with no exceptions
Missed setups included, and days when you "would not have had time" included. A backtest in which you cherry-pick the trades that suit you measures nothing. Note for each trade: the date, the entry reason, the result in risk multiples (R) and the drawdown reached.
Step 4: calculate your metrics
Enter everything into a spreadsheet to obtain the win rate, the win/loss ratio, expectancy and the maximum drawdown, detailed below.
Step 5: compare with the format of your target challenge
Profit target, allowed drawdown, maximum duration: the prop firm comparison tool lists these criteria firm by firm. If your worst historical streak exceeds the allowed drawdown, change format or strategy before paying, not your money management mid-flight.
The metrics that really matter
Four numbers are enough to judge a strategy:
- The win rate: the share of winning trades. It is worth nothing on its own: a strategy with a 40% win rate and a 2.5 win/loss ratio is comfortably profitable, while a strategy with a 70% win rate and gains three times smaller than its losses is a losing one.
- The win/loss ratio: average gain divided by average loss. Above 1.5, your strategy can withstand a modest win rate.
- Expectancy: (win rate x average gain) minus (loss rate x average loss). Positive and stable from one half-year to the next, it is the only true judge.
- The maximum historical drawdown: the largest cumulative loss sustained in the past. This is THE number to confront with your firm's drawdown rules.
How many trades make a backtest reliable?
Below 50 trades, a backtest is a statistical anecdote. At 100 trades, the broad strokes (expectancy, typical drawdown) start to become credible. At 200 trades and more, you see the real losing streaks and the true distribution of gains. If your strategy produces few signals, that is a useful signal in itself: during a time-limited challenge, a 2-trades-per-week strategy would need several months to reach an 8% target, which some validation windows do not allow. The question of duration is covered in detail in our article on the time needed to pass a challenge.
The 5 mistakes that ruin a backtest
1. Overfitting
Tuning seven parameters until the curve is perfect means memorising the past's answers. Your backtest must remain profitable with simple, few rules.
2. Forgetting costs
Spread, commissions and slippage turn "profitable" backtests into losing live accounts. Systematically subtract the real cost of each trade.
3. Testing on a single market regime
A trend strategy tested only on a bullish period proves nothing. Include at minimum a range phase and an episode of high volatility.
4. Hindsight bias
When replaying the past, you know what comes next. Force yourself to take each signal only if your rule was verified at the close of the current candle, without looking at the next one.
5. Changing strategy mid-test
A backtest is only valid if the rules stayed identical from the first to the last trade. Otherwise, you are testing ten incomplete strategies.
Backtest, forward test, then challenge: the right sequence
A satisfying backtest is not enough: it must be confirmed by a forward test, in other words a few weeks of trading in real conditions with no commitment, which is exactly what a demo account allows. That is the subject of our guide on using a demo account before a challenge. The logical sequence is therefore: backtest over 100 trades minimum, a 30-day forward test on demo, then launch the challenge. And if you are starting from zero, our complete guide to learning trading covers the training steps that come before all of this.
FAQ: backtesting in 5 quick questions
Can you backtest with free data?
Yes for forex and indices: TradingView, MetaTrader and most brokers provide sufficient free historical data. For futures, quality data (tick by tick, for example via Rithmic or CQG) is often paid, but remains infinitely cheaper than a lost challenge.
Does a backtest guarantee passing the challenge?
No, and no tool guarantees it: markets do not repeat the past. Backtesting simply improves your odds by verifying that your risk profile is compatible with the firm's rules before you pay.
Should I backtest candle by candle or on close only?
It depends on your strategy: strategies on timeframes above H1 test very well on close; intraday scalping requires dropping down to the lower level to place stops and targets realistically.
What risk per trade should I test?
The one you will actually apply, i.e. 0.5% to 1%. Backtesting at 2% risk to "stress" the strategy distorts the final comparison with the firms' drawdown rules.
How much time should you spend on a backtest before the first challenge?
Expect 10 to 20 hours of work for a clean first manual backtest of 100 trades: enough to save several challenge attempts, whose cumulative cost quickly reaches a thousand euros.
Verdict
Backtesting is not about proving that your strategy is good: it is about measuring its risk profile before confronting it with a firm's rules. With a sample of 100 trades or more, a known maximum drawdown and a positive expectancy, you do not eliminate the risk of failure, but you stop paying for challenges to discover your own strategy. And if you prefer to focus on execution, our complete prop firm training program and our support services on our challenges include this upstream validation work.
