Skip to content
You have a trading strategy. Perhaps you developed it yourself; perhaps you learned it in a course. The crucial question is: Does it work? And “work” does not mean “it worked on the last trade.” “Work” means: Does it have a statistical edge over hundreds or thousands of trades? That is exactly what backtesting answers.
Backtesting is the process of testing a trading strategy on historical data. You simulate how your system would have performed in the past without risking real money. In doing so, you collect data: win rate, average profit, average loss, maximum drawdown, and more. These numbers are your foundation. Without them, you are trading blind. In this article, I explain how to do backtesting properly, which mistakes you must avoid, and how to move from test results to live trading.
What is backtesting?
Backtesting means applying your trading strategy to historical market data and documenting every trade that would have occurred. You go chart by chart, day by day, and apply your rules exactly as you would in live trading.
In the end, you have a list of trades with entry, exit, profit or loss. From this list, you calculate statistical metrics that tell you whether your strategy is profitable in the long run or not.
Backtesting is not proof that a strategy will work in the future. Past performance never guarantees future results. But it is the best available way to validate a strategy before you risk real money. If a strategy does not even work in a backtest, it certainly will not work live.
Why backtesting is essential
Many traders skip backtesting. They have an idea, try it live for a few days, and when the first losses come, they discard it. Then they look for the next strategy. This pattern repeats endlessly.
Backtesting breaks this cycle. It gives you:
- Confidence: If you know your system was profitable over 500 historical trades, you can endure a losing streak without panicking
- Expectations: You know in advance how many losing trades in a row are statistically normal. No shock, no surprise
- Objective evaluation: No “I think this works.” Instead: “The data shows that it works”
- Optimization opportunities: You see which parameters of your system influence performance and can fine-tune them deliberately
- Risk management: You know the maximum drawdown and can adjust your position size accordingly
Without backtesting, you are relying on hope. With backtesting, you have statistics on your side.
Manual vs. automated backtesting
There are two basic approaches to backtesting. Both have their place.
Manual backtesting: You scroll through historical charts and apply your rules trade by trade. You document every entry, exit, stop-loss, and the result in a spreadsheet (Excel, Google Sheets, or specialized journal software).
Advantages of manual backtesting:
- You get to know your market extremely well
- You develop a feel for patterns and market behavior
- Discretionary elements can also be incorporated
- No programming effort required
Disadvantages:
- Time-consuming (50–100 trades can take days)
- Prone to unconscious bias (hindsight bias)
- Hard to reproduce
Automated backtesting: You code your rules into software that automatically applies them to historical data. This takes seconds instead of days.
Advantages:
- Fast and reproducible
- Objective (no hindsight bias)
- Can test thousands of trades over years
- Parameter optimization possible
Disadvantages:
- Requires programming skills or specialized software
- Only possible for fully rules-based systems
- Risk of overfitting (more on that in a moment)
My recommendation: Start with manual backtesting. It forces you to truly study your market. Once your system is clearly defined and you can automate it, switch to automated backtesting for larger datasets.
Tools for futures backtesting
Choosing the right tool depends on your approach:
| Tool | Type | Strength | Suitable for |
|---|---|---|---|
| NinjaTrader Strategy Analyzer | Automated | Deep integration with NinjaScript | Fully programmable strategies |
| NinjaTrader Market Replay | Manual/Semi | Tick-by-tick replay of historical data | Manual backtesting with order flow |
| ATAS Market Replay | Manual/Semi | Footprint and order flow in replay | Order-flow-based strategies |
| TradingView Bar Replay | Manual | Easy to use, large data pool | Fast visual testing of price action |
| Excel/Google Sheets | Documentation | Flexible, customizable | Trade journal and statistical analysis |
For futures traders in NQ and ES, NinjaTrader and ATAS are the best options. NinjaTrader offers the Strategy Analyzer for automated backtesting and Market Replay for manual testing. ATAS is particularly strong if you want to incorporate order flow data (footprint charts, delta) into your backtesting.
The most important metrics
After backtesting, you have a collection of trades. Now you need to ask the right questions. These metrics are crucial:
Win rate (hit rate): What percentage of your trades are winners? A 50% win rate is not bad as long as your winners are larger than your losers. Many successful systems have win rates between 40% and 60%.
Profit factor: Total profits divided by total losses. A profit factor above 1.0 means you make more than you lose. From 1.5, a system is considered solid. Above 2.0 is excellent. Below 1.3, it becomes difficult to remain profitable after commissions and slippage.
Expectancy: The average profit per trade when you add up all trades. Calculation: (win rate x average win) – (loss rate x average loss). Must be positive.
Max drawdown: The largest decline in your equity from a peak to a trough. If your system has gained $10,000 at its peak and then fell to $7,000, the max drawdown is $3,000 or 30%. This number is extremely important for your risk management and position sizing.
Average Winner / Average Loser (reward-to-risk): How large is your average win compared to your average loss? A 2:1 ratio means you make twice as much when you are right as you lose when you are wrong.
Sharpe ratio: Measures risk-adjusted return. The higher, the better. Above 1.0 is good; above 2.0 is very good.
Avoiding overfitting
Overfitting is the most dangerous mistake in backtesting. It happens when you keep adapting your system to historical data until it performs perfectly. The problem: It performs perfectly only on this specific data. In live trading, it collapses.
Imagine you test a strategy with 20 parameters (period settings, filters, time windows). If you optimize each parameter individually, you will eventually find a combination that looks fantastic on the historical data. But you have not found a profitable strategy. You have found a combination that happens to fit this data by chance.
How to avoid overfitting:
- Few parameters: The fewer parameters your system has, the more robust it is. Three to five adjustable values are enough for most systems
- Out-of-sample test: Split your data into two parts. Optimize on the first part (in-sample) and test on the second part (out-of-sample). If the system works on both, it is more robust
- Walk-forward analysis: Optimize on one time window, test on the next, move the window forward. Repeat this several times
- Logical rules: Every rule in your system should have a logical reason. “I buy when the 7-period RSI is below 23” is suspiciously specific. “I buy in oversold conditions” has a logical core
- Test different markets: A robust system works on NQ and ES. If it works only on a single instrument, it may be overfitted
Realistic expectations
Backtesting can lure you into a dangerous trap: unrealistic expectations. If your backtest shows an 80% win rate and a 5:1 reward-to-risk, you are either a genius or you made a mistake. Most of the time, it is the latter.
Common sources of error that make backtests look too good:
Ignoring slippage: In a backtest, you always get the exact price. In live trading, there is slippage, especially during fast moves. Factor in at least 1–2 ticks of slippage per trade.
Forgetting commissions: Every trade costs fees. With 200 trades per month, this adds up quickly. Include round-turn commissions in your backtest.
Hindsight bias: In manual backtesting, you tend to unconsciously know what happens next. You “obviously” see the winning trade and overlook the loser. What helps: use replay mode or hold a sheet of paper over the right side of the chart.
Survivorship bias: You test only the strategy version that “survived” in the end. The ten versions you discarded before do not count in your statistics. But they were part of your testing process.
Realistic expectations for a good futures system: 45–60% win rate, 1.3–2.5 profit factor, max drawdown of 15–25% of the account. Anything significantly above that should be questioned.
From backtest to paper trading to live
Backtesting is step one. But there are still important steps between a profitable backtest and profitable live trading:
Step 1: Backtesting (historical data)
Test at least 200–500 trades. Calculate all metrics. Make sure the result is profitable after slippage and commissions.
Step 2: Paper trading (real-time simulation)
Trade your system in real time on a demo account. At least 4–8 weeks, ideally 50–100 trades. This tests not only the system, but also you: Can you follow the rules when real emotions are involved?
Step 3: Micro live (minimal risk)
Trade with real money, but with the smallest possible position size. In the futures market, these are micro contracts (MNQ, MES). Real money feels fundamentally different from paper trading. This phase shows whether you can execute your system under psychological pressure.
Step 4: Scaling
If you are profitable after 50–100 live trades (micros) and your metrics match the backtest, you gradually increase position size. Not from one to ten contracts. From one to two. Then from two to three. Slowly.
This process takes time. Plan for 12–18 months from the first backtest to full live trading. That sounds long, but it protects your capital and builds real competence.
Backtesting in the TPTE Academy
In the TPTE Academy, backtesting is not an optional add-on. It is a core part of the curriculum. Every system you learn is backtested by you personally before you trade it live. You get the rules, the tools, the chart templates, and a structured guide for the entire process.
The reason is simple: You trust a system you have backtested yourself. And trust in your system is the foundation for getting through losing phases and being profitable in the long run.
If you want to see what professional backtesting looks like in practice and how you can apply it to your situation, book a free initial consultation. We will look together at where you are and what the next step that makes sense for you is.
