Every strategy works on paper. The question is whether it would have worked in real market conditions. Backtesting answers that question by running a strategy against past price data, so you can see how it might have performed before you commit real funds.

This guide explains what backtesting is, the common pitfalls to avoid, and how Traders Gym on ThinkTrader lets you backtest strategies on historical market data.

What is backtesting

Backtesting is the process of applying a trading strategy to historical price data to assess how it would have performed. You set the rules, the time period, and the instruments, then the system replays the market and records the outcomes.

A useful backtest tells you several things: how often the strategy would have triggered, what proportion of trades would have closed in profit, the average size of profitable trades compared to unprofitable ones, and the maximum drawdown the strategy would have produced, which is the largest peak-to-trough decline in account equity

Backtesting does not guarantee future results. Markets change, volatility regimes shift, and a strategy that performed well in one environment may underperform in another. What backtesting does is give you a structured way to test ideas before risking capital.

Common backtesting biases

One of the most common mistakes traders make with backtesting is treating it as proof rather than evidence. Several biases can make a backtest look stronger than the strategy actually is.

Look-ahead bias — Your test uses information that would not have existed at the time of the trade. Using a closing price to trigger a decision that should have been made mid-session is a classic example, and it silently inflates every result it touches.
Survivorship bias — The data only shows you what survived. Assets that were delisted, currencies that were re-pegged, and positions that collapsed entirely never make it into the dataset, which means your backtest is working from a tidier version of history than reality ever offered.
Overfitting — The strategy has been tuned so precisely to past data that it is picking up noise rather than pattern. It can look exceptional in backtest and collapse within days of going live, because the signals it relied on were never really there.
Curve-fitting through optimisation — Testing fifty variations of a strategy and picking the best performer feels like research. In practice you are likely crowning whichever version got lucky with historical noise, not the one with genuine edge.

To reduce these issues, hold out part of your data as an out-of-sample test set. Run the strategy on data the optimisation never touched. If it still performs, you have more confidence the result is meaningful.

How to design a useful backtest

Start with a clear hypothesis — A backtest works best when you have a specific question to answer. “Does a 20-period moving average cross with a confirming volume signal produce positive expectancy on EURUSD across 2020 to 2024?” is testable. “Is moving average crossover any good?” is not.

Choose a meaningful sample size — A handful of trades tells you nothing useful. Strategies with infrequent triggers need years of data before the results mean anything, and even then a small sample should make you cautious.

Account for trading costs — Spreads, commissions, swap rates and slippage all erode theoretical returns. A backtest that ignores these costs is not testing a strategy, it is testing a fantasy.

Define your risk management rules upfront — Where do you place stops? How do you size positions? How do you exit a profitable trade? Without answers to these questions baked into the test, you are not evaluating a strategy at all. You are evaluating an entry signal with no plan attached.

Traders Gym on ThinkTrader

Traders Gym is the backtesting environment built into ThinkTrader, available on both web and mobile. Awarded for Innovation by ForexBrokers.com in 2024, it is one of the few backtesting tools built directly into a trading platform rather than bolted on as an afterthought.

Run historical replays at x1, x5, x10, x50, x100 or x200 speed and compress months of price action into a single session. Switch between instruments, apply your own indicators and drawing tools, and place simulated orders that behave exactly as they would on a live account.

To get started, select an instrument, choose a date range, set your replay speed and begin. Pause at any point to study a setup, place a simulated order, then pick up where you left off.

Because Traders Gym uses the same chart engine and order ticket as your live account, the habits and workflows you build in the gym carry over directly when you trade for real.

What to do with backtest results

A backtest is the start of a process, not the end. If a strategy performs well in backtest, the next step is forward testing on a demo account using live market data. This surfaces issues a backtest cannot replicate: execution timing, real-time data feeds, and your own discipline under pressure.

Only once a strategy holds up in forward testing should you consider trading it live, and even then start with smaller position sizes than your backtest assumed.

Backtesting is a tool for building confidence in a process. It is not a substitute for sound risk management or for the experience of trading in real conditions.

FAQ

Which devices can I use Traders Gym on?

Traders Gym is available on ThinkTrader Web and the ThinkTrader mobile app on iOS and Android.

Can I use Traders Gym on all ThinkMarkets’ trading platforms?

Traders Gym is an exclusive tool that is only accessible on our ThinkTrader platform.

Can I use Traders Gym on a demo account?

Traders Gym is only accessible to live account holders. With a live account, you can access Traders Gym in both demo and live mode. If you only have a demo account and wish to use Traders Gym, you’ll need to create a live trading account.

Is Traders Gym free to use?

Yes, Traders Gym is free for clients with live trading accounts.

ThinkTrader Articles