Backtesting

Backtesting lets you simulate how a strategy template would have performed on historical market data. This is the safest way to evaluate a strategy before risking real capital.

The backtester uses real historical candlestick data from the exchange and runs your strategy's components, combination rules, exit rules, and trading session windows using the same logic as the live bot — giving you a realistic simulation of past performance.

Stop loss and take profit fills account for price gaps (using the candle open when it gaps past the level). The simulation does not model slippage or partial fills, so results on less liquid instruments should be treated as best-case estimates.

To run a backtest, configure the following parameters:

• Strategy Template — Select any strategy template (system or user-created) to test. The backtest will use all of the template's enabled components, combination rules, and exit rules.
• Symbol — The trading instrument to simulate on (e.g. BTCUSDT). The available symbols come from the exchange's instrument list.
• Exchange — The exchange to source historical data from. Currently supported: BitMEX.
• Days Back — How many days of historical data to use. Default is 14 days. Longer periods give more data points but take longer to run.
• Initial Equity — The starting capital for the simulation. Default is $10,000. This affects position sizing calculations.
• Leverage — The leverage to use for simulated trades. Default is 3x.
• Fee Rate — The trading fee rate applied to each trade. Default is 0.05% (0.0005). Fees significantly impact net profitability, so use a realistic value for your exchange tier.

Click "Run Backtest" to start. The backtest runs through three phases:

1. Starting — Initialising the simulation engine.
2. Fetching historical data — Downloading the required candlestick data from the exchange for all intervals needed by the template's components.
3. Running backtest — Processing each candle through your strategy and recording simulated trades. A progress bar shows the percentage complete.

Progress updates automatically. You can navigate away and return later — the backtest continues in the background.

When a backtest completes, the results page shows:

The bottom of the Backtesting page shows a list of your recent backtest runs with their status, template name, symbol, date, and results summary. You can click any completed backtest to review its full results, or delete backtests you no longer need.

Backtest statuses:

• Pending — Queued for processing.
• Fetching data — Downloading historical candlestick data.
• Running — Simulation in progress.
• Completed — Finished successfully with results available.
• Failed — An error occurred during the backtest. The error message is shown for diagnosis.

Backtesting is a valuable tool for evaluating strategies, but simulated results differ from live trading in several important ways. Keep the following limitations in mind when interpreting results:

• No guarantee of future performance — A strategy that performed well on historical data may behave very differently in live markets. Market regimes shift, volatility changes, and correlations break down. Backtest results should never be treated as a prediction.
• Slippage — The backtester assumes fills at exact price levels (candle close for entries, SL/TP levels for exits). In reality, orders may fill at worse prices due to market movement between signal and execution, especially during fast-moving markets or high-impact events.
• Liquidity — The simulation assumes your orders are always fully filled immediately. On less liquid instruments or with larger position sizes, real orders may experience partial fills, requotes, or wider spreads that erode profitability.
• Stop loss & take profit execution — While the backtester accounts for price gaps (filling at the candle open when it gaps past a level), real stop orders on exchanges are executed as market orders and may fill at prices worse than the trigger level, particularly during volatile periods or thin order books.
• Fees — The simulation uses a flat fee rate for all trades. Real fee structures vary by exchange tier, maker/taker status, and volume discounts. The actual fees paid in live trading may differ.
• Market impact — The backtester assumes your trades do not move the market. In practice, larger orders can shift prices, widening the spread and reducing the effective entry/exit quality.
• Exchange & infrastructure risk — The simulation does not account for exchange downtime, API failures, network latency, or rate limits that may prevent orders from being placed or cancelled in time during live trading.