Run your first backtest

A backtest runs a strategy over past market data to show how it would have traded. It's the quickest way to learn how a strategy behaves before you spend weeks watching it on sim.

1. Open the Strategy Analyzer

In the Control Center, click New → Strategy Analyzer.

2. Set up the test

• Backtest type: Backtest.
• Strategy: pick the strategy you added in Part 3, then fill in its settings the same way you would on a chart.
• Data series: choose the instrument, bar type and timeframe you plan to trade.
• Start and end date: use a long enough period to include different kinds of markets. A few months is a reasonable first test; longer is better.
• Include commission: turn this on so results reflect real trading costs.
• Slippage: adds a few ticks of worse fills to market and stop orders. A tick or two is a sensible, slightly pessimistic start.

Then click Run.

3. Read the results

The summary shows a lot of numbers. Start with these:

• Total net profit: the bottom line, after commission if you turned it on.
• Number of trades: too few trades and the results are mostly luck.
• Max drawdown: the biggest drop from a high point. Ask yourself honestly whether you could sit through it.
• Profit factor: money won divided by money lost. Above 1 means it made more than it lost over this period.

4. Don't fool yourself

• Past results don't guarantee future ones. Markets change.
• Beware of over-tuning. If you keep adjusting settings until the backtest looks perfect, you've probably fitted the strategy to the past, not found an edge. Change one setting at a time, and prefer settings that work reasonably across a range of values.
• Test a period you didn't tune on. Pick your settings on one stretch of dates, then run them on a different stretch and see if the results hold up.

Next in Start here

A good backtest is a starting point, not a green light. Part 5 is the checklist to work through before trading a strategy with real money.