How to Backtest a Crypto Trading Strategy (Before Risking Money)

Backtesting means running a strategy on historical data to see how it would have performed. Done right, it saves you from paying real money to learn a strategy doesn't work. Done wrong, it fools you into trusting one that doesn't. Here's how to do it right.

Read the drawdown and check for curve-fitting, not just the profit. A survivable strategy on realistic data beats a perfect-looking one tuned to the past.

On this page

Why backtest
Use quality data
Metrics that matter
Avoid curve-fitting
Forward-test too
FAQ

Why Backtesting Matters

A backtest answers a cheap question before an expensive one: would this strategy have survived real market history? It won't predict the future, but it will expose an obviously broken idea before you fund it.

It's also how you separate a bot with a real edge from one that only looks good in a screenshot.

Use Realistic, Quality Data

Test over several years, including volatile periods and news-heavy months.
Use high-quality data (ideally tick or accurate OHLC), not a cherry-picked window.
Include fees and spread — a strategy that's profitable without costs may not be with them.

Let the bots trade while you don't

Prometheus runs a managed, risk-controlled strategy 24/7 on your own exchange — non-custodial, with a 90-day money-back guarantee.

Get Lifetime Access →

The Metrics That Actually Matter

Max drawdown — the worst peak-to-trough drop. Can you survive it emotionally and financially?
Expectancy — average profit per trade. Positive is the whole game.
Profit factor — gross profit ÷ gross loss.
Number of trades — enough for the result to be meaningful, not luck.

Notice win rate isn't at the top — a 90% win rate with huge losers is a losing system.

Avoid Curve-Fitting (The Big Trap)

Curve-fitting is tuning a strategy so tightly to past data that it's perfect on history and useless live. Warning signs: a flawless equity curve, dozens of finely-tuned parameters, and results that fall apart on out-of-sample data.

Test on data the strategy wasn't tuned on. If it holds up, that's a good sign; if it collapses, it was curve-fit.

Then Forward-Test on Demo

A backtest is history; a forward test is now. Run the strategy in demo/paper mode for a few weeks in live conditions before committing real money. Prometheus includes a Profit Simulator to backtest the strategy and a demo mode to forward-test it.

Frequently Asked Questions

How do I backtest a crypto trading strategy?

Run it on several years of quality historical data (including volatile periods), with fees and spread included, then read the max drawdown, expectancy and profit factor — not just total profit. Finally, forward-test in demo mode before going live.

What is curve-fitting in backtesting?

Curve-fitting is over-tuning a strategy to past data so it looks perfect on history but fails live. Signs include a flawless equity curve and many finely-tuned parameters. Test on out-of-sample data the strategy wasn't optimized on to catch it.

Is a good backtest enough to go live?

No. A backtest shows how a strategy would have done in the past, not how it will do now. Always forward-test in demo/paper mode in current conditions before committing real money.

This article is for information only and is not financial advice. Cryptocurrency trading involves substantial risk of loss and is not suitable for everyone. Past performance does not guarantee future results. Never trade money you cannot afford to lose.