About
Coin Backtest is a free tool for testing cryptocurrency trading strategies against real historical market data. It was built on a simple conviction: most trading ideas that sound good never get tested honestly, and the ones that are tested are usually tested in ways that flatter them. This site exists to make skeptical, realistic backtesting available to anyone, without a paywall, an account, or a spreadsheet.
What it does
You build a strategy from indicators — entry triggers, filters, exits and position sizing — and the engine replays seven years of Binance candle data to show how that strategy would have performed. It reports return, maximum drawdown, win rate, profit factor and a trade-by-trade log, broken down across multiple time windows so you can judge consistency rather than a single lucky run.
How the engine stays honest
- Independently verified. The TypeScript engine that runs in your browser is checked against a separate Python reference implementation using golden tests. Both must produce trade-for-trade identical results, so the accounting is not left to chance.
- No look-ahead.Signals are evaluated only on closed candles, and trades open on the next candle's open. The engine can never "see" a price it could not have known at decision time.
- Real costs. Trading fees are charged on both sides of every trade by default, and leverage is modelled with isolated-margin liquidation — a wiped-out position is booked as a real loss.
- Fresh data.Historical candles are stored with the site and topped up with the latest closed bars from Binance's public data API on every run.
Why it runs in your browser
Every calculation happens on your own device. Your strategy configuration is never sent to a server, which keeps your ideas private and lets the tool stay free — there is no backend compute to pay for. The only external request is for public candle data; no personal information is transmitted.
What it is not
This is a research and education tool, not a financial adviser. A backtest describes the past; it does not predict the future, and a strategy that performed well historically can still lose money live. Nothing on this site is investment advice or a recommendation to trade. Cryptocurrency trading, especially with leverage, carries a high risk of loss. Test ideas here to challenge them, not to confirm what you already hope is true.
Contact
Suggestions and bug reports are genuinely welcome and reviewed roughly every two weeks. You can reach the site through the contact address in the footer.
Reading is good. Testing is better. Run a real backtest on 7 years of Binance data, free.
Try the backtester