Backtest your trading strategy before you risk money
Describe the rules in plain words, choose a market and timeframe, and get a report with costs, drawdowns and the number of trades. No code, no card, no promise that it will work.
The form takes about two minutes. Registration is only needed to save the strategy and run it.
- 01Rules, not settings
Describe the logic; the engine builds the setup.
- 02Costs included
Commission, spread and slippage are part of every run.
- 03Trial count on
Every variant raises the threshold for luck.
- 04Saved in Nodes
The form becomes a strategy draft you can inspect.
A backtest is a stress test, not a crystal ball
A rising equity curve proves almost nothing on its own. The useful question is narrower: what exactly did the rules do, after costs, on data the author had not seen?
What a backtest can tell you
A backtest replays a fixed set of rules over historical candles and records every trade. It shows whether the edge exists in the sample, how much of the profit survives commission and slippage, how deep the drawdowns were, and how the result compares with buy and hold. That is enough to reject weak ideas before they cost real money.
What it cannot tell you
It cannot promise that the edge will continue. A backtest is in-sample evidence; the future is out of sample. That is why the verdict needs a sealed holdout and a forward track, not a prettier curve.
How to describe a strategy the engine can test
Write rules as conditions, not opinions. “Buy when the trend is strong” cannot be executed. “Buy when the 20 EMA is above the 200 EMA and the last close is above the previous day’s high” can. Name the market, timeframe, direction and the exact trigger. If a rule needs a value, give the value.
- Market and timeframe
- Entry trigger
- Exit: stop, target or time stop
- Position size and risk per trade
- Session or filter rules, if any
The minimum we need
What the run includes
Every run uses historical candles and a realistic execution model: commission, spread, slippage and position sizing. The report shows trade count, win rate, profit factor, maximum drawdown, cost drag and return versus buy and hold. A number without a sample size is not a result — so the trade count is always on screen.
Costs are part of the strategy
A mean-reversion system can look profitable before fees and lose money after them. We show both numbers, so you can see how much of the edge the market keeps.
Why the number of attempts changes the verdict
The more variants you test, the easier it is to find one that looks good by chance. Ten attempts on the same idea are not ten independent confirmations. Normal Trading counts trials, groups near-duplicates, and raises the threshold as the count grows. That is the difference between a backtest and a search for a flattering chart.
What happens after registration
The form becomes a strategy draft in Nodes with the rules and market context attached. You can inspect it, edit the graph, run it again, and keep the full history of attempts. The counter cannot be reset by opening a new browser tab.
- The entry can be written as a condition.
- The exit is defined.
- The timeframe matches the idea.
- You know how many variants you have already tried.
If something is missing, the form still accepts the idea. We will mark what has to be made explicit before the run.
Four steps, no code
The form is not a lead magnet. It is the first half of the setup; the second half is the run.
Describe the idea
Write the entry, exit and risk in plain words. One rule per line.
Pick the market
Symbol, timeframe and direction. The same idea can be tested on several markets later.
Create an account
The draft is saved to your workspace and appears in Nodes. No strategy is invented for you.
Read the verdict
The report shows the numbers, the sample size and the checks the idea failed.
A negative result is a result. It saves the money you would have spent learning it live.
Questions before the first run
Is the backtest really free?
Do I need to know Pine Script or Python?
Which markets and timeframes can I test?
How long does a backtest take?
Will it tell me whether the strategy will be profitable?
What happens to my idea after registration?
Can I test a strategy I saw on YouTube?
Why do you need an account before showing the result?
Have a rule you keep meaning to test?
Write it down. The form will show you what the engine needs before it can answer.