Backtest
IntelliBacktest
Backtest a Pine strategy, optimize its inputs, find its best timeframe, or run it across the whole market — no login.
Loading…
About IntelliBacktest
Paste a TradingView Pine strategy and IntelliBacktest becomes a strategy lab with four modes. Backtest it on real candles (up to 5,000 bars) for net return, equity curve, win rate, profit factor, max drawdown and the trade list. Optimize sweeps one of its inputs across a range and ranks every setting. TF sweep runs it across several timeframes to find where it works. And Multi-coin runs it across the top coins at once and ranks them — the fastest way to see whether an edge generalizes or was just curve-fit to one symbol. All four are things TradingView gates, charges for, or won't do at all. It's an independent approximation (no fees, slippage or sizing), and optimize/sweep/multi-coin results are in-sample, so validate winners out-of-sample. Free, no login, and your code never leaves your browser.
How to use
- Paste a Pine strategy (strategy.entry / strategy.close, or long/short signals)
- Backtest: pick a symbol and timeframe and see the full result
- Optimize: choose a numeric input and a range — every setting gets ranked
- TF sweep / Multi-coin: find the best timeframe, or test across the top coins at once
FAQ
What makes this different from TradingView?
Two things TradingView gates or lacks: parameter optimization (its optimizer is a paid Premium feature) and testing across many timeframes at once (it does one at a time). Here both are free and in your browser — the trade-off is that this is an independent approximation, not a tick-exact match.
How does Optimize work?
It reads your strategy's numeric inputs, sweeps the one you pick from a start to an end value by a step, re-runs the backtest for each value, and ranks them by net return with the best highlighted. It caps at 60 runs so it stays fast.
How does the timeframe sweep work?
It runs the same strategy, unchanged, across each timeframe you list (e.g. 15m, 1h, 4h, 1D — custom ones too) and ranks them by net return, so you can see whether an edge is timeframe-specific.
Aren't optimized results misleading?
They can be. The best setting on past data often overfits and fails going forward, which is why the results are labelled in-sample. Use optimize to understand a strategy's sensitivity, not to cherry-pick a number — then validate on data you didn't optimize on.
Will results match TradingView's Strategy Tester?
Not exactly. It enters and exits at the signal bar's close, one unit long or short, with no fees, slippage, pyramiding or custom sizing. Use it to check logic and shape, then confirm exact numbers in TradingView.
Is my strategy uploaded?
No. Compilation, execution and every simulation run happen in your browser; only public candles are fetched.