Strategy Backtester
Walk-Forward Analysis catches overfitting. 1,000-path Monte Carlo models your worst-case drawdown. Grid-search finds robust parameters. Then overlay your live trades on the theoretical curve — all inside the same journal.
Or read first — what is Walk-Forward Analysis? →
A strategy that only works on the data it was built on isn't an edge — it's a memory. Walk-Forward Analysis splits your data: optimize on two-thirds, test on the third it never saw.
If the strategy falls apart on out-of-sample data, it's curve-fit. SignalDeck runs the split automatically and delivers a plain-English Pass/Fail verdict so you know before going live.
A single backtest path is a lottery result. Monte Carlo simulation shuffles your trade history 1,000 times and shows you the full distribution of possible outcomes — including the P5 worst-case drawdown you should be sized for.
Most backtesting tools show one path. One path hides 999 ways it can go wrong. Size your live position to survive the 5th percentile, not the median.
Grid-search runs up to 500 parameter combinations in a single pass. The goal isn't finding the combination that peaks on historical data — it's finding the combination that performs across a range of nearby values.
A parameter set that only works at exactly RSI(14) with OB=70 is fragile. A set that holds across RSI(12–16) with OB=68–72 is robust. SignalDeck shows you both.
Then link the backtest to your live journal strategy — and overlay the theoretical equity curve against your actual trades to see exactly where execution diverges from theory.
Test any strategy out of the box. Tune it with grid-search. Validate it with Walk-Forward. Then trade it live and compare the execution against the model.
Walk-Forward Analysis, Monte Carlo, grid-search, and a live journal overlay. No other trading journal combines them.
| Approach | Walk-Forward | Monte Carlo | Grid-Search | Journal Overlay |
|---|---|---|---|---|
| SignalDeck | ✓ | ✓ | ✓ | ✓ |
| Other trading journals (TraderSync, Edgewonk, TradesViz) | — | — | — | — |
| TradingView Pine Script | — | — | — | — |
| Python / Backtrader | Manual | Manual | Manual | — |
| Spreadsheet | — | — | — | — |
Competitor features change — verify current specs on their sites. See full comparisons →
Walk-Forward Analysis, Monte Carlo, and grid-search in a single backtest — then compare your live execution against the model. Free during beta.
Start free in SignalDeckWalk-Forward Analysis splits your historical data into in-sample (IS) and out-of-sample (OOS) windows. The strategy is optimized on the IS portion and then tested on OOS data it never saw. If performance collapses on OOS, the strategy is curve-fit. SignalDeck delivers a plain-English Pass/Fail verdict.
Monte Carlo shuffles your trade history 1,000 times to generate a distribution of possible equity curves. The P5 worst-case drawdown tells you what to expect in a genuinely bad variance run — before you're living it. Most backtesting tools only show a single path, which is misleading.
16 built-in strategies including SMA/EMA crossover, RSI mean reversion, Bollinger Bands, and MACD — tested against up to 5 years of data. Grid-search runs up to 500 parameter combinations per strategy.
Yes. SignalDeck lets you link any completed backtest to your live journal strategy and overlays the backtest equity curve against your actual trade history — showing you exactly where execution diverges from theory. Slippage, discipline violations, and timing differences all show up in the gap.