Forex backtesting software, compared honestly.
Seven head-to-head comparisons, each written against the competitor's own published documentation — and each with a section naming the cases where they are the better tool and we are not. If a comparison page never reaches that section, it was written by the marketing department.
One is a professional quant platform where you write the code. The other tests the strategy you already trade, without any.
Read the comparison →The best charts in the business, and a backtester that tells you exactly which assumption it made when your stop and target shared a bar.
Read the comparison →MetaTrader's tester is free, broker-native and runs the exact code you will trade. Here is where that is enough, and where it is not.
Read the comparison →One trains your hand. The other tests your rules. Most traders who need one do not need the other.
Read the comparison →StrategyQuant generates thousands of strategies and filters them. We take the one you already believe in and try to break it.
Read the comparison →Build Alpha's robustness testing is the most thoroughly documented in retail software. Here is what it does not tell you, and what it does better than we do.
Read the comparison →Both are no-code and both have AI. The difference is what happens when the backtest comes back looking good.
Read the comparison →Every factual claim about another product was checked against that vendor's own documentation or pricing page on 29 July 2026, and each page links its sources so you can check us. Where research could not confirm something, the page says what we could confirm and stops — "not documented" and "does not exist" are different claims and we do not blur them. Our own limitations appear on these pages too: we do not execute live trades, we model one conservative cost pair across instruments, and we cover 30 markets rather than every asset class.
Pick by the job, not the feature list.
Most of these tools are good at something specific. The fastest way to choose is to name what you are actually trying to do.
Out-of-sample grading, costs charged before the verdict, and a correction for how many variants you tried. If the answer is no, you want a tool that will say so.
We do not execute trades. You export to MQL or Pine and arm it yourself on your own broker — if execution is the whole job, start where execution lives.
A code-first platform with its own data and deployment will fit your hands better than our rule graph. Our Python path is an import route, not a research environment.
We cover 30 markets — FX majors and crosses, indices, metals, oil and crypto. Outside that, breadth beats our depth.
Your own trades or a backtest, re-sized onto the firm's risk and run against their rulebook — including the daily-loss rule counting open positions, which is where most simulations quietly pass an account that would have failed.
Nothing here competes with a great charting front end. Our chart reading is causal structure for grading decisions, not a workspace to trade from.
Four of those six point somewhere else. That is roughly the real ratio, and a comparison set that never reached that conclusion would not be worth reading.
These questions decide whether a backtest means anything, and they apply to every tool on this page including ours. Ask them wherever you end up.
A result measured on the same data the strategy was tuned on is a rehearsal. Ask where the held-out slice is and how many times it has been touched.
When one bar contains both your stop and your target, something has to decide the order. Finer data can measure it; an assumption should be the pessimistic one, and you should be told which happened.
Spread, commission and slippage should come off first, at levels the instrument really trades at. On a fast timeframe those costs can exceed the whole stop — at which point every strategy converges on the same loss.
If you tried forty variants and kept the best, the best-of-forty is not a like-for-like result. Ask whether the platform corrects for how hard you looked, and how.
Our answers are written up in full — read the methodology.
Or skip the reading and test your own strategy.
One full backtest free, no card, and an honest out-of-sample verdict — whatever it says.