Backtesting & Validation

Forex Backtesting, Validation & Robustness Testing

Finding out whether a strategy actually works, before capital is at risk.

A backtest is easy to pass and hard to trust. The interesting question is never the equity curve — it is how much of that curve survives out-of-sample data, realistic costs, and parameter perturbation. That is what we test for.

What this covers

Event-driven backtests that model fills, spread, slippage and financing
Walk-forward and out-of-sample protocols designed before results are seen
Monte Carlo and sensitivity analysis on parameters and trade ordering
A written verdict, including when the honest answer is that the edge is not there

Services in this area

16 pages

Forex Strategy Backtesting

Backtests that model spread, slippage, financing and realistic fills.

Forex Strategy Validation

An independent verdict on whether a strategy's edge is real or an artefact.

Strategy Optimization

Parameter work done so that it improves the strategy rather than the backtest.

Custom Backtesting Engine

An engine built for your instruments, costs and execution assumptions.

Python Backtesting Framework

A reusable framework so every future test starts from a known-good base.

Event-Driven Backtesting

Event-driven simulation that shares code paths with live execution.

Tick Data Backtesting

Tick-level testing where intrabar path and spread genuinely change the result.

Portfolio Backtesting

Testing strategies together, with shared capital and correlated drawdowns.

Walk-Forward Analysis

Rolling re-optimisation and out-of-sample testing that mimics live use.

Out-of-Sample Testing

Held-out data used once, at the end, as it was always supposed to be.

Monte Carlo Trading Analysis

Resampling trades and paths to see the range of outcomes you might have had.

Strategy Stress Testing

Behaviour under widened spreads, gaps, outages and regime breaks.

Parameter Sensitivity

Whether performance sits on a plateau or a spike — the difference matters.

Transaction Cost Modeling

Cost models built from your fills, not from a fixed assumed spread.

Trading Strategy Overfitting

Quantifying how much of the backtest is search bias rather than edge.

Backtest vs Live Performance

Reconciling the two, and attributing the gap to specific causes.

Tell us what you are trying to build

Describe the problem and we will tell you plainly whether we can help, roughly what it would take, and what it would cost. If it is not a fit, we will say so.

Test My Strategy