Strategy Optimisation & Testing
A strategy that performs in backtests is not necessarily a strategy that works in live markets. We apply rigorous statistical testing to validate your edge, detect overfitting, and give you genuine confidence before capital is deployed.
What's included
- Walk-forward analysis and out-of-sample validation
- Monte Carlo simulation and robustness testing
- Genetic and parameter optimisation
- Overfitting detection and curve-fitting analysis
- Stress testing across market regimes
- Comprehensive written test reports
How it works
Strategy Review
We receive your strategy code or specification and review the logic, parameter space, and existing backtest results. We flag anything that looks curve-fitted before testing begins.
Testing
We run the full testing suite walk-forward analysis, Monte Carlo, stress tests across different market regimes and compile the results.
Report & Recommendations
We deliver a written report with findings, statistical metrics, and clear recommendations on whether the strategy has a genuine edge and what, if anything, to change.
Common questions
My backtest looks great. Why do I need this?
A strong backtest result on its own is not sufficient evidence of a real edge. Parameter fitting to historical data is common and produces misleading results. Walk-forward and out-of-sample testing give you a much more honest picture.
What platforms and formats do you accept?
We work with MT4/MT5 strategies, Python-based backtests (Backtrader, Vectorbt, custom), and TradingView Pine Script strategies. Send us what you have and we'll confirm compatibility.
How long does testing take?
Most projects complete in 1–2 weeks depending on the complexity of the strategy and the number of tests required. We'll give a precise timeline after the initial review.
What if the strategy fails the tests?
We'll tell you clearly. The report will explain where it breaks down, whether it's fixable, and what changes we'd recommend. An honest failure result saves capital.