ZenWave is a systematic capital platform. Engines are developed and operated under a Walk-Forward Analysis (WFA) protocol where out-of-sample (OOS) results govern acceptance.
Walk-Forward Analysis divides historical data into sequential in-sample (IS) and out-of-sample (OOS) periods:
ZenWave engines use multi-slice anchored WFA. Each engine is evaluated across multiple slices (A1–A4, B1–B5) for parameter stability and regime alignment.
Initial research, validation, and deployment were conducted on USDJPY due to its liquidity profile, execution consistency, and identifiable macro regime behavior.
Additional markets may be evaluated under the same protocol.
Risk controls are protocol-defined and enforced. No discretionary overrides are permitted. Risk is treated as a first-class constraint.
Position sizing is derived mechanically from a predefined risk percentage and stop-loss distance.
Reported risk is calculated from stop-loss exposure relative to account equity. Actual exposure may differ due to minimum lot size granularity (0.01 lots), broker price rounding, and execution effects such as slippage. Parameters are frozen during live deployment.
No martingale. No grid logic. No hedge stacking. No loss-based sizing.
Stop-loss and take-profit required on every position (server-side).
Server-side execution only.
Engine isolation enforced.
The platform operates multiple engines with defined time horizons:
These engines are combined into ZenWave Book, the unified portfolio account with its own risk cap and allocation rules.
AI-assisted tooling supports candidate generation and independent review. Decision authority remains human.
Generates candidates and assists with feature exploration and implementation.
Reviews WFA results, verifies slicing boundaries, checks for data leakage, and flags overfitting indicators.
Engine selection, parameter acceptance, and risk allocation are approved by the operator.
Published claims are supported by reproducible artifacts and a documented review process. Non-negotiable principles:
Performance is measured only on out-of-sample data. In-sample numbers are never reported as expected performance.
WFA slices, parameter ranges, and OOS equity curves are published in quantitative reports and supporting media.
Live results are tracked on cTrader Copy — publicly visible balance, equity, and trade history.
Review tooling verifies slicing boundaries, checks for data leakage, and flags potential overfitting indicators.
Methodology is fixed prior to live deployment and is not modified in response to performance fluctuations.