How a strategy earns the right to trade.
Ten stations from idea to record. The system's default state is NO TRADE — this page shows, in plain language, what it takes to change that.
Ten stations. One default: no trade.
A promising idea is not a tradeable strategy. Each station below must pass before the next one matters — and a strategy that fails goes back or into the record, not into an account.
Strategy Definition
Every strategy begins as a written hypothesis: entries, exits, invalidation rules, market context, timeframes, risk assumptions. Anything ambiguous or discretionary is flagged before testing begins — not discovered after.
Rule Formalisation
The hypothesis becomes deterministic rules with defined inputs, dependencies, decision states and rejection criteria. Unknown parameters are documented as unresolved — never filled with convenient assumptions.
Evidence-Based Analysis
Every decision must be supported by current, verifiable market evidence — depending on the strategy:
- Market structure across timeframes
- Price action & liquidity zones
- Volume & auction profiles
- Order flow & absorption
- Volatility & market regime
- Spread, liquidity & execution conditions
Missing, outdated or contradictory evidence has exactly one output:
No tradeSequential Validation
The analysis runs as a sequence of specialised checks — context, structure, confirmation, execution quality, risk — each evaluated separately. A later signal can never override a failed mandatory condition, and analytical components may propose or reject a trade — they do not possess order authority.
A failed mandatory condition ends the sequence:
No tradeHistorical Testing
The rules run against years of historical data under realistic assumptions — fees and funding, bid–ask spreads, slippage and latency, partial fills and rejected orders, tick and lot-size restrictions, liquidity limits. The objective is not the most profitable configuration in hindsight, but whether the edge survives realistic costs.
- Anchor —Profit factor > 1.20 under stressed costs
Robustness & Walk-Forward
Promising results are re-tested across periods, assets and market regimes. Parameters are established on training data, then frozen before out-of-sample testing. Sensitivity analysis, walk-forward validation and stress testing expose overfitting, unstable parameters, and strategies carried by a handful of exceptional trades.
Paper & Shadow Trading
A strategy that survives history is observed on live market data without placing real orders: signal timing, data quality, operational stability, order and stop behaviour, expected versus observed execution, reconciliation after interruptions. Paper success means nothing until it repeats in the present.
Controlled Execution
Only strategies that complete every previous station may be considered for limited live deployment — and execution remains subject to independent safeguards the strategy cannot bypass: position and exposure limits, data-quality checks, server-side protection orders, persistent kill-switches, and daily and weekly de-risking brakes.
- Anchor —−5% from the account's peak = full shutdown
- Anchor —Written broker approval before any funded trade
Unsuitable execution conditions have the same output as missing evidence:
No tradeAuditability
Every relevant decision is versioned and traceable — inputs, rule evaluations, rejected signals, risk decisions, orders, fills, manual interventions. A material change creates a new strategy version and starts validation again.
Continuous Learning
Failed and inconclusive strategies are not silently discarded. Their assumptions, test conditions and failure reasons stay in the research record — so the validation process itself improves without rewriting historical outcomes.
A strategy should advance because the evidence supports it — not because the system was forced to produce a trade.
When evidence is incomplete, risk cannot be measured, or execution conditions are unsuitable, the correct outcome is
NO TRADEHistorical testing, paper trading and technical validation do not guarantee future performance. This methodology is designed to improve transparency, reproducibility and risk control — not to eliminate market risk.
See it applied.
Our first reference strategy is moving through this pipeline now. Create your account and follow the process — including the setbacks.
Create accountOpen registration. Not available to US persons.