Strategy Lifecycle
DraftStatus: Draft
Version: 0.1.0
Last Updated: 2026-05-16
Owner: Axodus Trading
Purpose
Every trading strategy must have a lifecycle. A strategy should not move from idea to user-facing execution without research, testing, risk review, access policy, monitoring, and reporting.
Scope
This page applies to internal, treasury-facing, user-facing, educational, and experimental strategies.
Lifecycle Stages
- Concept: define hypothesis, market type, and initial risk assessment.
- Research: analyze market logic, data requirements, assumptions, and failure modes.
- Backtesting: test on historical data and document limitations.
- Paper trading: test live conditions without real capital.
- Risk review: define limits, position sizing, leverage rules, and kill switch.
- Limited pilot: restrict live use under monitoring.
- Governance or product approval: approve material internal, treasury, or user-facing release.
- Active: make the strategy available under defined scope.
- Iteration: update parameters or logic with versioning and review.
- Pause or deprecation: stop strategy due to risk, performance, policy, or maintenance.
Strategy Statuses
Statuses include Concept, Research, Backtesting, Paper Trading, Risk Review, Limited Pilot, Pending Approval, Active Internal, Active User Facing, Paused, Deprecated, Failed, and Needs Review.
Versioning
Material strategy logic changes require a new version. User-facing changes require disclosure when material. Treasury strategy changes require reporting when material. High-risk changes require review before release.
Promotion Gates
Promotion gates require defined hypotheses, data sources, backtest reports with limitations, paper trading observations, risk limits, pilot review, disclosures, monitoring readiness, access policy, and approval when material.
Backtest Requirements
Backtests should include data period, market source, assumptions, fees, slippage, liquidity constraints, leverage if any, drawdown, number of trades, robustness notes if available, overfitting risk, limitations, and a statement that backtests do not guarantee future performance.
Deprecation Conditions
Deprecation may occur because of persistent underperformance, excessive drawdown, market regime change, exchange API incompatibility, security issue, governance policy change, unacceptable user risk, infrastructure instability, or replacement by a better strategy.