Strategy research system
The Strategy Library, Strategy Object, strategy lifecycle, and promotion gates.
Crypto Strategy Library
V1 divides the library into eight categories spanning mechanisms, products, and data perspectives. Research must still explain the source of returns; the categories are not independent:
A common research chain
- Observation
- Mechanism
- Hypothesis
- Data
- Signal
- Backtest
- Falsification
Strategy Object
Every strategy must have a Strategy Object with all fields filled in. A field you cannot fill shows where the research remains incomplete.
Save your research using the editable, copyable, downloadable Strategy Object template. Complete three hypotheses in the Chapter 20 workshop; Phase 5 explains how to produce and assess a complete Backtest Report.
Frequently skipped fields:
- Mechanism: not “this indicator works,” but “who pays you, and why are they willing to pay?”
- Failure: specify before launch what would make you accept that the strategy is wrong.
- Capacity: how much capital the strategy can deploy. Beyond that size, your own impact costs consume its returns.
Strategy lifecycle
- Idea
- Research
- Testing
- Validated
- Paper
- Shadow
- Live Small
- Production
- Scaled
A strategy does not run forever once completed. Markets change, participants learn, strategies become crowded, and alpha decays. The lifecycle gives each strategy an explicit state at all times.
Promotion gates
A profitable backtest is not sufficient for Live. Every promotion must pass the following eight gates again:
| 1 | Economic logic Economic Rationale | Can you explain who pays you and why they are willing? |
| 2 | Statistical significance Statistical Significance | Can the result be distinguished from luck? How many attempts found it? |
| 3 | Out-of-sample OOS | Does it hold on completely unseen data? |
| 4 | Robustness Robustness | Does it hold across parameters, markets, and periods? |
| 5 | Costs Cost | What remains after fees, slippage, funding, and impact? |
| 6 | Risk Risk | What is the worst-case loss and correlation with the existing portfolio? |
| 7 | Execution Execution | Can actual execution approach backtest assumptions? What is capacity? |
| 8 | Market regime Regime Analysis | Which regimes produce profits and which cause failure? |
Connections to the course
- Foundation Phase 4 develops strategy hypotheses; Phase 5 produces your first backtest report.
- Advanced B takes a complete strategy study through the research chain.
- Advanced F manages strategies at different lifecycle stages.