Phase 4 introduction
Strategy thinking. Turn a trading idea into a hypothesis with a mechanism, a signal, and a way to be disproved.
What this phase establishes
“Buy when this indicator crosses upward” is a rule, not a strategy. A strategy must explain: who pays you, why will they keep paying, and when will they stop?
This phase develops the first half of the research chain:
- Observation
- Mechanism
- Hypothesis
- Data
- Signal
- Backtest
- Falsification
Six chapters
Whose money funds a sustainably profitable strategy?
What is wrong with trading whenever an indicator signals?
What economic mechanism makes a strategy profitable?
How can a computer evaluate a feeling such as the market is too hot?
Why does the same strategy profit last year and keep losing this year?
What would show that your strategy is wrong?
Phase project: 3 Strategy Hypotheses
Write three hypotheses, each specifying on one page:
- Observation: the market phenomenon you observed
- Mechanism: why it happens, who is on the other side, and why they pay
- Hypothesis: under which conditions price will behave how, and for how long
- Signal: conditions built from data available at that time
- Applicable regime: trending or ranging, high or low volatility
- Invalidation: observations that would make you accept it is wrong
Use different return sources—for example, momentum, Carry, and Flow. See Strategy research system.
Phase project workshop
Complete three hypotheses with different return sources in the Chapter 20 workshop, then copy or download strategy-hypotheses.md. Each must include an observation, mechanism, conditional prediction, data timing, signal, regime, and invalidation action. Complete fields do not establish the mechanism; a peer must be able to identify evidence that would disprove it.
| Acceptance item | Passing standard | Requires rework |
|---|---|---|
| Source | Three distinct categories with actual compensation sources explained | Merely changing three indicator names |
| Mechanism | Observable participants, constraints, and behavior | Explaining rises by rises |
| Data and signal | Explicit publication time, window, threshold, and missing-data rule | Future values or hindsight regime labels |
| Regime | Boundaries and costs of misclassification | Keeping only profitable environments |
| Falsification | Contrary evidence and stop/restart conditions | Moving the criteria after failure |
Strategy Object template
Expand one hypothesis into the unified research object. Label unfinished fields “Unknown” with a plan to obtain evidence. Bring back backtest, OOS, and cost results in the next phase. Text remains only on the current page; copy or download before refreshing.
Use the course's standard fields. Keep unverified items unknown and explain the gaps. Template prompts are not results.
Edits exist only on this page and clear on refresh. Copy or download promptly.
Course versionV1-docs; sourcep3:report:strategy-object.md；Non-chapter tool; no chapter or node inferred.
Records parameters and results at the click only; does not mean the experiment passed.View snapshot to save
Phase self-test
Answer independently first. If you cannot explain an item, return to its chapter.
Continue to Phase 5 · Backtesting foundations, keeping drafts and every failed attempt.
Knowledge nodes
| concept_id | Nodes | Foundation | Advanced | 3D |
|---|---|---|---|---|
| TRD-STRAT-001 | Return source Edge Sources | Chapter 19 | B | — |
| TRD-STRAT-002 | Hypothesis Hypothesis | Chapter 20 | B | — |
| TRD-STRAT-003 | Mechanism Mechanism | Chapter 21 | B | — |
| TRD-STRAT-004 | Signal Signal | Chapter 22 | C | — |
| TRD-STRAT-005 | Market regime Market Regime | Chapter 23 | C | Macro knobs |
| TRD-STRAT-006 | Refutation Falsification | Chapter 24 | B | — |