Strategy Origins — who created what (canonical provenance map)
directive: *"I need to be able to tell whether a strategy is one I created, a proprietary
strategy, or a standard (regular) strategy — the distinction should be clear."*
The three origin classes (BINDING vocabulary):
- 🧑 FOUNDER — the founder's own concept, dictated by him (the house implemented and tested it).
- 🏠 HOUSE — proprietary, patearn-built: designed inside this project, not public knowledge.
- 📚 CLASSIC — standard/"bookish": public-domain families from the literature, admitted
only after re-proving themselves on our own data.
Mixed lineage is stated explicitly (e.g. "classic base + the desk layer") — never blurred.
Labeling rule (BINDING): every page inthe strategies folderdeclaresOrigin:in its
header; every surface that shows a strategy's output states the origin class in its fence
line. New strategies may not ship without the label.
The map
| Strategy / construct | Origin | Where it lives | Status |
|---|---|---|---|
| DVPT positioning (delivery-value-per-trade, power baselines, key price, ×power) | 🧑 FOUNDER (concept) + 🏠 implementation | /dash/stocks · screener "pos" group | live, descriptive |
| Accumulation (signed accumulation/distribution) | 🧑 FOUNDER doctrine + 🏠 model | /dash/mep | live; descriptor-only |
| Wolfe waves | 📚 CLASSIC base (Bill Wolfe) + 🧑 FOUNDER §B strength rebalance & fractal gate | /dash/wolfe/scan + trades | live; BULL selection edge OOS-validated; book rejected |
| CPR structure pillar | 📚 CLASSIC base (pivot-range school) + 🏠 multi-TF amplification | /dash/cpr | live, descriptive |
| Relative Strength suite (Rotation Map · RS-band · rotation · Mansfield · capture · size-index) | 📚 CLASSIC families (Rotation Map — de Kempenaer · Mansfield/Weinstein RS · rotation) + 🏠 measurement | /dash/rotation-map · rotation · rsband · capture-map | live, descriptive |
| Harmonic (XABCD / PRZ) | 📚 CLASSIC (Gartley/Carney) + 🏠 PIT detector | /dash/harmonic | live, descriptive · backtest-gated |
| STREAM BAND (13-EMA banks + 5-EMA trigger) | 🧑 FOUNDER | Screen+ "rev" group (band state · stretch pctile) | falsified as signal (ledger 07-13); live as context |
| FRACTAL FLOOR / CEILING | 🧑 FOUNDER | Screen+ "rev" group + reclaim/slip pills | falsified as entry (07-14…14c); live as risk-geometry context |
| Buyback tender quota play | 🧑 FOUNDER (charter §2.4) | /dash/buyback-calc | live, personal-scale |
| The Union (RS turn ∪ RS trend, stock-level) | 🧑 FOUNDER (both signal theses — oversold-RS turn + persistent RS; cash-out sizing) + 🏠 HOUSE (union construction, PIT harness, falsification + pre-registration) | the union notes (no live surface by design) | RESEARCH — pre-registered lead, SEALED, not deployed |
| pt14 quality (14-pattern durability) | 🏠 HOUSE | /dash/stocks · screener "qual" | live; filter/veto, not a ranker |
| Concall credibility (concall guidance-accuracy) | 🏠 HOUSE | /dash/concalls · screener "cci" | live; factor falsified → descriptive/veto |
| C capital-allocation | 🏠 HOUSE | screener "ca" · /dash/momentum-scan C-blend | live; blend/veto , never a ranker |
| Conviction shortlist (cross-pillar synthesis) | 🏠 HOUSE | /dash/conviction | live |
| Launchpad (momentum + contracting vol screen) | 🏠 HOUSE | /dash/launchpad | validated screen; no fundable edge net |
| Seasonal tape (certification-gated calendar residuals) | 🏠 HOUSE (framework) | /dash/seasonal-tape | live; 0-certified IS the finding |
| Reversal context columns (implementation) | 🧑 FOUNDER concepts + 🏠 honesty fences | Screen+ | live, descriptive |
| Sector Rotation (V8 champion · V17 defensive-fill candidate) | 🧑 FOUNDER (concept + every lever) + 🏠 implementation/harness | research modules (sector_rotation*); page the sector rotation notes | research, CONDITIONAL; long-only |
| SPRINTER-25 (MOM12) | 📚 CLASSIC (Jegadeesh-Titman 12-mo momentum) | /dash/model-portfolios | model portfolio since 2012 |
| PACER-25 (RISKADJ) | 📚 CLASSIC (vol-adjusted momentum school) | /dash/model-portfolios | model portfolio since 2012 |
| CRAFTSMAN-25 (QUAL_MOM) | 📚 CLASSIC blend, 🏠 delivery-leg formulation | /dash/model-portfolios | model portfolio since 2012 |
| STEADY-25 (LOWVOL_MOM quarterly) | 📚 CLASSIC (low-vol + momentum, "conservative formula") | /dash/model-portfolios + /dash/momentum-scan/slow | model portfolio; NET champion |
| Momentum / RISKADJ engine (the benchmark ranker) | 📚 CLASSIC (Jegadeesh-Titman · vol-adjusted momentum) + 🏠 measurement | /dash/momentum-scan | live; gross benchmark, not fundable net of cost |
| Factor league (families ranked incl. failures) | 📚 CLASSIC content, 🏠 measurement | /dash/factor-league | live |
| Value / deep value / quality-standalone / deliv-mom rows | 📚 CLASSIC | factor league (verdict rows) | FAILED/REJECTED on our data — shown with numbers |
| Classic screens catalog (in flight, sibling lane) | 📚 CLASSIC | /dash/classics (when it lands) | building |
External sources — the standing policy
Methodology from outside is welcome; data from outside is not. Encoding a public, "bookish"
strategy from books/papers is exactly how the CLASSIC class grows — but every such strategy must
feeds), (2) pass the standard exam (walk-forward vs the 0.89 hurdle + cost realism) before any
portfolio treatment, and (3) carry its 📚 label and its measured verdict — including failures
(value's negative alpha stays on the league page precisely because it is a classic).
Documentation loop (the desk directive, BINDING)
Every strategy discussion — wins, failures, refinements, dictated rules — lands in that
strategy's the strategies folder page (or the ledger for results) in the SAME session it happens.
Nothing lives only in chat. The loop: **discuss → document → implement per the doc → test →
record the verdict in the ledger → update the page's status.** This file is the provenance
index over that loop.