# [OPERATOR-AUTHORED TITLE]

> **CAPTURED MARKET DATA — IMMUTABLE LINEAGE BOUND.** This editable scaffold was generated from the canonical captured Binance Public Data Spot BTCUSDT dataset bound in this run. It is a historical result under the frozen specification and does not establish future profitability.

Scaffold render version: `captured_article_scaffold_v1`  
Scaffold render identity: `6ae8e7a98ff06edfcd977f9f1558dd14cec4b46299736cbb765beb67f23678cc`  
Family identity: `3744822a0e3cb1b34384c40dbe45b626f0d41a7d4af762d5fbeff0da471ac98a`

The scaffold render identity identifies this deterministic template contract only; it is **not** the numerical family identity and is **not** the M6 renderer source/build fingerprint.

## Standfirst — OPERATOR AUTHORED

[OPERATOR: Summarize why this source/strategy family matters and what question the test was intended to answer. Do not infer this from the numerical result.]

## What was tested

A frozen `study_family_v1` family of 5 `log_close_sma` members was evaluated LONG/FLAT on `BTCUSDT` `1d` daily bars with next-daily-open execution and an identically costed buy-and-hold benchmark. Every member was fixed before evaluation and starts independently FLAT at the common evaluation start.

| # | Member | Decision role | Parameters | Member criterion |
| ---: | --- | --- | --- | --- |
| 1 | vma_20 | primary | lookback=20 | `excess_return_v2` |
| 2 | vma_50 | robustness | lookback=50 | `excess_return_v2` |
| 3 | vma_100 | robustness | lookback=100 | `excess_return_v2` |
| 4 | vma_150 | robustness | lookback=150 | `excess_return_v2` |
| 5 | vma_200 | robustness | lookback=200 | `excess_return_v2` |

## Frozen decision rule

Family rule: `primary_plus_robustness` / `family_rule_v1`, k=0, n=5, primary=vma_20.

No member may be selected, removed, promoted, retried, or reordered after observing results. Overlapping lookback members on one asset are strongly dependent; this family rule is not a calibrated family-level confidence test and individual HAC bounds do not create a family-wise error rate.

## Results at a glance

Family decision: **DID_NOT_MEET_CRITERIA**  
Family completion: **COMPLETE**

| Member | Status | Scientific outcome | Strategy net return | Benchmark net return | Strategy max drawdown | Benchmark max drawdown | HAC 95% lower bound | Drawdown condition |
| --- | --- | --- | ---: | ---: | ---: | ---: | ---: | --- |
| vma_20 | `EVALUATED` | DID_NOT_MEET_CRITERIA | 546.444439% | 2260.989703% | -69.834191% | -76.629251% | -0.116192% | Not applicable to criterion |
| vma_50 | `EVALUATED` | DID_NOT_MEET_CRITERIA | 2643.621539% | 2260.989703% | -60.037418% | -76.629251% | -0.058426% | Not applicable to criterion |
| vma_100 | `EVALUATED` | DID_NOT_MEET_CRITERIA | 3152.030957% | 2260.989703% | -44.346165% | -76.629251% | -0.051450% | Not applicable to criterion |
| vma_150 | `EVALUATED` | DID_NOT_MEET_CRITERIA | 2217.955796% | 2260.989703% | -45.970465% | -76.629251% | -0.063127% | Not applicable to criterion |
| vma_200 | `EVALUATED` | DID_NOT_MEET_CRITERIA | 974.128651% | 2260.989703% | -61.850943% | -76.629251% | -0.092721% | Not applicable to criterion |

The complete primary and diagnostic-cost metric tables, both Sharpe conventions, member reasons, and any operational failure text remain in `report.md`; this scaffold does not silently omit unfavorable members from the frozen decision.

## Cost context

Shared modeled numerical costs: **15 bps (0.15%) per executed leg** primary; diagnostic cases: 10 bps (0.10%) per leg, 20 bps (0.20%) per leg. Family validation requires these numerical cost assumptions to be identical for every member. Each strategy case is compared with an identically costed buy-and-hold benchmark. These are modeling assumptions, not verified account charges and not an inferred fee/slippage decomposition.

Shared declared cost rationale (verbatim JSON string):

    "Fixed total transaction-cost assumption of 15 bps per executed leg, with 10 and 20 bps declared cost-sensitivity cases. These values are research assumptions for comparability and robustness and are not claimed to reconstruct Binance's historical fee schedule."

## Chronological context

- Evaluation interval: UTC `[2019-01-01, 2026-01-01)`, equivalent to 2019-01-01 through 2025-12-31 inclusive.
- Evaluated daily intervals: **2557**.
- Common first evaluation open: `2019-01-01`. Terminal liquidation open: `2026-01-01`.
- Shared captured dataset coverage begins `2018-06-15` because the largest declared member warmup is **200 days**; those extra rows are capture coverage only and do not extend shorter members' state.
- Per-member warmups:
  - vma_20: 20 days, first required warmup date `2018-12-12`.
  - vma_50: 50 days, first required warmup date `2018-11-12`.
  - vma_100: 100 days, first required warmup date `2018-09-23`.
  - vma_150: 150 days, first required warmup date `2018-08-04`.
  - vma_200: 200 days, first required warmup date `2018-06-15`.

## Available source and lineage facts

The source/date declarations below are preserved from operator-supplied member specifications; this scaffold does not verify source URLs, wording, transcription, or prior-inspection declarations against external material.

Member disclosures differ, so they are reported per member rather than presented as family-wide facts:

### Disclosure — vma_20

- Source title (literal): "Technical trading rules in the cryptocurrency market"
- Source author (literal): "Klaus Grobys; Shaker Ahmed; Niranjan Sapkota"
- Source date (literal): "2020"
- Source URL (literal): "https://doi.org/10.1016/j.frl.2019.101396"
- Date-selection reason (literal): "Human-approved later-sample interval [2019-01-01, 2026-01-01), immediately following the source study's 2016-2018 sample and using the platform's seven-calendar-year maximum. The interval was frozen before Strategy Lab evaluated this research case."
- Prior inspection declaration: `known`.

### Disclosure — vma_50

- Source title (literal): "Technical trading rules in the cryptocurrency market"
- Source author (literal): "Klaus Grobys; Shaker Ahmed; Niranjan Sapkota"
- Source date (literal): "2020"
- Source URL (literal): "https://doi.org/10.1016/j.frl.2019.101396"
- Date-selection reason (literal): "Human-approved later-sample interval [2019-01-01, 2026-01-01), immediately following the source study's 2016-2018 sample and using the platform's seven-calendar-year maximum. The interval was frozen before Strategy Lab evaluated this research case."
- Prior inspection declaration: `known`.

### Disclosure — vma_100

- Source title (literal): "Technical trading rules in the cryptocurrency market"
- Source author (literal): "Klaus Grobys; Shaker Ahmed; Niranjan Sapkota"
- Source date (literal): "2020"
- Source URL (literal): "https://doi.org/10.1016/j.frl.2019.101396"
- Date-selection reason (literal): "Human-approved later-sample interval [2019-01-01, 2026-01-01), immediately following the source study's 2016-2018 sample and using the platform's seven-calendar-year maximum. The interval was frozen before Strategy Lab evaluated this research case."
- Prior inspection declaration: `known`.

### Disclosure — vma_150

- Source title (literal): "Technical trading rules in the cryptocurrency market"
- Source author (literal): "Klaus Grobys; Shaker Ahmed; Niranjan Sapkota"
- Source date (literal): "2020"
- Source URL (literal): "https://doi.org/10.1016/j.frl.2019.101396"
- Date-selection reason (literal): "Human-approved later-sample interval [2019-01-01, 2026-01-01), immediately following the source study's 2016-2018 sample and using the platform's seven-calendar-year maximum. The interval was frozen before Strategy Lab evaluated this research case."
- Prior inspection declaration: `known`.

### Disclosure — vma_200

- Source title (literal): "Technical trading rules in the cryptocurrency market"
- Source author (literal): "Klaus Grobys; Shaker Ahmed; Niranjan Sapkota"
- Source date (literal): "2020"
- Source URL (literal): "https://doi.org/10.1016/j.frl.2019.101396"
- Date-selection reason (literal): "Human-approved later-sample interval [2019-01-01, 2026-01-01), immediately following the source study's 2016-2018 sample and using the platform's seven-calendar-year maximum. The interval was frozen before Strategy Lab evaluated this research case."
- Prior inspection declaration: `known`.

- Available deterministic identities: family identity above plus member numerical/source identities in `report.md`.
- Captured dataset lineage: provider `Binance Public Data`; market `Spot`; source `"Binance Public Data Spot monthly kline archives plus daily kline archives for the trailing partial month"`; interval `1d`.
- Immutable hashes: normalized bars `7920dbed13fa862d5beade58b981d647318c859d778c756e62358039cedcc49c`; capture manifest `21f579e0113ee397f1946083ef0e0692a1978845e9760dfa97036a7091ca7d54`; per-row source attribution `fd38f79394fec7ca3c72431b2f471a102f7f50258f1e6bb0b37a0c737a907640`; pre-run provenance `715e03437132a099035865351d4ca5c2ffae5bfdd0ca27aae6b1d1a1baa1db80`.
- The canonical workspace preserves these bound artifacts and verifies them offline; raw source archives remain external evidence identified by the preserved capture manifest rather than being re-downloaded during verification.

## Source critique and replication fidelity — OPERATOR AUTHORED

[OPERATOR: Compare the frozen transcription with the original source. Record ambiguities, mismatches, unverifiable assumptions, and any reason exact replication is or is not credible. Do not infer source fidelity from a successful decode or numerical result.]

## Interpretation — OPERATOR AUTHORED

[OPERATOR: Explain what the completed historical result does and does not support. Address the family rule, uncertainty, dependence among members, costs, benchmark choice, and whether the result is useful for the research question. Do not convert historical criteria into a claim of future profitability.]

## Limitations

- The captured dataset is provider-derived historical evidence bound by the hashes listed above; capture-source interpretation remains operator responsibility.
- Offline verification checks copied normalized bars and provenance artifacts; it does not re-download or replace provider data.
- The family contains strongly dependent variants on one asset; the family rule is not a calibrated multiple-testing correction or confidence probability.
- Results are historical outputs of the frozen mechanism/accounting assumptions and do not establish future profitability.
- No chart files are generated or claimed by this scaffold.
- Editorial/source judgments remain operator-authored and are intentionally absent from deterministic generation.

## Conclusion — OPERATOR AUTHORED

[OPERATOR: Write the publication conclusion after reviewing the source, factual report, assumptions, and limitations. Preserve an unfavorable or inconclusive result as such; do not select a preferred family member after seeing outcomes.]

## Local factual artifacts

- `input.json` — exact submitted family JSON preserved by canonical captured-data admission.
- `assessment.md` — deterministic structural/family transcription assessment.
- `report.md` — complete deterministic factual family result.
- `article-scaffold.md` — this editable M5 starting point. Preserve subsequent operator edits separately from deterministic regeneration.