Objective
Independently replicate the distinct RecessionAlert row CHARTS > Bitcoin | RecessionAlert from the checked-in canonical snapshot at data/recessionalert.sqlite, without live-site requests, credential use, private account data, or mutation of the crawl database.
Approach
The primary input is the manifest-selected page body opened read-only from data/recessionalert.sqlite, cross-checked against research/findings/recessionalert_inventory.md and research/artifacts/recessionalert/manifest.json. The page digest is sha256:48ab3995ad8abafa4060fa4a72dd658607ddd205dd28dc1e040abe7ab38b2139; it is 116,181 bytes, HTTP 200, and fetched at 2026-08-22T23:39:03Z. The crawl log contains five same-URL page vintages, with the latest digest selected. The target page has zero canonical asset rows.
The page links to a separate Excel Data page. Four unique workbook artifacts are hash-pinned locally, but the snapshot does not identify a native workbook, sheet, or column for this distinct Bitcoin row. They are recorded as adjacent provenance and are not substituted for a target numeric series.
Disclosed rule and interpretation
The cached page is a multi-model context suite rather than one formula. BOCA-3 and BOCA-4 examine daily Bitcoin-blockchain address holdings in five categories. BOCA-3 highlights Whales and Fish growing while exchange balances decline; BOCA-4 highlights Whales, Crabs, and Shrimps growing against declining exchange balances. Falling exchange holdings are described as bullish and rising exchange holdings as bearish. The page's stated ownership bands are Humpback Whales above 10,000 BTC, Whales 1,000–10,000 BTC, Fish 10–100 BTC, Crabs 1–10 BTC, and Shrimps below 1 BTC; the apparent 100–1,000 BTC band is not disclosed.
The technical chart is described as updated every 30 minutes and combines SuperTrend, Donchian support/resistance, Parabolic SAR, Aroon Oscillator, and ZIG/ZAG moves for rallies and corrections exceeding 10%. The Mayer Multiple is current Bitcoin price divided by its 200-day moving average. The page calls a bounce at MM=1 a BUY signal, describes MM<1 as bearish and MM>1 as bullish, calls MM<0.6 a good cyclical BUY zone, MM>2.4 overbought, and MM>3.6 near a cyclical market top. The Puell Multiple is daily coin issuance in USD divided by its 365-day moving average. The Power Law Corridor uses regressions through three bubble peaks and daily price minus one standard deviation.
The trend-direction model compares daily BTC/USD OHLC corrections and rallies by magnitude and duration. The page claims 242 short-term events no smaller than 2% and 87 medium-term events no smaller than 13% over 12 years since 2010. It explicitly says not to use the probability model as an actionable signal; traditional technical analysis remains the action method and probabilities are context. These component claims are retained as related but distinct; no common aggregation rule is invented.
Native span, publication lag, and causal assumptions
The page claims daily BOCA, ETF-flow, Puell, and probability updates and a 30-minute technical-overlay update. It claims a probability history since 2010, but no source-linked Bitcoin numeric observations are captured. The four adjacent workbook captures are attached to the separate Excel Data page and cannot establish target native coverage or point-in-time linkage.
Article publication date and timestamp are undisclosed. The 2026-08-22 archive fetch is archival delay, not an economic-data publication timestamp; an article-to-fetch interval is not computable. Underlying release lag, calculation cutoff, timezone, vintage, and revision calendar are unknown. Any future overlay must use point-in-time information, then execute no earlier than the next available bar with at least one available-bar lag. Same-close/lookahead is unsupported. Unknown inputs remain unavailable rather than imputed, carried forward, or neutral.
Validation
Focused offline suite tests/test_recessionalert_bitcoin.py (7 passed) verifies source-body SHA agreement, the five crawl revisions, disclosed rule text and thresholds, all four adjacent workbook hashes without merging them, unknown publication lag, unavailable-state policy, late-input policy, zero target-page asset rows, and read-only database access. The executed notebook research/experiments/1361.ipynb reproduces the source and artifact hash assertions. Controller receipts check-1 and check-2 record the complete test module as temporal and behavioral evidence.
No numeric performance, subperiod stability, adjacent-parameter, false-signal, or market-outcome metric is claimed because no target series, common state, point-in-time releases, or independent outcome window is pinned.
Results
Public evidence-gap report only. Machine-readable source record: research/findings/specs/recessionalert-bitcoin.json. Experiment manifest: research/experiments/1361.json. Decision: insufficient_evidence; registry action: not_registered; playable: false; current value: unavailable. The safe result is published under Other / Research rather than as a playable Other-indicators card. No raw HTML, workbook, image, chart, nonce-bearing URL, credential, cookie, or private account data is redistributed.
Explicit gaps
- No source-linked machine-readable daily or 30-minute Bitcoin component series is attached to the target row.
- No verifiable native numeric span exists for BOCA, exchange balances, ETF flows, technical outputs, valuation, issuance, OHLC, or probability outputs.
- The Excel Data link does not identify a native workbook, sheet, column, or revision vintage for this row.
- The apparent 100–1,000 BTC ownership band is not disclosed.
- BOCA factor formulas, categorization, aggregation, and color-state thresholds are incomplete.
- Technical parameters and precedence for SuperTrend, Donchian, Parabolic SAR, Aroon, and ZIG/ZAG are undisclosed.
- Power Law regression sample selection, fit window, and update mechanics are undisclosed.
- Mayer/Puell source definitions, initialization, warm-up, timestamp treatment, and missingness are undisclosed.
- Probability-model event inclusion, database, calculations, out-of-sample boundary, and revision policy are undisclosed.
- Daily and 30-minute calculation/release clocks, information cutoffs, timezones, vintages, and publication lag are unknown.
- No common state machine, persistence, confirmation, hysteresis, unavailable transition, or aggregation precedence is disclosed.
- No target instrument, allocation, holding period, costs, fills, exit, re-entry, or cash behavior is disclosed.
- The source explicitly warns that the probability model is not an actionable signal.
- No performance or market-outcome metric can be evaluated without the target series and outcome window.
- No proxy, synthetic extension, chart OCR, unrelated workbook merge, third-party acquisition, or live refresh is authorized.
Web plan and explicit exception
Web plan:
- canonical section: Other / Indicators, with this non-promoted result discoverable under the Other / Research library;
- supported access path: static generated result/experiment pages and repository spec/tests;
- browser evidence: generated evidence table and discoverable result entry, with no fabricated signal controls, current value, or overlay statistics;
- static/local boundary: public pages explain the disclosed rules and limitations; local tests and notebook verify offline source binding;
- exception: a live/playable indicator UI would be misleading while target series, release timing, current value, common aggregation, and execution map are absent. Reconsider only if an authorized source refresh supplies those inputs.
Files changed
research/findings/1361.mdresearch/findings/specs/recessionalert-bitcoin.jsonresearch/experiments/1361.jsonresearch/experiments/1361.ipynbtests/test_recessionalert_bitcoin.pyresearch/reports/experiments.pyresearch/results/1361.md- Generated experiment/result pages and payloads under
docs/via canonical producers