Research Changelog

Dated changes to BSAHI's measurements, grades, and methods. Corrections are recorded here with their cause. Newest first.

2026-09-17

• Integrity audit added (tools/research/integrity_audit.py): heights/dates monotonicity, anchor consistency, unit agreement, provenance, layer classification. Runs as a gate.
• Units correction (Observed→corrected): production_cost_ratio.py stored the hash-rate series (unit TH/s) under the field hashrate_ehs — a 10⁶ mislabel. The power calculation was correct; the label was not. Now stores hashrate_ths and a derived hashrate_ehs, and the audit asserts the relation.
• Evidence Matrix published (/research/evidence-matrix): 27 headline numbers classified Observed / Reconstructed / Modelled with source, date and grade.
• UTXO measurement pipeline added (utxo_state_measure.py) and a first Observed gettxoutsetinfo row captured (h671,462 · 72,234,156 UTXOs · 4.39 GB). The 2013-2026 UTXO table remains Reconstructed (grade D).
• SCCR hardening: exact sensitivity bands over N×T×C (below 1.0 in 22/27 combinations; N dominates the uncertainty at 10×), a bootstrap 95% CI for the mean of 155 real blocks (0.2131–0.2707), and a named-archetype electricity scenario table. Published at /research/sccr-sensitivity.
• Reproduction verified: JS = Python = C on the frozen capture (155/155 heights, max |diff| 4.9e-7); recorded in data/reproduction_verification.json.
• Per-block strip renamed to per-block SPOT validation with an explicit sample-size caveat; sample size increased from 3 to 12 blocks per era.
• Methodology paper v1.0 published (/research/methodology); versioned dataset index (/data/datasets.json) and replication invite (/research/replicate) added.

2026-09-16

• Historical series frozen: difficulty 2009→2026 (1,614 points) and mempool congestion 2016→2026 (1,498 points) from primary sources; row 10 (hashrate) B→A, row 3 (congestion) B→B*.
• Production-cost scenarios: the single assumed electricity price replaced by a scenario table across $0.03–$0.15/kWh.
• Per-block spot validation strip published for 2010–2026.

2026-09-15

• UTXO series reachability documented (no continuous public source found; grade D recorded rather than estimated); VCI chain leg cross-checked to 2.7%.
• Pool hashrate reachability documented; attribution validated internally 7/7.

2026-09-10

• BIP-110 Post-Lock-In Case Study published (1,862 words, primary-source chronology); author byline corrected to the Independent Researcher (not a council).

2026-09-09

• UTXO cost ratio (reconstructed era table) published; grade D recorded.

2026-08-21

• Regime event observed: SCCR rose ~0.16 → ~0.44 in ~five days, measured live.

2026-08-11

• SCCR trend note: the ratio fell from ~0.28 to ~0.238 across the first measured week (early series; uncertainty stated).

2026-08-02

• SCCR measurement begins. First live readings; the daily tracker installed.

Earlier

• Model correction (pre-launch): a 10× time-horizon error in our own storage-cost model was found and fixed; every dependent value was regenerated and the conclusion survived. This precedent is why corrections are published here.

← All research · ← Back to Learn