Benchmark Series · Methodology

How the benchmark series is built

Two free, recurring publications, the CCP Margin Monitor and the Stablecoin Reserve Scorecard, recompute figures from disclosures each subject already publishes, and attach a hash-anchored receipt to every number. This page states the source corpus, the derivation formula, and the receipt format for each published metric, and is versioned so a later formula change never retroactively re-narrates an older issue as wrong.

Derived, Never Mirrored Hash-Anchored Receipt Issue #1 Free Zero PII
๐Ÿ”’ All inputs are processed locally in your browser. No data is transmitted. Do not enter real personal data โ€” use synthetic or anonymised inputs only.
What this series is not: it recomputes a figure from fields a CCP or issuer has already disclosed publicly. It does not audit whether the underlying disclosure itself is truthful, and it does not redistribute the source disclosure table. Every published number is a computed derivation, cited back to its source field, never a copy of it.
Deliverable A

CCP Margin Monitor

Per-CCP, per-quarter derived figures (initial-margin-to-notional ratios, margin-model coverage deltas quarter-over-quarter, default-fund-to-IM ratios) recomputed from each CCP's own published CPMI-IOSCO Public Quantitative Disclosure (PQD), plus the CCP Global aggregator (ccp-global.org/pqd). Both source corpora are already public and free; nothing in this series pays for or resells access to them.

Where the underlying comparator kernel has landed, the derive step is a chain that consumes it directly and emits a receipted composite. Where a specific figure isn't yet covered by an automated kernel, the issue states so explicitly ("hand-extracted from CCP PQD, pending kernel automation") rather than silently presenting it as machine-derived. Cadence target: quarterly, aligned to CCP PQD publication windows (CCPs typically publish about 60 days after quarter-end). This is a target, not a publication-date guarantee.

Deliverable B

Stablecoin Reserve Scorecard

Per-issuer, per-month derived figures (reserve composition ratios, T-bill-yield-implied income compared against published reserve income, attestation-to-attestation drift flags) recomputed from public monthly reserve attestation reports. The scorecard rides the same proof-of-reserves verification (Merkle-SUM) and reserve-disclosure-check kernels used elsewhere in the suite, reading each attestation's published fields as caller-supplied input, never a hardcoded rule table.

The T-bill-implied income figure is independently computed from a live/dated T-bill rate input, and is stated as a sanity check against the issuer's own published reserve income, not an assertion that the issuer's figure is wrong. Where commentary references a compliance-cost figure sourced to a single journalism cluster rather than a regulator, auditor, or issuer filing, it is cited inline by publication name and never presented as a computed number; the Scorecard's own kernels do not emit it. Cadence target: monthly, aligned to attestation publication.

Receipt Format

What ships with every published number

Every figure in every issue carries the same six fields, verifiable offline with the standard's existing verifier. No new verification surface for this series.

FieldWhat it states
SubjectThe CCP or stablecoin issuer name (public-disclosure subject, not a private engagement client).
Disclosure periodThe PQD quarter or attestation month the figure was computed from.
Source field(s)The specific disclosure field(s) the figure derives from.
Computed valueThe derived figure itself: a ratio, a delta, or a drift classification, never a copy of the source table.
execution_hashCommits to the exact input/output pair; any third party can replay the kernel and get a byte-identical result.
verify_urlLinks to the standard's offline-verifiable receipt check for that figure.
Versioning

Methodology changes never rewrite history

This page carries a version string. When a derivation formula changes, the version bumps and the change is noted here. Every published issue cites the methodology version it was computed under, so a later formula revision doesn't retroactively re-narrate an older issue's figures as wrong; it was correct under the methodology in force when it published.

Methodology version: v1.0 (2026-08-03): initial publication, CCP Margin Monitor + Stablecoin Reserve Scorecard.

No coverage percentage and no cadence SLA are stated anywhere in this series: cadence is a target, and each issue states its own "as of" scope honestly rather than claiming completeness it hasn't earned.

Related

Explore adjacent tooling