Data definitions
Every metric ReserveBeat publishes is either a figure stated in a source document, rendered verbatim, or a value computed from stated figures. This page gives the exact computation behind each computed metric — its inputs, its arithmetic, its rounding, and what the result does not mean — so a reader can re-derive any of them from the same sources.
Computed
Computed metrics
Figures read from an issuer's attestation report or measured from a public blockchain are never adjusted, netted, or recomputed: they render exactly as stated or as measured. Only the metrics listed here are computed, and each one is labeled as computed wherever it appears.
- Attested coverage ratio
- The total reserve figure an attestation report states, divided by the tokens-outstanding figure the same report states at the same as-of date, quantized to six decimal places with banker's rounding. Only a single-token-scope report produces one. A document stating just one of the two sides yields no ratio at all rather than a half-computed one, and no on-chain measurement ever enters the division.
- All-network coverage ratio
- A different figure from the attested coverage ratio, which a token page shows only where it publishes per-network supply: the same stated total reserve divided by the on-chain supply ReserveBeat itself measures across every published network nearest the report's as-of date, quantized to six decimal places with banker's rounding. Unlike the attested ratio it does contain an on-chain measurement, and no ratio renders when any published network has no reading near that date.
- Attested reserve assets tracked
- The exact sum of the total reserve figure each tracked token's latest active attestation report states. The addition runs over string decimals brought to a common scale, so no floating-point value ever touches a reserve figure, and a single unreadable input refuses the whole sum rather than publishing a partial one. As-of dates differ by issuer, a consolidated-scope report counts once at its stated scope, and a token with no attested series adds nothing. The homepage prints the sum split by scope: the part from single-token reports and the part from consolidated group reports, each with the tokens it covers.
- Attestation status
- The age of a token's most recent report month, measured against the cadence expected for that token — the cadence recorded for its attestation source, or the default cadence when none is recorded. Past that limit the status reads STALE; within it, CURRENT. Past it by a further stated withholding grace, the status reads WITHHELD_STALE: the board stops serving that token's report-derived figures and states the withholding instead, while its on-chain supply still renders. By default the limit is forty-five days after the report month ends for a monthly cadence or when no cadence is recorded, one hundred and six days for a quarterly cadence, one hundred and ninety-eight days for a semiannual cadence and three hundred and eighty days for an annual cadence: each cadence's own period plus the same fourteen days of publication grace. The withholding grace is a further thirty days by default, so the figures are withheld past seventy-five days for a monthly cadence, one hundred and thirty-six days for a quarterly cadence, two hundred and twenty-eight days for a semiannual cadence and four hundred and ten days for an annual cadence. Every token card and token response states its own two limits, as stale_after_days and withheld_after_days. A token with no attestation series at all reads NONE, which is an absence rather than a value. The status states age, not lateness: an issuer publishing on its usual schedule can read STALE while its next report is not yet due, which is what overdue answers.
- Overdue
- Whether a token's next attestation report is late against the token's own schedule. The next report is expected by the end of the reporting period after the latest report month, stepped by the token's expected cadence, plus the publication lag that latest report itself showed. Overdue reads true once today's UTC date is past that expected-by date plus a stated grace window with no newer report held, and false until then. Held means captured by this board, so the grace window also absorbs the days between an issuer publishing and the board capturing. It is null when no expected-by date can be built: a token with no attestation series, or a latest report that states no issue date to measure a lag from.
- Publication lag
- The number of days between the latest as-of date a report states and the issue date the same report states. Both dates come from the document; neither is inferred.
- Capture lag
- The number of days between a report's stated issue date and the moment ReserveBeat captured and hash-anchored the document. It measures this site's own latency, not the issuer's.
- Disclosure score
- A count, not a rate: how many dimensions of a fixed, versioned rubric a published disclosure meets, rendered as the number met over the number of dimensions with the rubric version beside it. Every dimension records the stored fact it was decided from. A report that has never been scored renders no score rather than an assumed one.
- Reconciliation delta
- An attested figure set against the token's own on-chain supply, read at the last block at or before the time the report states its figures are as of. Where the document itemizes the token's own network, the delta is the supply read there minus that itemized figure, with its ratio to the attested figure quantized to six decimal places against a published tolerance. Where the document states only a total, the supply read on the token's own network is bound-checked against that total instead of differenced, with that network's share of the total beside it. Without a read at that block, ReserveBeat's recorder snapshot within fifteen minutes of the stated time stands in, and the row says which. Where neither exists — the report states no time of day, the read has not landed yet, or the read was refused — the row states the absence and the reason, never a pass.
- Regulator-filed divergence
- The absolute difference between the net-asset figure an SEC-filed N-MFP report states and the fund figure it is compared against, divided by the filed figure and expressed in basis points to two decimal places, against a per-issuer tolerance. A filing stating zero net assets yields no divergence figure, and none is stated when no comparison ran (NOT_APPLICABLE, WITHHELD_STALE). The comparison adjusts neither side and enters no ratio.
- Board counts
- Every count the board states — the tokens on the board, the board filter's All, Current and Stale, and the attestation reports parsed — is a count of records held: never sampled, projected, or rounded. Current and Stale follow each token's attestation status; a token whose status is neither shows under All only.
Stated
Rendered as stated
- Stated figures
- Total reserve and tokens-outstanding figures, every reserve composition row, as-of and issue dates, accountant names, and engagement descriptions all render exactly as the source document states them. On-chain supply is measured rather than derived: read from the token's public contract at a quorum of independent providers, carrying the block number and timestamp of the reading.
- Precision and rounding
- A stated figure keeps every digit its document states — ReserveBeat never rounds a source figure for display. In the API every stated amount is written as a decimal string at a fixed scale of six decimal places, so a whole-dollar figure gains six zeros after the point while its value is unchanged, and a line the document prints as a dash or as a zero amount is served as zero — only where the document states that line, since a value it does not state is null. Where a computed value must be rounded, the precision is fixed at six decimal places with banker's rounding, applied identically on every page, in every export, and in the API.
Trust