Active official Bitcoin adoption
A jurisdiction counts in the default series only when it has at least one verified, currently effective government policy or program that explicitly involves Bitcoin and creates a meaningful government role in holding, purchasing, accepting, protecting, mining, or formally recognizing it.
Introduced bills, speeches, nonbinding resolutions, generic cryptocurrency regulation, private-sector activity, and passive custody of seized Bitcoin do not qualify by default.
Event-sourced government series
Each jurisdiction is reconstructed from ordered events. Effective policies activate one or more categories; pauses, repeals, expirations, or withdrawals deactivate them. A jurisdiction is counted once even when several categories are active. Corrections supersede incorrect events while retaining audit history.
Ownership confidence
- Tier A: a representative survey asks specifically about Bitcoin ownership.
- Tier B: Bitcoin ownership is derived within one compatible survey and sample.
- Tier C: a model combines unrelated sources and is hidden by default.
Tier is a measurement-method classification. The separate low, medium, or high review confidence records the Atlas review judgment; it is not a statistical confidence level. A numerical confidence interval appears only when the source reports a compatible one.
Ownership population scope and owner counts
The public comparison currently requires a total-resident-population survey universe and binds every estimate to the exact World Bank total-population observation for the declared year. Adult-only, internet-user, investor-only, customer, account, or transaction samples cannot be multiplied by total population to create a public owner count.
When scope is compatible, estimated owner count is the reviewed ownership percentage multiplied by that retained population observation and rounded to a whole person. It is a derived estimate—not an observed headcount—and compounds uncertainty in both inputs.
Ownership comparability, rights, and corrections
Historical estimates are grouped only inside an explicitly declared methodology series. Estimates from different questions, samples, population scopes, or derivation methods are not joined into a continuous trend. Tier C models remain hidden unless a visitor explicitly enables them.
Publication requires a Bitcoin-specific measure, representative design, source and raw manifest lineage, documented fieldwork and weighting, and reviewed permission for commercial transformed republication. Corrections append a new record, supersede the old value, and preserve both directions of the correction history.
Addresses, wallets, and people
An address is not a wallet, person, or legal entity. A person can control many addresses, and a custodian can control addresses for many people. Address counts are described only as address activity unless a licensed entity-adjusted methodology supports another claim.
Supply last active at least one year
The on-chain series groups spendable Bitcoin transaction outputs by age using block timestamps. An output qualifies once it is at least 365 days old. The BTC value is the cohort's supply; the percentage divides it by spendable UTXO supply.
This is supply dormancy, not a holder count. Same-owner transfers reset the moved output's age, while lost coins can remain old indefinitely. It is also separate from entity-adjusted long-term-holder models that use an approximately 155-day probabilistic threshold.
Illustrative fixtures
Fixture mode exists only to review application behavior before credentials or verified data are available. Fixture responses are marked illustrative, never mixed with canonical data, and blocked from Vercel production.