Galaxy Research’s October ownership map places about 11.1 million bitcoin, or 55.1% of roughly 20.1 million mined BTC, in an unattributed bucket. The number is real within Galaxy’s methodology. The common reading of it, that most bitcoin therefore sits in individual self-custody, goes further than the data can prove. coindesk.com
The distinction is consequential. An address can be visible on Bitcoin while its beneficial owner remains unknown. The unattributed group can contain individual cold wallets, early miners, lost coins, unidentified institutions, over-the-counter desks and custody structures spread across many addresses. It measures the limit of current labeling, not a clean demographic category.
Galaxy’s snapshot, reported at block 971,382, assigns 17.9% of supply to exchanges, 6.0% to ETFs and funds and 5.6% to the Satoshi-era “Patoshi” pattern. The remaining named and miscellaneous categories account for 15.4%. Galaxy head of research Alex Thorn argued that the residual still indicates substantial self-custody, while also acknowledging that the precise owner mix cannot be observed directly.
more than half of all BTC supply is sitting in addresses with no known attribution
despite all the institutionalization and centralized ownership over the years, still a ton of bitcoin (maybe even most) is held by individuals in self-custody 🔐 pic.twitter.com/OYGFMuQVvQ
— Alex Thorn (@intangiblecoins) October 14, 2026
Institutional holdings are easier to underestimate than the top-line chart suggests. In a separate October report, Galaxy calculated that digital-asset treasury companies alone hold at least 1.073 million BTC, including 848,500 at Strategy and roughly 120,000 across Metaplanet, Twenty One and Strive. Holdings can be attributed at the company level without every operational address being labeled immediately, and service providers can reorganize wallets without changing beneficial ownership. Galaxy
The methodological trap is temporal. If an analytics firm labels a previously unknown cluster tomorrow, the unattributed share falls even when no bitcoin moves. Conversely, a transfer from a labeled exchange to a fresh address can enlarge the unknown bucket without proving that retail ownership increased. Any trend claim must separate actual transfers from improvements or regressions in entity attribution.
The snapshot still provides useful information. At minimum, public researchers cannot confidently connect more than half of mined supply to a named entity. It also shows that the institutionalization of Bitcoin through ETFs, public companies and custodians has not made ownership fully legible. What it does not establish is how much of that opaque supply is active, spendable, liquid or controlled by individuals.
TOKEN RECON ASSESSMENT
The 55.1% figure is best treated as a ceiling on known ownership, not a census of self-custody or a proxy for sell-side liquidity. The strongest future confirmation would be a consistent time series that decomposes changes into on-chain transfers and newly labeled clusters. Without that split, a shrinking unknown bucket could reflect better analytics rather than institutional accumulation, while a growing one could reflect address rotation rather than new retail custody.
Sources
CoinDesk: Over 11 million bitcoin has no known owner
Alex Thorn, Galaxy Research, original X post
Galaxy Research: Bitcoin treasury companies hold at least 1.073 million BTC
