Currently 62.0% of the evaluable supply sits in wallets holding over 100 BTC (sharks, orcas and whales), in the most recent snapshot (October 25, 2026, block 973,007).
Snapshot computed directly from our own Bitcoin node — not a third-party API.
“Addresses holding more than 100 BTC hold 62.0% of bitcoin supply (UTXO set snapshot of 25 Oct 2026, block 973,007).” NodeWitness, https://nodewitness.com/en/charts/bitcoin-holder-distribution-chart/
Journalist? Ready-to-use images and CSV in our press kit.
Each weekly snapshot of the full UTXO set, ordered by the block height it corresponds to — not by the calendar day it was generated: since the series was rebuilt by rolling the UTXO set backward in time, the two orderings don't always match. Hover over the chart for the detail of each snapshot; click a cohort in the legend to isolate or hide it. The dashed white line is BTC's market price on the real date that height corresponds to (right axis, log scale) — useful for reading the colors against the price cycle.
Across the 59,879,495 addresses holding any bitcoin in the latest snapshot, the average balance is 0.34 BTC. That average is misleading: 58% of those addresses hold less than 0.001 BTC and 93% hold less than 0.1 BTC, so a typical address holds far less than the average. And an address is not a person: one holder can spread coins across many addresses, and a single exchange or ETF address can hold coins for thousands of people.
Each band is a cohort of addresses grouped by balance, from dust up to whales, and its height is the share of supply that cohort holds. Week-to-week wiggles mean little. Trends over months do: small and mid-sized cohorts gaining share means supply is spreading out, large cohorts gaining share means it’s concentrating.
Addresses aren’t people, and that matters more here than anywhere. Exchanges and ETF custodians pool thousands of customers into a handful of giant addresses, which inflates the top cohorts. A single holder can just as easily split coins across a hundred addresses. A sudden jump in the whale band is far more often a custodian reshuffling wallets than a whale buying.
Ranking addresses by balance goes back to the early Bitcoin “rich lists” of the 2010s. The shrimp-to-whale naming came later, as on-chain data providers turned those lists into cohorts. The underlying question is unchanged: is bitcoin concentrating or dispersing?
The live question since January 2024 is whether US spot ETFs and other large custodians are pulling supply into the top cohorts. Our weekly series, which starts in 2018, lets you follow that directly. Just remember that many of those big addresses hold coins on behalf of a crowd, not one owner.
Groups every address with a balance by how much BTC it holds in total, from dust (under 0.001 BTC) to whales (over 10,000 BTC). It doesn't say who owns each wallet, only how much BTC it controls.
Weekly, we take a full snapshot of the UTXO set from our own Bitcoin Core node — complete, unpruned, run by us, not a third-party API. UTXOs are grouped by address (same script = same address, no need to decode it) and each address's total balance is classified into the nine cohorts shown above.
It runs from January 12, 2018 to today: 455 weekly snapshots, rebuilt by rolling back the UTXO set to past block heights. About 4 scattered weeks within that range are still unprocessed: processing them is paused, with no date yet, documented in our Score version history. And nothing before 2018: extending further back would need more rollbacks, with no date set yet.
The change in supply share held by large cohorts (sharks and whales) is one of two sub-signals of our own node's data, the single highest-weighted input to the NodeWitness cycle Score. See the full methodology in how the Score works.
On this chart, an address holding more than 10,000 BTC. 1,000 to 10,000 BTC is usually called an orca or small whale, and anything over 100 BTC counts as a large wallet. These are industry conventions, not official categories, and data providers draw the lines slightly differently.
No. It counts funded addresses, not people. One person can control many addresses and a single exchange address can hold coins for thousands of customers, so address counts don’t convert into user counts.
The methodology and what concentration actually means are in how we measure Bitcoin holder distribution. Part of the largest wallets are exchanges, so read it alongside exchange flows and reserves, and compare each period with the monthly Score history.