Measure · Network profile

Bitcoin

Layer 1 · Proof of work

Decentralization Index

7.2out of 10

MeasureWeightScore
Nakamoto coefficient 25% 3.0
Full nodes 15% 9.7
Client diversity 10% 0.0
Hardware to run a node 10% 10.0
Token concentration 15% 9.7
Insider allocation at launch 15% 10.0
Control over upgrades 10% 10.0

How the index works · method v1.0

Insider allocation at launch

10.0Fair launch

No premine and no presale. Every coin was issued through open mining or staking.

Fair launch: no premine and no presale. The whitepaper (October 31, 2008) describes new coins entering circulation only as a reward to whoever creates a block, with no central issuer. Satoshi Nakamoto's v0.1 release announcement (January 8, 2009) let anyone generate coins with the software and set the starting difficulty so an ordinary PC could do so. The genesis block's 50 BTC reward can't be spent. Context: Satoshi mined many of the earliest blocks while few others were mining. Sergio Lerner's 2013 analysis of blocks 0 to 36,288 gave a rough, by-eye estimate that about 1 million BTC were mined by a single entity, which he believes is Satoshi. Because this was open mining that anyone could join on the same terms, our method counts it as public, not as an insider allocation. Vesting is 'none' because there is no insider allocation to lock up.

Sources: [1][2][3][4]

How we score insider allocation

All measures

MeasureValueAs ofHow we got it
Nakamoto coefficient
Fewest block producers who could together disrupt the network.
3
Proof of work, so this counts mining pools, by share of blocks found over mempool.space's one-month window, read November 3, 2026 (4,657 blocks, about October 3 to November 3, 2026). The window rolls forward, so a later reading gives slightly different counts. Foundry USA found 26.1% (1,217 blocks), AntPool 20.3% (947) and F2Pool 16.7% (778). The top two hold 46.5%, so three pools are needed to pass 50%. Individual miners can switch pools, so counting pools overstates concentration somewhat. Conflict: researcher b10c (April 2025, https://b10c.me/blog/015-bitcoin-mining-centralization/) and mainnet.observer (https://mainnet.observer/charts/mining-pools-antpool-and-friends/) report that several smaller pools appear to mine on AntPool's block templates (the list of transactions a pool tells its miners to work on). The pools they name include Binance Pool, SECPOOL, Luxor, Braiins, CloverPool and Poolin. Grouping those pools with AntPool gives that group 32.3% of blocks in this window. Together with Foundry, that is 58.5%, which would make the coefficient 2. That grouping rests on observed template similarity and isn't confirmed by the pools, so we report the per-pool figure.
November 3, 2026 Researched
Block producers
Validators, miners or pools that produce blocks.
19
Named mining pools and solo-mining services that found at least one block in mempool.space's one-month window, read November 3, 2026 (about 31 days, October 3 to November 3, 2026, slightly longer than our 30-day definition). This excludes the Unknown entry (18 blocks not attributed to any pool) and includes three solo entries with one block each (Solo CK, FutureBit Apollo Solo and Braiins Solo). The six largest pools found 83% of blocks. Several smaller pools may mine on AntPool's templates (see the Nakamoto coefficient note), so fewer independent parties choose which transactions go into blocks.
November 3, 2026 Researched
Full nodes
Independent computers that check every rule.
18.7K
Nodes in the crawler snapshot taken November 3, 2026 at 13:53 UTC (timestamp 1793714013, block 974,389) that the crawler validated (its total_nodes figure). The same snapshot found 19,033 reachable nodes before validation. We use the smaller, validated number. The bitnodes.io API now redirects to btcnodes.io. btcnodes.io lists Rodrigo Martínez as its contact (https://btcnodes.io/), and its crawler is his fork of the original Bitnodes code (https://github.com/brunneis/bitnodes), so it isn't run by Bitnodes' original creator. Coin Dance reported 18,346 public nodes the same day (https://coin.dance/nodes). Only reachable (listening) nodes are counted, so the true number of full nodes is higher.
November 3, 2026 Researched
Largest client share
Nodes running the most popular software. Lower is safer.
90%
Bitcoin Core ran 16,485 of 18,346 public nodes (89.86%) on Coin Dance, read November 3, 2026, with all Core versions counted together. Bitcoin Knots ran 1,832 (9.99%), and btcd, Bitcore and others 29 nodes (under 0.2%) combined. Bitcoin Knots' own documentation says its development 'generally takes place as part of Bitcoin Core' and is merged into each Knots release (https://github.com/bitcoinknots/bitcoin), so the two share most of their code and real software diversity is lower than this split suggests.
November 3, 2026 Researched
Hardware to run a node
What it takes to verify the network yourself.
Home computer
Category: a typical home computer. Bitcoin.org's recommended minimum for Bitcoin Core is 750 GB of disk, 2 GB of RAM, a desktop or laptop processor (or some ARM chips, the low-power processors used in phones and small computers, above 1 GHz), 500 MB a day of download and 5 GB a day of upload, after a one-time download of about 740 GB. With custom settings the page lists 7 GB of disk and 256 MB of RAM. Figures as shown for Bitcoin Core 31.0.
November 3, 2026 Researched
Who can change the rules
Whether one party can upgrade or pause it.
No single party
Bitcoin.org's FAQ says 'Nobody owns the Bitcoin network' and 'Bitcoin is controlled by all Bitcoin users around the world'. There is no foundation, company or admin key that can change the rules or pause the network; a rule change takes effect only if node operators choose to run software that enforces it. Caveat: Bitcoin Core's maintainers publish the software that about 90% of reachable nodes run (see client share), so in practice most upgrades are proposed and released through one project.
November 3, 2026 Researched
Held by top 100 addresses
Excluding known exchanges and contracts where possible.
11%
Computed from CoinCarp's Bitcoin rich list, read November 3, 2026 (the page gives no data date). The 100 largest addresses held 3,095,125 BTC. We excluded the 14 addresses CoinCarp labels as exchanges: Binance 8, Bitfinex 2, OKX 2, Crypto.com Exchange 1 and Gate 1. Together they held 869,351 BTC. That leaves 2,225,774 BTC, or 11.1% of the 20.105 million BTC in circulation on October 29, 2026 (https://api.blockchain.info/charts/total-bitcoins?timespan=10days&format=json). Limits: most top-100 addresses carry no label, so other custodians may remain in the figure. We also didn't replace the 14 excluded addresses with the next largest, which would add at most about 0.6 percentage points. CoinCarp labels no bridge, burn or contract addresses in the top 100. Including exchanges, the top 100 hold about 15.4% of circulating supply. CoinCarp's own percentages use the 21 million cap as the base (its headline top-100 figure is 14.77%).
November 3, 2026 Researched
Transactions per second
30-day average of user transactions, not consensus votes.
8.3
21,570,846 confirmed transactions over the 30 days from September 29 to October 28, 2026 (Blockchain.com, Confirmed Transactions Per Day), divided by 2,592,000 seconds. Bitcoin records no consensus or vote transactions, so nothing was excluded. The count includes one coinbase transaction (the transaction that pays the miner's block reward) per block (about 150 a day, under 0.03% of the total). Daily counts ranged from 535,739 to 913,196. Blockchain.com's daily series ran about six days behind the day we read it (November 3, 2026). So the window ends on October 28, 2026.
October 28, 2026 Researched
Daily active addresses
30-day average. Addresses are not people.
478.7K
Average of Blockchain.com's daily Number Of Unique Addresses Used from September 29 to October 28, 2026 (14,361,547 in total over 30 days). Blockchain.com describes it as the number of unique addresses used on the Bitcoin blockchain each day, which means addresses that appeared in a transaction. Addresses aren't people: one person can use many.
October 28, 2026 Researched
Typical fee
Median for a simple transfer over 30 days.
–
Left blank. We found no accessible source that publishes a 30-day median fee for a simple transfer in US dollars. BitInfoCharts and Coin Metrics, which publish a USD median, refused our requests, and Blockchain.com's USD fee chart is an average, not a median. For context only, mempool.space's block fee data for the month to November 3, 2026 (https://mempool.space/api/v1/mining/blocks/fee-rates/1m) shows median fee rates mostly between 0 and 2 sat/vB (satoshis, or hundred-millionths of a bitcoin, per unit of transaction size), rarely above 5. We didn't convert this to US dollars.
– –
Time to finality
Until a payment practically cannot be reversed.
~60 minutes
Bitcoin's finality is probabilistic: each new block makes a reversal less likely, but never impossible. We use the common convention of 6 confirmations, the default the Bitcoin Wiki describes and that it says most exchanges and merchants require, at the 10-minute average block interval: about 60 minutes. Real times vary because block intervals vary. The wiki suggests many more confirmations (144 blocks, about a day) for high-value irreversible sales. The wiki also says that, with mining as it is today, more than 6 confirmations are required. The first confirmation takes about 10 minutes on average.
November 3, 2026 Researched
Outages, last 24 months
Times the network halted or stopped producing blocks.
0
No network halts between November 2024 and November 2026. Bitbo's uptime tracker lists two downtime incidents in Bitcoin's history, on August 15, 2010 (the value overflow incident, CVE-2010-5139; a CVE is a public ID for a software security flaw) and March 11, 2013 (CVE-2013-3220), and shows 100% uptime for every year from 2014 through 2025 (99.93% in 2013, the year of its last listed incident). A news search found no reported halt of the main network in 2026. On August 8, 2026, nodes running BIP-110 software (a proposed one-year rule that would restrict non-financial data on the blockchain) split from the main chain at block 961,632. They followed a minority chain that produced two blocks and then stalled at block 961,633, while the main network kept producing blocks normally (The Block, https://theblock.co/news/ecosystems/2026-08-08-bitcoins-bip-110-supporters-split-onto-minority-chain-as-main-network-pulls-ahead-411213; Decrypt, https://decrypt.co/375177/bitcoin-anti-spam-fork-splutters-halt-mining-just-two-blocks). We don't count it as an outage. Bitbo's 'days since last downtime' counter showed an out-of-date figure when we read it, so we relied on its incident list.
November 3, 2026 Researched
New coins per year
Yearly issuance as a share of supply.
0.8%
The block subsidy has been 3.125 BTC since block 840,000 (April 20, 2024). The next halving, to 1.5625 BTC, is at block 1,050,000. The Bitcoin Wiki's schedule gives 164,062.5 new BTC for the current 52,500-block period (from block 945,000, reached April 14, 2026), a 0.82% increase. Cross-check: 164,062.5 BTC against the 20.105 million BTC in circulation on October 29, 2026 (Blockchain.com) is also 0.82%. Blocks came slightly faster than every 10 minutes over the past month (about 150 a day, per mempool.space), which raises actual issuance a little.
November 3, 2026 Researched

See how Bitcoin stacks up

Compare it side by side with up to three other networks.

Comparing networks isn't advice to buy their coins. Spotted an error? Email us with a source and we'll correct it.