This article explains the Nakamoto Coefficient in plain English. Rankings shift as validator sets and mining pools change. Nothing here is a recommendation to buy, sell, or stake any asset. Always do your own research.
Every blockchain claims to be decentralized. The question is whether you can actually measure it. The Nakamoto Coefficient is the most widely used single number for that job: it tells you how many independent parties would need to work together to take control of a network.
This guide covers four things: what the coefficient is, how to calculate it step by step, the top 10 blockchains on Chainspect as of July 7, 2026, and a deeper look at Internet Computer (ICP) — the chain most PoS-only trackers miss.
What Is the Nakamoto Coefficient?
The Nakamoto Coefficient was introduced in 2017 by Balaji Srinivasan, then CTO of Coinbase. He named it after Bitcoin's pseudonymous creator because it measures the same property Satoshi cared about: how hard is it for a small group to seize control?
In plain terms, the coefficient is the minimum number of independent entities that must collude before they can:
- Censor transactions (refuse to include certain transfers)
- Halt the chain (stop blocks from finalizing)
- Rewrite history (on Proof-of-Work chains with majority hashrate)
Higher number = more decentralized. A chain with a coefficient of 40 needs 40 separate organizations to coordinate an attack. A chain with a coefficient of 3 needs only three. That is a very different risk profile, even if both chains process billions of dollars in volume.
Important caveat upfront: the Nakamoto Coefficient is a useful starting point, not a complete audit. It does not measure governance, token ownership, client software diversity, or where validators physically sit. We cover those limits at the end.
How Is It Calculated?
The concept is simple. The data collection is the hard part. Here is the process in four steps.
Step 1: Pick the control threshold
Different consensus models use different danger zones:
- Proof-of-Work (Bitcoin, Litecoin): 51% of hashrate. Whoever controls a majority of mining power can rewrite the chain.
- Proof-of-Stake (most modern chains): 33% of staked tokens. A third of the stake is enough to halt finality or censor transactions on BFT-style networks.
Step 2: List every controlling entity
Do not count raw validator nodes. Count who actually controls the stake or hashrate. That means aggregating by:
- Mining pools (for PoW)
- Staking providers like Lido or Coinbase (for PoS)
- Exchange-run validators
- Independent node operators running their own infrastructure
A chain with 2,000 validators might have only 15 real operators behind them. Entity-level data is what matters.
Step 3: Sort largest to smallest and add up
Rank entities by their share of total stake or hashrate. Start adding from the top until you cross the threshold. Count how many entities it took.
Worked example (Proof-of-Stake, 33% threshold):
- Entity A: 12% of stake → running total: 12%
- Entity B: 9% → running total: 21%
- Entity C: 7% → running total: 28%
- Entity D: 5% → running total: 33% ✓
It took 4 entities. The Nakamoto Coefficient is 4.
Step 4: Use consistent, public data
Trackers like Chainspect and Nakaflow do this continuously using on-chain staking data and known operator mappings. Figures in this article come from Chainspect as of July 7, 2026, using entity-level aggregation at the 33% threshold for Proof-of-Stake chains (and 51% for Proof-of-Work where noted).
- List entities by stake or hashrate share, not raw node count
- Add from the top until you hit 33% (PoS) or 51% (PoW)
- The count of entities at that point is the Nakamoto Coefficient
- Higher = better. More collusion required to attack
Top 10 Blockchains by Nakamoto Coefficient (July 2026)
The ranking below covers the ten highest-scoring major networks on Chainspect as of July 7, 2026. Unlike Nakaflow, Chainspect includes Internet Computer (ICP) alongside standard Proof-of-Stake chains. Bitcoin and Ethereum are discussed separately below because they sit outside this top 10 on Chainspect's current snapshot.
Two Rankings Worth Scrutinizing Before You Trust the Table
1. TON at 79 — the headline number needs context. A coefficient of 79 means Chainspect counts 79 separate entities before reaching 33% of stake. That is more than triple Avalanche in second place. TON's validator set is genuinely broad — hundreds of independent operators run nodes, and stake is fragmented across many small delegators rather than a few liquid-staking giants like Lido on Ethereum. But three caveats apply:
- Telegram adjacency: TON's growth is tightly linked to Telegram's 900M+ user base. That is distribution, not necessarily the same as organic, arms-length decentralization. The TON Foundation and early ecosystem players still hold meaningful influence even if no single entity crosses 33%.
- Younger network: TON mainnet launched in 2018 but saw its major growth spurt in 2023–2025. Institutional stake and exchange-run validators may concentrate over time, pulling the coefficient down — as happened on Solana and Sui during their scaling phases.
- Methodology sensitivity: Chainspect's entity mapping may count more TON operators as independent than other trackers do. Nakaflow does not currently list TON, so cross-checking is impossible — treat 79 as a Chainspect-specific reading, not an industry consensus.
TON likely deserves the #1 slot on this dashboard, but 79 is not a settled fact — it is the best available snapshot from one tracker on one date.
2. Polkadot is missing from the top 10 — and that is almost certainly a data error. On Nakaflow, Polkadot consistently scores 42, the highest of any major chain, thanks to Nominated Proof-of-Stake spreading stake across ~600 algorithmically selected validators. Chainspect currently shows Polkadot at 1 — the same reading as Base and Arbitrum — which implies a single entity already exceeds 33%. That contradicts years of on-chain data and Polkadot's own validator-set design.
The most likely explanation: Chainspect's stake feed for Polkadot returned incomplete or null data on July 7, 2026, causing the calculator to default to a minimum count. Chainspect previously rated Polkadot at 176 in August 2025 community reporting — the opposite extreme. If you care about Polkadot specifically, trust Nakaflow's 42 over Chainspect's 1 until the feed is fixed. We list Polkadot in honourable mentions below with both numbers.
General rule: Different trackers produce very different numbers for the same chain. Solana reads 18 on Chainspect vs 10 on Nakaflow. Always check which data source, entity definition, and threshold a ranking uses before comparing chains.
| Rank | Blockchain | NC | Consensus |
|---|---|---|---|
| 1 |  TON (TON) | 79 | Proof-of-Stake |
| 2 |  Avalanche (AVAX) | 24 | Proof-of-Stake |
| 3 |  Sui (SUI) | 19 | Proof-of-Stake |
| 4 |  Solana (SOL) | 18 | Proof-of-Stake |
| 5 |  Cardano (ADA) | 15 | Ouroboros PoS |
| 6 |  Aptos (APT) | 14 | Proof-of-Stake |
| 7 |  Internet Computer (ICP) | 14 | Proof-of-Useful-Work |
| 8 |  Tezos (XTZ) | 13 | Liquid Proof-of-Stake |
| 9 |  TRON (TRX) | 13 | Delegated Proof-of-Stake |
| 10 |  Algorand (ALGO) | 12 | Pure Proof-of-Stake |
What Each Blockchain Is (Top 10 Profiles)
1. TON (TON) — Nakamoto Coefficient: 79
What it is: The Open Network, originally rooted in Telegram's blockchain ambitions and now governed by the TON Foundation and community validators. TON is a fast Layer-1 integrated deeply with Telegram — mini-apps, in-chat payments, on-chain games, and USDT transfers at scale. It is one of the few chains where user growth came from an existing messenger audience rather than crypto-native onboarding alone.
Why it ranks first: Chainspect counts 79 independent entities before anyone reaches 33% of stake. TON uses a Proof-of-Stake model with a large validator pool and no dominant liquid-staking layer equivalent to Lido. Stake is spread across hundreds of smaller operators and delegators, which inflates the entity count relative to chains where three staking providers control most of the network.
The asterisk: A score of 79 does not mean TON is 3× more decentralized than Avalanche (24) in any absolute sense. It means Chainspect's entity mapping finds more independent operators on TON than on any other chain it tracks. Telegram's ecosystem influence, the network's relative youth at scale, and the absence of a second tracker to cross-check all warrant healthy skepticism. See the caveats section above.
Context: TON is the headline story of this ranking. Whether 79 holds as institutional capital and exchange validators enter the set is the key question for 2026–2027. For a full explainer on the chain and its Telegram integration, see our TON blockchain guide.
2. Avalanche (AVAX) — Nakamoto Coefficient: 24
What it is: A high-speed Layer-1 focused on sub-second finality and custom subnets. Each subnet can have its own rules, making Avalanche popular for gaming, finance, and enterprise use cases that need dedicated environments.
Why it scores well: Roughly 620 validators share stake across the primary network, and entity-level analysis shows no single operator dominates. Twenty-four parties must collude to reach 33%, a consistently strong reading.
Context: Avalanche has held a top-tier decentralization score for years. It combines real on-chain usage with one of the healthiest distribution profiles among high-throughput L1s.
3. Sui (SUI) — Nakamoto Coefficient: 19
What it is: A Move-language Layer-1 designed around "objects" rather than accounts. Built by ex-Meta engineers, it targets gaming, social, and DeFi apps that need fast, cheap transactions with an intuitive programming model.
Why it scores well: Stake is distributed across enough independent validators that 19 entities must collude to reach 33%. Sui has moved from experimental to established in the L1 race.
Context: Sui sits in the upper-mid tier alongside Aptos (its Move-language sibling) and well above exchange-linked chains like BNB.
4. Solana (SOL) — Nakamoto Coefficient: 18
What it is: The highest-throughput major Layer-1, processing thousands of transactions per second. Solana powers a large share of on-chain retail activity outside Ethereum, with deep DeFi, NFT, and memecoin ecosystems.
Why it scores 18: Solana runs hundreds of validators, but entity-level analysis on Chainspect shows 18 organizations must collude to reach 33%. That is meaningfully better than Nakaflow's reading of 10 for the same chain, illustrating how tracker methodology changes the picture.
Context: Solana's decentralization is debated constantly. Critics point to hardware requirements and past outages. Supporters note the validator set has matured. Chainspect's 18 puts it in the upper half of this ranking, not the bottom.
5. Cardano (ADA) — Nakamoto Coefficient: 15
What it is: A research-driven Layer-1 using the Ouroboros Proof-of-Stake protocol. Cardano emphasizes peer-reviewed upgrades and a large global community, with on-chain governance through its Voltaire era voting system.
Why it scores well: Cardano has more than 2,000 stake pools, and its protocol design encourages many smaller operators rather than a handful of large ones. The entity-level coefficient of 15 reflects that spread.
Context: High pool count does not automatically mean high decentralization (pools can share operators), but Cardano's score has held steady in the mid-teens for years, which is respectable for a top-10 market-cap chain.
6. Aptos (APT) — Nakamoto Coefficient: 14
What it is: Another Move-language Layer-1, launched by former Diem (Facebook Libra) engineers. Aptos focuses on parallel transaction execution and low latency, competing directly with Sui for the high-performance smart-contract market.
Why it scores well: Fourteen entities must collude to reach the 33% threshold. That puts Aptos in the same upper-mid band as Sui and ICP.
Context: Aptos launched with significant venture backing, which often concentrates early stake. A coefficient of 14 suggests the validator set has diversified beyond the initial insider cohort.
7. Internet Computer (ICP) — Nakamoto Coefficient: 14
What it is: A decentralized compute platform — not a standard financial ledger. Applications run entirely on-chain as "canisters" on node machines hosted by independent providers worldwide. Chainspect is one of the few major dashboards that tracks it; Nakaflow does not.
Why it scores 14: Fourteen node-provider entities must collude to reach 33% on Chainspect's Proof-of-Useful-Work model. That ties it with Aptos and lands it squarely in the top 10. See the dedicated ICP section below for how infrastructure decentralization differs from NNS governance decentralization — the numbers are not the same.
Context: If you hold or follow ICP, rank #7 on a mainstream dashboard is meaningful. But the full picture requires looking at DRE, geographic spread, and the NNS governance coefficient separately.
8. Tezos (XTZ) — Nakamoto Coefficient: 13
What it is: A self-amending Layer-1 where protocol upgrades are proposed and voted on by "bakers" (validators) directly on-chain. Tezos has been live since 2018 and is known for formal verification tooling and institutional partnerships in art, finance, and gaming.
Why it scores well: Thirteen independent bakers must collude to reach 33% of stake. Tezos's liquid proof-of-stake model lets token holders delegate without locking funds, which has supported a broad baker ecosystem over time.
Context: Tezos does not dominate DeFi volume headlines, but its mid-teens coefficient and long operational history make it a steady decentralization story.
9. TRON (TRX) — Nakamoto Coefficient: 13
What it is: A high-throughput chain focused on stablecoin transfers and entertainment dApps, with a delegated proof-of-stake model using 27 "Super Representatives" elected by TRX holders.
Why it scores 13: Despite the small SR count, Chainspect's entity-level stake aggregation shows 13 parties must collude to reach 33%. That reflects how stake is distributed behind the visible validator layer.
Context: TRON processes enormous stablecoin volume (especially USDT). Its decentralization score is better than many assume, though critics still point to governance concentration around the founding team.
10. Algorand (ALGO) — Nakamoto Coefficient: 12
What it is: A Pure Proof-of-Stake chain designed by Turing Award winner Silvio Micali. Algorand uses cryptographic sortition to randomly select committees for each block, aiming for fast finality without forks.
Why it scores well: The committee-based design produces moderate entity-level spread. Twelve independent parties must coordinate to reach 33%, rounding out the top 10.
Context: Algorand has never captured the DeFi volume of Solana or Avalanche, but its decentralization score has been stable and predictable, which matters for institutions evaluating infrastructure risk.
Internet Computer (ICP): A Deeper Look
ICP earns a spot in the top 10 table, but it deserves more than a one-paragraph profile. Internet Computer is not a standard Proof-of-Stake validator chain. It runs on node machines operated by independent node providers across dozens of subnets worldwide. That is a fundamentally different architecture — and it is why Nakaflow skips it entirely while Chainspect includes it.
What Is ICP?
The Internet Computer Protocol (ICP), built by the DFINITY Foundation, is a decentralized compute platform. Entire applications (frontend, backend, and data) run on-chain as "canisters" on node machines hosted in independent data centers. The network is governed by the Network Nervous System (NNS), an on-chain DAO where ICP holders stake tokens in neurons and vote on upgrades, subnet changes, and economic parameters.
We covered ICP's tech stack, Caffeine AI, and tokenomics in depth in our project spotlight. Here we focus purely on decentralization.
How ICP's Nakamoto Coefficient Is Measured
DFINITY's Decentralization Response Engine (DRE) calculates the Nakamoto Coefficient across multiple dimensions, not just one:
- Node providers (who owns the hardware)
- Geographic distribution (country, continent, city)
- Data center ownership (whether one facility dominates)
For each dimension, the engine counts how many entities must collude to control more than one-third of nodes in a subnet. When subnets are reconfigured, DRE actively selects node replacements that maximize the Nakamoto Coefficient rather than letting concentration grow unchecked. That is unusual — most chains passively report decentralization; ICP's protocol tries to optimize it.
ICP's Scores: Infrastructure vs Governance
Chainspect's 14 measures node-provider concentration at the 33% infrastructure threshold — the number in our top 10 table. That is a solid mid-tier reading, tied with Aptos.
On the governance subsystem, the picture is different. DFINITY's 2023 analysis of NNS voting power calculated a Nakamoto Coefficient of 134 for publicly known neurons at the 51% voting-power threshold: meaning 134 separate entities would need to collude to reach a majority of governance votes. That included the DFINITY and ICA foundations (then ~22.7% combined) plus hundreds of independent stakers. Post-Genesis participants already controlled a majority of voting power at that point.
Do not conflate the two. A chain can have well-distributed node providers (NC 14) and still face governance questions about foundation influence. Read both numbers.
- Ranked #7 on Chainspect at NC 14 (node-provider infrastructure, 33% threshold)
- Not on Nakaflow — ICP uses node machines and subnets, not PoS validators
- NNS governance NC: 134 (DFINITY analysis, 2023, found-neuron subset, 51% threshold)
- DFINITY's DRE actively optimizes node-provider and geographic distribution per subnet
- Different subsystems (nodes vs governance) produce different numbers — compare like with like
Why ICP matters in this ranking: Without Chainspect, a reader scanning Nakaflow's top 10 would see Polkadot at 42 and never encounter ICP at all. Chainspect's inclusion fixes that blind spot. ICP's node-provider model, geographic spread across independent data centers, and high governance coefficient put it firmly in the decentralization conversation — even though it does not fit a standard PoS staking dashboard.
What About Bitcoin and Ethereum?
These two dominate crypto by market cap but do not fit neatly into the PoS top 10 table.
Bitcoin (BTC) — Nakamoto Coefficient: 4 on Chainspect; ~3 pools on Hashrate Index
Chainspect rates Bitcoin at 4 entities at the 51% hashrate threshold. Hashrate Index pool data from early July 2026 tells a similar story: Foundry USA (~24.5%) + AntPool (~17.9%) + F2Pool (~13.5%) = ~56%. Either way, a handful of mining pools could theoretically control a majority — lower than every chain in the top 10 table.
But pools are not miners. Hashrate is liquid and can redirect within hours if a pool misbehaves. Bitcoin also has 21,000+ independent full nodes validating every block. The coefficient undersells Bitcoin's defence in depth, even as pool concentration is a real and worsening trend.
Ethereum (ETH) — Nakamoto Coefficient: 1 on Chainspect
Ethereum has 1M+ validators, more than any rival. But validators ≠ operators. Chainspect's entity-level reading is just 1 at the 33% threshold — meaning a single staking entity (or tightly linked group) already exceeds the danger zone on their model. Lido alone controls roughly 23% of staked ETH (per its February 2026 update), and the top few liquid-staking providers combined approach 33%. Raw validator count favours Ethereum; entity concentration favours several PoS rivals in the table above.
Honourable Mentions (Outside the Top 10)
Chains that did not make the top 10 but are worth knowing about on Chainspect (July 7, 2026):
NEAR Protocol (9): Sharded L1 with stable but unremarkable decentralization
BNB Chain (7): Binance-linked validators; high volume, lower decentralization
Polygon (4): Popular Ethereum sidechain; only four entities reach 33%
Polkadot (1 on Chainspect; 42 on Nakaflow): Would rank #1 on Nakaflow. Chainspect's reading of 1 is almost certainly a broken stake feed — NPoS spreads stake across ~600 validators by protocol design. Use Nakaflow's 42 for Polkadot until Chainspect fixes the data
Base (1): Coinbase's L2; a single entity exceeds the threshold
Chains like Avail and Monad appear on Nakaflow's top 10 but are not listed on Chainspect's decentralization dashboard, which is another reason we switched to Chainspect as the primary source here — it covers ICP and a broader set of live mainnets.
What the Number Does Not Tell You
Before you treat the ranking as gospel, remember what it misses:
- Client software: One bug in a dominant client can halt the chain regardless of validator count
- Geography and cloud hosting: Ten "independent" validators on the same AWS region are not truly independent
- Governance: A decentralized validator set with a centralized foundation is a mixed picture
- Token ownership: Wide validator spread does not help if 60% of supply sits in 50 wallets
- The Nakamoto Coefficient = minimum entities needed to seize control
- TON leads Chainspect at 79 — impressive, but scrutinize the Telegram link and single-tracker source
- ICP ranks 7th at 14 on infrastructure; NNS governance NC was 134 in DFINITY's 2023 analysis
- Polkadot would be #1 on Nakaflow (42) — Chainspect's reading of 1 is a data error, not reality
- Ethereum reads just 1 on Chainspect — liquid staking concentration is a real risk
- Use rankings as a starting point, then dig into who runs the nodes, which tracker you trust, and where
For more on protecting yourself across any chain, see our crypto security series and the MiCA guide for European users.
ThriveInMarkets publishes educational market commentary only and does not provide personal investment advice. Nakamoto Coefficient figures are snapshots from public trackers and can change as stake and hashrate shift. Thresholds, entity definitions, and data sources vary between providers; verify against primary sources before making decisions.
