Back to: Certified Decentralization Analyst
The Nakamoto Coefficient is the single most useful decentralization number in crypto. It answers: how many independent entities do you have to corrupt or control to compromise the network?
A coefficient of 1 means one entity can take over. A coefficient of 24 means you’d need 24 independent actors working together. Higher = safer.
How it's computed
Rank the biggest holders of some resource (hash power, stake, validators, clients). Add them up largest-first until the cumulative share passes the compromise threshold (usually ~33–34% or 51% for different attacks). The count you reached is the Nakamoto Coefficient for that resource.
The two traps
1. Node count ≠ coefficient. “10,000 validators!” is meaningless if 20 pools control 51% of the stake. The coefficient counts *independent decision-makers*, not machines. 2. One resource ≠ the whole network. A chain can have a great validator coefficient (320 independent validators) and a software coefficient of 1 — if 100% of clients run a single codebase. Always ask *which resource* a coefficient refers to.
Real examples from W3D
– Bitcoin (infra ≈4): ~17,800 nodes, but mining pools concentrate hash — a handful of pools approach the threshold. Strong overall, thanks to other pillars. – Polkadot (>90): the most diffuse validator set in crypto. – Solana (≈19): meaningfully good validator dispersion, hurt elsewhere by cloud hosting. – Arbitrum (4 inherited): security inherited from Ethereum’s L1 — its own coefficient would be 1 (single sequencer).
Practicing as an analyst
When you see a Nakamoto Coefficient, ask: – Which resource is it measuring? – What’s the threshold (49%? 34%)? – What does the *weakest* resource look like?
A chain’s true decentralization is its **worst** Nakamoto Coefficient across
> resources — the attacker takes the path of least resistance.
Next lesson: the Infrastructure pillar in depth.