Nakamoto Coefficient

What Is the Nakamoto Coefficient?

The Nakamoto Coefficient is a quantitative metric that measures how decentralized a blockchain network is. Named after Bitcoin's pseudonymous creator Satoshi Nakamoto, the coefficient represents the minimum number of independent entities that would need to collude to disrupt or control the network. The concept was formalized in a 2017 essay titled "Quantifying Decentralization" by Balaji Srinivasan and Leland Lee.

The intuition is straightforward: if three mining pools control more than half of Bitcoin's hashrate, the Nakamoto Coefficient for mining is 3. Those three pools could theoretically coordinate a 51% attack. A network where 50 independent validators would need to collude has a coefficient of 50 and is considered substantially more decentralized.

Before this metric existed, discussions about decentralization were largely qualitative. The Nakamoto Coefficient gives researchers, investors, and developers a single number to compare networks and track centralization trends over time.

How It Works

Srinivasan and Lee proposed a three-step methodology for calculating the Nakamoto Coefficient:

1. Enumerate the essential subsystems of the decentralized system (mining, nodes, development, exchanges, client software, governance)
2. For each subsystem, determine how many entities would need to be compromised to control it
3. Take the minimum across all subsystems as the overall Nakamoto Coefficient

The key insight is using the minimum. A network might have 10,000 full nodes but only 3 mining pools controlling majority hashrate. The coefficient reflects the weakest link: 3, not 10,000.

Calculating for Proof-of-Work Chains

For proof-of-work blockchains like Bitcoin, the calculation focuses on mining pool concentration. The standard threshold is 51% of total hashrate:

1. List all mining pools ranked by their share of total hashrate
2. Sum their shares from largest to smallest
3. Count how many pools it takes for the cumulative share to exceed 51%

As of mid-2025, Bitcoin's mining-pool-based Nakamoto Coefficient was approximately 3, with Foundry USA (~27-31%), AntPool (~17%), and ViaBTC (~15%) collectively exceeding 51% of block production.

// Simplified Nakamoto Coefficient calculation
function nakamotoCoefficient(
  shares: number[],
  threshold: number = 0.51
): number {
  const sorted = [...shares].sort((a, b) => b - a);
  let cumulative = 0;
  for (let i = 0; i < sorted.length; i++) {
    cumulative += sorted[i];
    if (cumulative >= threshold) {
      return i + 1;
    }
  }
  return sorted.length;
}

// Bitcoin mining pools (approximate shares)
const btcPools = [0.28, 0.17, 0.15, 0.12, 0.08, 0.06, 0.04];
console.log(nakamotoCoefficient(btcPools));
// Output: 3 (0.28 + 0.17 + 0.15 = 0.60 > 0.51)

Calculating for Proof-of-Stake Chains

For proof-of-stake networks like Ethereum, the calculation uses validator stake distribution instead of hashrate. One important nuance: the critical threshold differs. Most PoS networks rely on Byzantine fault tolerance, which requires two-thirds agreement among validators. This means an attacker controlling one-third of staked value can halt the network. Some analysts therefore use 33% as the threshold for PoS chains rather than 51%.

// PoS threshold: 33% for BFT-based consensus
const ethValidators = [0.12, 0.10, 0.09, 0.08, 0.07, 0.06, 0.05];
console.log(nakamotoCoefficient(ethValidators, 0.33));
// Fewer entities needed at the 33% threshold

This threshold distinction is methodologically significant. A network reporting a Nakamoto Coefficient of 10 under a 51% threshold is not directly comparable to one reporting 10 under a 33% threshold.

Subsystem Analysis

The original framework calls for evaluating multiple subsystems, not just mining or staking. Srinivasan and Lee identified several dimensions of decentralization:

The overall coefficient is the minimum across all these subsystems. In practice, most published Nakamoto Coefficients only measure the mining or staking subsystem because that data is easiest to obtain on-chain.

Why It Matters

Decentralization is often cited as the primary advantage of blockchain systems over traditional infrastructure. But without a way to measure it, claims of decentralization remain subjective. The Nakamoto Coefficient provides a concrete benchmark.

• Network security assessment: a low coefficient signals vulnerability to coordinated attacks. If only 2 or 3 entities need to collude, the network's censorship resistance and consensus integrity depend on the goodwill of a small group.
• Trend tracking: monitoring the coefficient over time reveals whether a network is becoming more or less decentralized. A steadily declining coefficient is a warning sign.
• Protocol comparison: investors and developers can compare the decentralization of different networks using a consistent metric, though the threshold used must match.
• Regulatory context: as regulators evaluate which digital assets qualify as sufficiently decentralized, quantitative metrics like the Nakamoto Coefficient may inform classification decisions.

For Bitcoin specifically, the concentration of mining pools has been a persistent concern. A Nakamoto Coefficient of 3 means that just three organizations collectively produce the majority of blocks. While this does not mean those pools would collude (individual miners within a pool could switch pools), it highlights a structural centralization pressure that the mining centralization research explores in detail.

Use Cases

Investor Due Diligence

Institutional investors increasingly evaluate the Nakamoto Coefficient as part of their due diligence on blockchain networks. A low coefficient introduces counterparty risk: if a handful of validators or mining pools can censor transactions or reorganize the chain, the network's value proposition as a censorship-resistant settlement layer weakens.

Protocol Design

New blockchain protocols use the Nakamoto Coefficient as a design target. Some networks implement mechanisms specifically to increase their coefficient: delegation caps that limit how much stake any single validator can attract, geographic distribution requirements, or delegated staking structures that spread influence across more operators.

Mining and Staking Analysis

Researchers and analysts track the coefficient to identify centralization trends in mining and staking. Dashboards like D-Central's Bitcoin Pool Centralization tracker publish the metric alongside the Herfindahl-Hirschman Index (HHI) to give a more complete picture of concentration. Changes in the coefficient can signal shifts in hashrate geography, new pool entrants, or consolidation.

Governance Evaluation

Decentralized autonomous organizations and on-chain governance systems apply the same logic to governance token distribution. If a small number of token holders can pass any proposal, the governance is effectively centralized regardless of how many participants exist.

Risks and Considerations

Snapshot Limitation

The Nakamoto Coefficient captures decentralization at a single point in time. Mining pool shares, validator sets, and stake distributions all change continuously. A coefficient measured today may not reflect the network's state next week. Short time windows can cause significant swings, making it important to track the metric over longer periods rather than relying on individual readings.

Geographic Blindness

The coefficient treats all entities equally regardless of jurisdiction or physical location. A network with 20 validators all hosted in a single data center in one country scores the same as a network with 20 validators spread across 15 countries. In practice, the geographically concentrated network is far more vulnerable to regulatory action, infrastructure failure, or state-level coercion. Research on Bitcoin hashrate geographic redistribution illustrates why location matters.

Entity Ambiguity

Defining what counts as a single "entity" is not straightforward. A mining pool like Foundry USA coordinates thousands of individual miners. Should it count as one entity or many? The pool operator controls block template construction, but individual miners can leave at any time. Similarly, a staking service may custody tokens from thousands of delegators but operate under unified management. The coefficient changes dramatically depending on how entities are defined.

Single-Subsystem Bias

While the original methodology calls for measuring multiple subsystems and taking the minimum, most published scores only measure mining or staking. This can paint an incomplete picture. A network might have a healthy mining coefficient but extreme client diversity problems: if 90% of nodes run the same software implementation, a single client bug could halt the network regardless of how many miners exist.

Threshold Disagreement

There is no universal agreement on the correct attack threshold. Proof-of-work chains typically use 51% based on the 51% attack model, while proof-of-stake chains might use 33% (to halt finality) or 67% (to forge blocks). Cross-chain comparisons are meaningful only when the same threshold is applied consistently.

Complementary Metrics

The Nakamoto Coefficient works best alongside other decentralization measures rather than in isolation. The Gini coefficient captures inequality in the overall power distribution. The Herfindahl-Hirschman Index (HHI) measures market concentration. Shannon entropy quantifies the unpredictability of which entity produces the next block. No single metric captures decentralization fully; the Nakamoto Coefficient answers one specific question: how many entities must collude?

This glossary entry is for informational purposes only and does not constitute financial or investment advice. Always do your own research before using any protocol or technology.