ARK Invest and Glassnode have published a joint research report examining the decentralization architectures of the three largest blockchain networks — Bitcoin, Ethereum, and Solana. Titled The Decentralization Spectrum: Design Tradeoffs in Digital Assets, the report moves away from treating decentralization as a binary label and instead places each network on a measurable continuum defined by four design features: auditability, security, governance, and ownership. The analysis is supported by a six-dimension quantitative scoring framework built from Glassnode on-chain data and other network sources, underscoring how network design choices can be assessed through observable data rather than broad labels alone.

The report argues that every architectural decision — from block size to staking model — places a network at a specific point on that spectrum, and that this position reflects deliberate tradeoffs. A high-throughput payment network and an institutional settlement layer, the report says, are designed to sit at different points on the spectrum. When scores across all six dimensions are averaged, Bitcoin ranks highest on overall decentralization, Ethereum sits in the middle, and Solana prioritizes performance and coordination speed over distributed control.

The blockchain trilemma: a network can strengthen two of decentralization, security, and scalability, usually at a cost to the third. Read our joint report with @ARKInvest on how Bitcoin, Ethereum, and Solana compare. pic.twitter.com/ilvLTxi6RR — glassnode (@glassnode) September 1, 2026

The blockchain trilemma: a network can strengthen two of decentralization, security, and scalability, usually at a cost to the third. Read our joint report with @ARKInvest on how Bitcoin, Ethereum, and Solana compare. pic.twitter.com/ilvLTxi6RR

Control thresholds, infrastructure geography, and scoring methodology

On the security dimension, the report measures what it calls the critical control threshold — the minimum number of independent entities that would need to collude to meaningfully alter each ledger. Bitcoin and Ethereum each cross that threshold at just three entities, while Solana requires 19. The authors caution, however, that mining pool and staking delegation structures make direct comparisons with actual operational control more complex.

The report also points to an asymmetry in exit fluidity. A Bitcoin miner can liquidate a 1% network position in roughly 30 seconds by powering down hardware, while the equivalent unstaking process on Ethereum can take weeks during periods of network stress.

Geographic and infrastructure distribution show additional differences. Bitcoin has the most balanced physical node footprint of the three, with approximately 63% of its nodes operating via Tor, a setup that provides meaningful resistance to coordinated coercion. Ethereum shows a notable concentration in cloud infrastructure, with roughly 20% of its nodes hosted on AWS alone. Solana’s validator infrastructure is operated almost entirely within data centers, a design choice that supports performance but concentrates physical exposure.

The six scored dimensions — ownership distribution, exit fluidity, network verification overhead, critical resilience threshold, blockchain reconstruction overhead, and geographic and provider resilience — show that no single network leads across every category. Bitcoin scores highest on auditability, ownership distribution, and geographic resilience. The full report includes a methodology appendix and governance case studies covering Bitcoin’s SegWit upgrade and Solana’s SIMD-228 proposal.