BIS Study Reveals Significant Variability in Bitcoin On-Chain Transfer Estimates

The findings challenge the reliability of on-chain metrics that underpin a multi-trillion-dollar cryptocurrency market.

• BIS published a working paper on September 15, 2026, revealing that Bitcoin on-chain transfer estimates can vary by up to six times based on methodological choices.
• The study analyzed around 100 billion blockchain records from Bitcoin, Ethereum, and Tron, highlighting the complexity of measuring economic activity in crypto.
• Researchers emphasized the need for transparent methodologies to improve the accuracy of on-chain metrics used by market participants and regulators.

• On-chain metrics like transaction volumes and token counts have been essential for assessing economic activity in the cryptocurrency sector.
• Previous analyses noted discrepancies in measurement approaches, but this is the first comprehensive study quantifying the scale of these variations.
• The UTXO model used by Bitcoin and the smart contract complexity of Ethereum contribute to the challenges in achieving consistent metrics across different blockchains.

The Bank for International Settlements (BIS) recently published working paper No. 1377, which scrutinizes the methodologies used to estimate Bitcoin on-chain transfer values. The research team processed extensive transaction data from Bitcoin, Ethereum, and Tron using the Mercurius data platform, analyzing approximately 100 billion blockchain records. They tested various methodologies for estimating Bitcoin transfer values, including unadjusted counts of all outputs, adjusted measures excluding likely change sent back to sending addresses, and conservative approaches that subtracted the largest output.

The results were striking: estimates of Bitcoin transfer values differed by a factor of six depending on the methodology employed. This variability raises critical questions about the reliability of on-chain metrics that analysts, market participants, and regulators have relied upon to gauge economic activity in the cryptocurrency and decentralized finance (DeFi) ecosystems. The study also highlighted the structural challenges in data aggregation, smart contract programmability, and cross-chain comparability.

In addition to Bitcoin, the researchers identified approximately 13 million active Ethereum contracts, including 1.4 million tokens, and noted divergent stablecoin usage patterns. Ethereum stablecoins are predominantly utilized within smart contracts, while simpler holdings are more common on the Tron network. This divergence underscores the complexity of the crypto landscape and the need for clear methodological assumptions to align raw blockchain records with meaningful economic indicators.

The implications of these findings are profound. As the cryptocurrency market continues to grow, the accuracy of on-chain metrics will become increasingly important for investment decisions, regulatory frameworks, and market analysis. Without standardized methodologies, stakeholders may be operating on flawed data, leading to misguided strategies and potential financial losses.

• Investors: Those relying on on-chain metrics for trading decisions may face increased risks due to unreliable data.
• Analysts: Professionals interpreting market trends will need to reassess their methodologies and assumptions.
• Regulators: Authorities tasked with overseeing the cryptocurrency market may struggle to enforce policies based on inaccurate metrics.
• Data providers: Companies offering blockchain analytics services will need to adapt their reporting tools to incorporate clearer methodological choices.

• Standardization efforts: Watch for initiatives aimed at creating standardized methodologies for measuring on-chain metrics, which could enhance data reliability.
• Market reactions: Monitor how investors and analysts adjust their strategies in response to the findings, particularly in the wake of potential regulatory changes.
• Technological advancements: Keep an eye on developments in blockchain technology that may address the structural challenges highlighted in the study, improving data aggregation and analysis.