A new BIS analysis finds that Bitcoin on-chain transfer values can differ by up to six times depending on calculation method, raising questions about the reliability of widely cited blockchain metrics and their use in market analysis
How much Bitcoin actually moves on-chain? A new working paper from the Bank for International Settlements (BIS) shows that the answer depends almost entirely on how you measure it. The BIS researchers found that estimates of Bitcoin transfer values can swing by as much as six times, depending on the method used to interpret blockchain data. This gap highlights a core problem for anyone relying on on-chain metrics to judge real economic activity in crypto markets.
How measurement changes the numbers
The issue comes from how Bitcoin transactions are structured. When someone spends Bitcoin, any leftover amount is usually sent back as "change" to the sender's own address. Some ways of measuring count this change as a separate output, which can inflate the total value transferred. Other approaches try to exclude it, aiming to capture only what is sent to other people. The BIS study, which looked at about 100 billion blockchain records across Bitcoin, Ethereum, and Tron, found that this single choice in methodology can create a sixfold difference in reported transfer values according to Cointelegraph.
BIS analysis shows that Bitcoin on-chain transfer values can differ by up to six times depending on how blockchain data is processed.
This isn't just a technical detail. How change outputs are handled directly affects how much economic activity appears to be happening on the network. For traders, analysts, and regulators, these differences can distort perceptions of liquidity, adoption, and risk. BIS warns that common crypto metrics, including market cap and total value locked (TVL), may give a false sense of precision that the underlying data does not support. As a result, these indicators should be seen as rough estimates, not exact measures of economic activity.
Market cap and stablecoin complications
The BIS researchers also looked at other popular crypto indicators. They found that Bitcoin's usual market capitalization-calculated by multiplying the total supply by the latest on-chain price-can be up to four times higher than "realized capitalization," which values each coin at the price it last moved. This shows how market cap figures can overstate the value actually at risk in the network.
Stablecoins add more complexity. The same token, like USDT, can be used for different things depending on the blockchain. The BIS study found that USDT on Ethereum is more tied to decentralized finance (DeFi), while USDT on Tron is used more for payments and as a store of value. Adding up stablecoin activity across blockchains can mix together very different types of economic behavior, making it harder to see how these tokens are actually used.
The BIS study is based on roughly 100 billion blockchain records from Bitcoin, Ethereum, and Tron, making it one of the largest on-chain analytics samples to date. This scale allows for a more nuanced understanding of how protocol-level differences and data processing choices impact reported metrics.
Ethereum's smart contract blind spot
Ethereum's smart contract system brings its own measurement problems. Of the 67.5 million active contracts the BIS analyzed, about 54 million could not be categorized using the researchers' framework. This leaves a huge part of Ethereum's on-chain activity without a clear label, making it harder to interpret network data or compare it to other blockchains.
Some analytics providers are trying to address these gaps. Visa's Onchain Analytics dashboard, for example, now shows both total and adjusted stablecoin transaction volumes. The adjusted number tries to filter out distortions from high-frequency trading, bots, bridge routing, and internal exchange operations. Over a recent 30-day period, the dashboard showed $6.4 trillion in total stablecoin transaction volume across tracked networks, but only $313.1 billion after adjustment.
Why on-chain metrics are noisy
The BIS paper's authors argue that on-chain indicators should be seen as "noisy approximations rather than direct measures of economic activity." This warning applies not just to Bitcoin, but also to Ethereum, Tron, and the stablecoins that move across them. The findings echo concerns raised in other research, such as EgonCoin's earlier breakdown of how market share and liquidity can be misrepresented by headline figures.
For U.S. investors, companies, and regulators, the message is clear: blockchain data is only as reliable as the methods used to interpret it. Relying on raw transaction counts, market cap, or total value locked without understanding their limits can lead to flawed analysis and poor decisions. As the crypto industry matures, the need for transparent, standardized, and context-aware metrics will only become more important.
According to the BIS study, the gap between total and adjusted stablecoin transaction volume was especially wide in the most recent 30-day window, with $6.4 trillion in total volume shrinking to $313.1 billion after removing high-frequency and non-economic activity. This shows how much of the reported on-chain activity may not reflect real user transactions or actual value transfer.
On-chain metrics are a staple of cryptocurrency analysis, but interpreting them is tricky. Unlike traditional financial systems, where transaction data is standardized and regulated, blockchain records are open but ambiguous. Change outputs, internal transfers, smart contract interactions, and cross-chain bridges all add noise that can hide the true scale and nature of economic activity. For anyone using these metrics to make investment, regulatory, or business decisions, understanding their limits is essential for avoiding costly mistakes.
