On September 16, the Bank for International Settlements (BIS) published a working paper specifically discussing the statistical deviations of on-chain blockchain metrics. Based on nearly 10 billion blockchain records from the three major networks: Bitcoin, Ethereum, and Tron, the paper points out that commonly used on-chain data has systematic errors in depicting real economic activity and emphasizes that related metrics should be regarded as "approximations" rather than precise measurements. According to AiCoin data, at the time of the paper's release, the price of Bitcoin was approximately $75,923, hovering within a densely leveraged exposure area: the upper range of $77,400–$77,800 had around 994.86 BTC in potential short liquidation exposure, totaling about $77.081 million, which is about 1.95% from the current price, with a single address holding around 325 BTC constituting a local concentration risk point; the lower range of $72,600–$72,800 had about 874.79 BTC in potential long liquidation exposure, totaling about $63.558 million, around 4.1% from the current price. Existing materials do not provide evidence that the release of the BIS paper and the short-term changes in Bitcoin's price or liquidation exposure have a direct causal relationship. One side sees regulators and academic institutions questioning the credibility of on-chain metrics, while the other side sees market participants building highly leveraged structures based on these metrics and price distributions. How Bitcoin risk is priced amidst this data controversy and position tug-of-war has become a core tension that needs continuous observation.

BIS's Verification: Nearly 10 Billion On-Chain Records Reassessed

The working paper published on September 16 directly engages in “verification” of mainstream public chains. According to a single source, the BIS research team selected Bitcoin, Ethereum, and Tron, covering both UTXO and account models, analyzing nearly 10 billion on-chain records to re-statistical common metrics such as transaction volumes and address activity. Bitcoin, Ethereum, and Tron currently occupy substantial positions in the market with infrastructure status, and this cross-model, large-sample unified approach makes the paper's conclusions highly representative of the entire industry. It also clearly challenges the practice over the past few years of using on-chain data as substitutes for measuring "real economic activity."

The crux of the difference lies in the underlying ledger structure. Bitcoin uses a UTXO model, where each transaction corresponds to multiple unspent transaction outputs, including both actual payments and “change” automatically returned to oneself. Wallets also consolidate or split UTXOs at different points in time, which changes the transaction output structure and thus alters any statistics constructed by "output amount" and "output count"; whereas Ethereum and Tron operate on an account model, directly adding or subtracting from address balances, with transfer paths being more linear statistically. The paper's estimates show that due to different handling of UTXOs, Bitcoin's "transaction volume" metric can differ by as much as six times at extremes. The BIS thus points out that various on-chain metrics should be regarded as approximate data reflecting underlying activities rather than precise measures of real economic activity.

Bitcoin Transaction Volume with Sixfold Error, Metric Blind Faith Dispelled

Under the UTXO model, a Bitcoin transaction often includes actual payments, change, and multiple old UTXOs as merged inputs, and different statistical approaches to these components directly widen the gap in "on-chain transaction volume." The BIS working paper provides calculations showing that under extreme circumstances, merely based on whether change outputs and merged outputs are included in "transaction volume," Bitcoin's on-chain transaction volume can show a difference of up to six times. According to AiCoin data, the "on-chain transaction volume" commonly referenced in the market often does not fully disclose the specific processing logic for UTXOs, meaning investors may unknowingly use incompatible versions of these metrics.

This sixfold error level exposes what was originally viewed as "hard data" in transaction volume to be filled with statistical choices. Once researchers or traders equate these volume metrics directly with actual monetary flows, it can systematically overestimate network economic activity and adoption or conversely underestimate on-chain activity when employing more conservative benchmarks. More critically, when risk models, valuation frameworks, and even market sentiment are based on these "volume curves" to assess Bitcoin's usage intensity and potential risk exposure, measurement error is no longer just a technical detail; it permeates the cognitive structure of the entire network's security and risk pricing.

The Tug of War Around 75,000: Concentrated Distribution in Liquidation Zones

According to AiCoin data, at the time of the BIS working paper's release, Bitcoin's current price hovered around $75,923, stuck between the potential liquidation zones of leveraged positions on both sides. The upper range of $77,400–$77,800 contains around 994.86 BTC in potential short liquidation exposure, with a nominal value of about $7.081 million, just about 1.95% from the current price; the lower range of $72,600–$72,800 contains about 874.79 BTC in potential long liquidation exposure, with a nominal value of approximately $6.558 million, about 4.1% from the current price. The price is nearly "sticking" between the concentrated areas of both shorts above and longs below, any moderate price fluctuation upwards or downwards has the potential to first touch one side's high-density leveraged zone, changing the local risk distribution.

It is worth noting that within the upper short liquidation zone of $77,400–$77,800, there exists a single address holding about 325 BTC, which becomes a prominent concentration risk point within that zone. This means that while the overall nominal scale appears controllable, individual large addresses' passive reductions or liquidations may amplify the intensity of the price and liquidation connection, and risk models evaluating such "sticking distributions" of long and short structures need to see both total amounts and percentage distances of the two upper and lower zones, without neglecting the single-point concentrated risk exposed behind statistical averages.

When Flawed On-Chain Data Meets Fragile Leverage Positions

With the BIS reminding that on-chain metrics are approximations and not precise, the question becomes: how do statistical deviations distort risk perceptions when they encounter highly sensitive leveraged structures? The market has long been accustomed to using metrics such as on-chain transaction volume and active address counts to assess Bitcoin's activity, adoption, and risk preferences, while BIS's research shows that just the on-chain transaction volume for Bitcoin can differ by about six times under different UTXO processing methods, meaning investors could derive vastly different interpretations of "fund flows" and "on-chain activity" from the same set of original records. If one side's interpretation is overestimated or underestimated, and then incorporated into trading and risk control models, it amounts to calibrating expectations of future volatility with inputs containing statistical errors.

According to AiCoin data, when BTC is priced around $75,923, the upper range of $77,400–$77,800 contains about 994.86 BTC in potential short liquidation exposure, while the lower range of $72,600–$72,800 contains about 874.79 BTC in potential long liquidation exposure, with both sides less than 5% from the current price. Additionally, within the upper range, a single address holds about 325 BTC, making the leveraged positions highly sensitive to minor price changes. In this "sticking distribution" environment, if risk teams mistakenly regard the significantly deviated on-chain transaction volume as accurate signals of fund flows, they may underestimate the probability of triggering dense liquidation zones or incorrectly assess the spillover effects of localized concentrated exposures. However, existing materials do not provide any evidence of immediate changes in Bitcoin's price or leveraged positions following the release of the BIS paper; a more reasonable positioning of this research is to remind market participants to recalibrate how they utilize on-chain data rather than directly attributing it as a causal explanation for recent price fluctuations.

Next Steps for Using On-Chain Data: Calibrate Metrics Rather Than Abandon Tools

With the BIS having publicly disclosed the statistical deviations in on-chain data, the more pragmatic approach for institutions and traders is to view various on-chain indicators as "error-prone approximations." Before using liquidation exposure, transaction volume, and other data, they should actively consult statistical platform explanations regarding methodology, sample selection, and deduplication methods instead of assuming their precise mapping of real monetary behavior. In the future, regarding the differences between UTXO and account models, BIS and other research teams will likely continue to propose new statistical frameworks or alternative indicators. If these methods can reduce the deviation of critical metrics like Bitcoin transaction volume by up to six times, it will directly improve the data foundation relied upon for risk pricing. Regarding the current market structure, according to AiCoin data, the current BTC price is close to the narrow distance between the upper short liquidation band of $77,400–$77,800 and the lower long liquidation band of $72,600–$72,800, with about 325 BTC concentrated exposure within the upper range making these two leveraged sensitive zones continuously trackable risk coordinates. However, the liquidation data itself is still influenced by statistical methods and platform differences, providing structural information rather than infallible truths. Whether this fragile balance can be broken to reshape the market's perception of Bitcoin risk depends on the interaction between price evolution and subsequent improvements in on-chain statistics.

Join our community to discuss and become stronger together!
Exclusive Hyperliquid benefits for AiCoin: https://app.hyperliquid.xyz/join/AICOIN88
Exclusive Aster benefits for AiCoin: https://www.asterdex.com/zh-CN/referral/9C50e2
On-chain Telegram community: https://t.me/AiCoinWhaleData
On-chain community: https://www.aicoin.com/link/chat?cid=N6OVMor5g
AiCoin on-chain Twitter: https://x.com/aicoinwhaledata

免责声明：本文章仅代表作者个人观点，不代表本平台的立场和观点。本文章仅供信息分享，不构成对任何人的任何投资建议。用户与作者之间的任何争议，与本平台无关。如网页中刊载的文章或图片涉及侵权，请提供相关的权利证明和身份证明发送邮件到support@aicoin.com，本平台相关工作人员将会进行核查。