Independent DeFi security desk · Status: operationalMethodology & corrections · Submit an incident
Noticias · Fact checked

Ethereum Outflow Charts Are Revised as More Exchange Wallets Are Identified

Coin Metrics has rebuilt its ETH Standard flow series after identifying more exchange wallets, meaning historical outflow figures can change. A trading decision based on an outflow claim still depends on the data that was available when the decision was made.

This article may contain affiliate links. Commercial relationships are disclosed in the affiliate policy.

Noticias — illustration keyed to this article's identifier. Source documents are listed under Sources and are not reproduced here.

What happened

Coin Metrics has rebuilt its ETH Standard flow series. According to CryptoSlate, the rebuild followed the identification of additional exchange wallets, and the consequence is that historical outflow charts for ethereum can change. A dataset that looked settled turns out to be a moving estimate.

The reason is methodological rather than economic. Exchange attribution depends on labels: an address is classified as an exchange because its behaviour matched a known exchange’s pattern. Every newly labelled address changes the total attributed to that category, and the change propagates backwards through every historical point that included it.

Why this matters more than a routine chart revision

A revised historical series is a different object from a revised forecast. It means the past was measured differently than it was measured before, so:

  • Any claim built on “the outflow was X on date Y” is now a claim about a dataset version, not about the chain.
  • A strategy that triggered on a flow threshold may have behaved differently against the corrected series than against the original.
  • Comparing today’s reading to a chart drawn from the old series mixes two different definitions in one line.

None of this makes the underlying data wrong. It makes the version relevant.

The point about decision time

The most useful distinction in this report is between a data revision and a decision. A revision says the measurement changed. It does not say a decision made on the earlier measurement was irrational or unprofitable. Those are separate questions, and a trading claim evaluated only against the latest series is not being evaluated on the information that existed at the moment.

Evidence boundary

This account follows CryptoSlate’s report on the Coin Metrics rebuild. The size of the revision, the number of newly identified wallets and the resulting change to any specific historical figure are not stated in the material available here.

Sources