---
title: "Coin Metrics Rebuilds Ethereum Flow Data, Highlighting Backtesting Risks"
description: "Coin Metrics has recomputed its Ethereum Standard Flow Metrics from genesis, a move that complicates historical backtesting for traders who rely on exchange outflow data as a signal."
author: "CryptoResearch AI"
published: "2026-10-05T00:03:18.944Z"
updated: "2026-10-05T00:03:18.944Z"
category: "ethereum"
reading_time_minutes: 1
content_type: "editorial"
canonical: "https://cryptoresearch.news/news/coin-metrics-rebuilds-ethereum-flow-data-highlighting-backtesting-risk"
tags: ["Ethereum", "Data", "Trading", "Backtesting", "Coin Metrics"]
---

# Coin Metrics Rebuilds Ethereum Flow Data, Highlighting Backtesting Risks

> Editorial content, written by CryptoResearch.

Coin Metrics has recomputed its Ethereum Standard Flow Metrics from genesis, a move that complicates historical backtesting for traders who rely on exchange outflow data as a signal.

Coin Metrics announced on Oct. 1 that it has rebuilt its Ethereum Standard Flow Metrics from the network's first block. The update, which covers both daily and hourly frequencies, is part of the provider's Ethereum Point-in-Time release.

The rebuild creates a potential issue for researchers and traders who use historical exchange outflow data to backtest trading strategies. Because the updated history includes wallet addresses discovered after the fact, a chart viewed today may reflect knowledge that was not available at the time of a past trading decision.

For a backtest to be accurate, it must use the information available at the moment a trade would have occurred. Coin Metrics notes that while its Standard metrics use all currently known addresses for an entity, its Point-in-Time series only includes addresses known during a specific historical interval. Later discoveries do not retroactively change Point-in-Time data.

The provider did not supply specific revision amounts or details on how these changes might impact strategy performance. Consequently, analysts must now account for data vintage—the specific version of the data used—when evaluating historical signals.

Other data providers also face similar challenges. CryptoQuant has noted that its exchange flow values can change as wallets are identified and validated through clustering updates. Similarly, Glassnode has illustrated the risks of data-vintage bias, showing in a hypothetical backtest that using revised balances can lead to different performance results compared to using Point-in-Time data.

To properly measure the effect of this Ethereum data rebuild, researchers would need to compare pre-rebuild Standard values with the new post-rebuild history while keeping trading rules and parameters fixed. Without such a controlled comparison, it remains unclear how these revisions impact the effectiveness of trading signals based on exchange outflows.

## Sources

- [Crypto slate](https://cryptoslate.com/ethereums-past-outflow-charts-can-change-when-more-exchange-wallets-are-identified/)

---

Published by CryptoResearch. Canonical version: https://cryptoresearch.news/news/coin-metrics-rebuilds-ethereum-flow-data-highlighting-backtesting-risk