How to Find Ethereum Whales and Follow Their Money

An Ethereum whale is a wallet that holds or moves enough ETH to matter. There is no fixed line. In this guide we use 10,000 ETH, worth about $25 million in October 2026.

Finding whales is easy. Telling them apart from exchanges is the hard part, and it is where most whale alerts go wrong. Below we find the week's largest moves, label them, pick out a real whale, and follow its money. We use the Bitquery MCP server, so each step is a question in plain English. Each step shows the question, the tool the AI called, and what came back. We ran every query on 15 October 2026. All times are UTC.

Step 1: find the big transfers

We asked: "List the 50 largest ETH transfers between 8 and 14 October 2026."

The AI used eth_transfers_raw_sql, which reads Ethereum transfers directly.

What came back	Figure
Moves looked at	the 50 largest, each over 5,000
Coins moved in those 50	1,368,131
Largest single transfer	50,000 ETH, 8 Oct

Step 2: label both sides

We asked: "Label the senders and receivers of those transfers."

The AI used labels_for_addresses, which checks addresses against Bitquery's address labels: exchange wallets, funds, protocols and more.

What the labels showed	Figure
Moves with an exchange or custodian wallet on one side	at least 45 of 50
Coins in those moves	1,220,131 (89%)
Moves between two wallets of the same exchange	at least 23 of 50
Exchanges and custodians seen	Binance, Bitfinex, Coinbase, Kraken, OKX and others

At least 23 of the 50, carrying 569,467 ETH, went between two wallets of the same exchange. The largest, 50,000 ETH, went from one Bitfinex wallet to another, and 32,489 came back 12 seconds later. A whale alert that skips this step reports bookkeeping as news.

Step 3: pick a real whale

Five of the 50 transfers had no labelled exchange on either side. Two came from a wallet that Bitquery's labels, from one source, tie to Abraxas Capital: 20,000 ETH on 8 October and 28,000 ETH on 9 October, both into Aave, a lending protocol.

Wallet	Label
0xed0c6079229e2d407672a117c22b62064f4a4312	Abraxas Capital fund wallet

We asked: "What is this wallet's history, and what does it hold?"

The AI used eth_address_profile. We also read the wallet's token balances from a public Ethereum node.

What came back	Figure
Active since	8 April 2024
Transfers received / sent	21,194 / 26,912
Plain balance	3 ETH
Deposited in Aave (aEthWETH)	58,654 ETH

The plain balance says 3 ETH. The real position is 58,654, about $147 million, held as an Aave deposit token. A whale search that only reads plain balances misses this wallet completely.

Step 4: follow the money

We asked: "Where did this wallet send and get its coins in the last 30 days? Include coins moved by contracts, and label each counterparty."

The AI used eth_flows_between with no second address set, which lists every counterparty, then labels_for_addresses.

Counterparty (label)	In, ETH	Out, ETH
Aave contracts	134,950	199,241
Lido withdrawal queue	45,425
Lido stETH		3,000
Bitfinex	34,500	31,425
Binance	22,500	17,000
Unlabelled wallet 0xb99a…bcf5		13,929

The wallet moves coins between exchanges, Aave and Lido, an ETH staking service. It took 45,425 out of Lido's withdrawal queue on 5 and 6 October, and put far more into Aave than it took back out.

We asked: "Did this wallet trade on DEXes in the last 30 days?"

The AI used trader_profile. The answer: one swap, worth $1,875, on Uniswap. This whale lends and stakes. It barely trades on DEXes. Reading its Aave deposits as a buy or sell signal would be a mistake.

Step 5: watch for fake counterparties

The same 30 days of history held something else. Several tiny transfers, 0.00001 ETH each, came from addresses that copy the first and last characters of the wallet's real counterparties.

Real counterparty	Look-alike that sent dust
Binance deposit 0xc29bf263…56fe	0xc29b02c5…56fe, 0xc29b37aa…56fe
Aave 0xd01607c3…5722	0xd016dd06…5722
Aave 0xbd7d6a9a…1704	0xbd7d0110…1704

This is address poisoning. The attacker hopes someone copies the look-alike from the wallet's history and sends real money to it. Fake "USDT" tokens with look-alike symbols showed up in the same history. When you follow a whale, check full addresses and token contracts, never the first few characters or a symbol.

What whale data can and cannot tell you

Whale tracking shows where large amounts sit and how they move. It does not show why. A transfer to an exchange may be a sale, a loan or a move between the firm's own accounts. A deposit into Aave may back a loan taken elsewhere. Treat whale moves as questions to look into, not as trade signals. Nothing here is investment advice.

Run this yourself

Connect the Bitquery MCP server to your AI client with one URL: https://mcp.bitquery.io. Tracing and labels sit on the AI Investigation MCP plan, at $149 per month on our pricing page. The setup guide covers each client.

Client	How to connect
Claude Desktop / Claude web	Settings, Connectors, add custom connector https://mcp.bitquery.io
ChatGPT	Settings, Connectors, add a connector with the same URL
Cursor	Settings, MCP, Add new MCP server
Claude Code	claude mcp add --transport http bitquery https://mcp.bitquery.io

Then try these prompts:

List the 20 largest ETH transfers in the last 7 days and label
both sides. Leave out transfers between exchange wallets.

Who are the top holders of WETH right now? Label them.

Where did 0xed0c6079229e2d407672a117c22b62064f4a4312 send ETH
in the last 30 days? Include ETH moved by contracts.

Did this wallet trade on DEXes this month?

For developers: the GraphQL APIs

If you want to build your own whale tracker, the same data is in Bitquery's GraphQL APIs. The docs keep tested, current examples:

• Ethereum Token Holder API for the top holders of any token
• Ethereum Balance API for a wallet's balances and their history
• ERC20 Token Transfers API for large transfers of any token
• Ethereum DEX Trades API for a wallet's swaps

Frequently asked questions

What is an Ethereum whale?

A wallet that holds or moves enough ETH to affect the market. There is no fixed threshold. This guide uses 10,000 ETH, about $25 million in October 2026.

How do I find Ethereum whale wallets?

List the largest transfers or the top holders of ETH and WETH, then label every address. Remove exchange and protocol wallets, and check deposit tokens such as Aave's aEthWETH, which a plain balance check misses.

Are most whale alerts real whales?

Often not. Between 8 and 14 October 2026, at least 45 of the 50 largest transfers had an exchange or custodian wallet on one side.

Should I copy what whales do?

No. A whale's move may be a loan, a hedge or an internal transfer, and the ledger does not say which. Use whale data as a starting point for research.

Related reading

• Where did the Twitter hack Bitcoin go? A full trace
• How to investigate an Ethereum address
• FairWin: who won and who lost in the Ethereum Ponzi

First published in March 2024 by Divyasshree. Rewritten on 15 October 2026: the GraphQL walkthrough was replaced by an MCP walkthrough with links to the maintained API docs, every figure comes from a new run of the Bitquery MCP server and a public Ethereum node, and AI tools helped run the queries and draft the text. Bitquery sells the MCP server named above. The wallet is named only by the label Bitquery's data gives it. Nothing here is investment advice.