A snapshot with three different quantities
Binance’s October 1 proof-of-reserves page reports 688,572.68 BTC in net account balances and 691,003.593 BTC in on-chain wallet balances. It separates the latter into 679,550.124 BTC held on the exchange and 11,453.469 BTC held through third-party custody. The displayed reserve ratio is 100.35%.
Those labels matter. Net account balances represent the user-balance side of the snapshot. On-chain wallet balance represents assets attributed to the exchange and its disclosed custody arrangements. The reserve ratio compares those quantities at the recorded time; it is not a measure of trading volume, revenue, profit, or market direction.
What the verification system is designed to test
The exchange says it builds a Merkle tree from user records and uses a zk-SNARK mechanism in the assessment. A customer receives a record identifier and Merkle leaf for a selected verification date. That allows the customer to check whether their account balance was included in the liabilities dataset without publishing every customer’s balance.
The public page also provides a Merkle root and links to address data. These components answer different questions. The Merkle path concerns inclusion in the account-balance set. Address ownership and disclosed wallet balances concern the asset side. A useful review checks both instead of treating a large wallet total as proof that a particular customer record was counted.
What the snapshot does not establish
A point-in-time reserve page is narrower than a full financial-statement audit. The displayed data does not, by itself, provide a complete view of the company’s liabilities, corporate structure, internal controls, asset encumbrances, or obligations outside the covered account-balance calculation. Binance states that excess assets in the listed reserve addresses should not be read as the total amount of the company’s own assets.
The snapshot also cannot explain why customer BTC balances changed from the prior month. Deposits, withdrawals, transfers between venues, and changes in customer behavior can all move the total. The number alone does not show that users intend to sell, that confidence has risen, or that liquidity conditions have improved. Those interpretations require separate evidence.
A practical reading order
Start with the audit timestamp and asset, then compare net account balances with the on-chain wallet balance and stated ratio. Check whether third-party custody is broken out rather than folded into an unexplained total. Next, use the customer verification flow to test inclusion of a specific account record. Finally, retain the date: a later deposit or withdrawal does not retroactively change what the October 1 snapshot measured.
This reading order keeps a useful disclosure in proportion. The page offers testable evidence about selected balances, wallet coverage, and account inclusion at one time. It should be used for those questions, not stretched into a general claim about solvency or future market behavior.
Source: BTCUSA.