> ## Content Index
> Fetch the complete content index at: https://blog.presolt.com/llms.txt
> Use this file to discover other available public pages before exploring further.

# Bitcoin Exchange Reserves Depend on Who Is Counting
- URL: https://blog.presolt.com/bitcoin-exchange-reserves-depend-on-who-is-counting/
- Published: 2026-10-29T07:22:00.000Z
- Updated: 2026-11-11T07:22:36.000Z
- Description: Trackers put Bitcoin on exchanges anywhere from 1.3M to 3.67M BTC. Here is why exchange reserve data differs and how to read it carefully.
- Author: Presolt Team
- Tags: Research, On-Chain Analysis, Metrics, Data, Markets, Bitcoin, Crypto, Exchange Reserves, Exchange Flows, Proof of Reserves

On October 27, two well-known analytics firms measured the same thing on the same day and got different answers. [CryptoQuant data reported by U.Today](https://u.today/bitcoin-sees-massive-1-billion-exchange-exodus?from=article-links&ref=blog.presolt.com) put Bitcoin's net exchange outflow at about 12,570 BTC. [Santiment, as reported by DailyCoin](https://dailycoin.com/bitcoin-price-81k-options-expiry-fed-impact/?ref=blog.presolt.com), put it at 14,942 BTC and called it the largest single-day drain in five weeks.

Both numbers are reasonable. Neither is wrong. They differ because "Bitcoin on exchanges" is not something the blockchain reports directly. Every tracker has to decide which addresses belong to exchanges, and those decisions shape the result. This post explains how exchange reserves and exchange flows are built, why estimates differ so much, and how to use them without reading too much into them.

## What exchange reserves and exchange flows measure

Exchange reserves are an estimate of how much Bitcoin sits in wallets controlled by centralized exchanges. Exchange flows track the change: coins moving into those wallets are inflows, coins moving out are outflows, and netflow is inflows minus outflows over a set period, usually one day.

The common reading goes like this. Coins on an exchange can be sold quickly, so rising reserves may point to more supply available for sale. Coins leaving exchanges usually go to self-custody, cold storage, or an institutional custodian, so falling reserves may point to holders planning to keep their coins longer.

That reading is a starting point, not a conclusion. A flow shows where coins moved. It does not show why they moved.

## How the number is built

Every exchange reserve figure is the result of three steps.

1\. Identify exchange addresses. Analytics firms label addresses using information exchanges publish, such as Proof of Reserves address lists, and their own clustering methods, which group addresses that appear to be controlled by the same entity.

2\. Sum the balances. The tracker adds up the Bitcoin held in every address it has labeled for each exchange.

3\. Choose coverage and smoothing. Each provider decides which exchanges to include and whether to show raw daily balances or averages over time.

[Glassnode's latest on-chain report](https://research.glassnode.com/the-week-onchain-week-42-2026/?ref=blog.presolt.com) is clear about these choices. Its reserve figures are 30-day averages of Bitcoin held on each exchange's published Proof of Reserves addresses, limited to 11 exchanges holding at least 1,000 BTC there. Glassnode also warns that the figures may not capture all of an exchange's reserves, especially where an exchange does not disclose its official addresses.

## Market snapshot

As of 02:38 UTC on October 29, 2026, Perplexity Finance data showed Bitcoin at $79,704.98, down 1.92% over 24 hours, with a market capitalization of about $1.59T. Ether traded at $2,293.74, down 2.70%. All ten of the large assets we track were lower over the period, with Cardano, Avalanche, Sui, and Chainlink each down more than 4%.

The backdrop was the Federal Reserve's decision on October 28 to keep its target range at 3.75% to 4.00%. [Crypto.news reported](https://crypto.news/bitcoin-fed-holds-rates-october-2026/?referrer=www.coinrithm.com/en/cryptocurrencies/near-protocol&ref=blog.presolt.com) that the vote was 10 to 2, with Lisa Cook and Beth Hammack dissenting in favor of another quarter-point hike. Large exchange outflows in the days before the meeting drew a lot of attention, which makes this a useful moment to look at what those flow numbers can and cannot support.

## A worked example using four published estimates

Here is how much Bitcoin different sources say is sitting on exchanges right now.

- [Santiment data cited by Yahoo Finance](https://finance.yahoo.com/markets/crypto/articles/whales-bought-86-702-bitcoin-203017644.html?ref=blog.presolt.com) puts exchange holdings at about 6.5% of supply, or around 1.3 million BTC.
- A daily dashboard published by [TokenPost](https://www.tokenpost.com/news/investing/41700?ref=blog.presolt.com) showed exchange reserves of 2,490,619 BTC on October 28.
- CryptoQuant data reported by [TokenPost](https://www.tokenpost.com/news/investing/31721?ref=blog.presolt.com) put exchange supply at 2.68 million BTC, down from 2.726 million on September 20.
- A dataset cited by [Crypto Briefing via TradingView](https://www.tradingview.com/news/cryptobriefing:889b2420b094b:0-bitcoin-balances-on-centralized-exchanges-climb-by-11-593-btc-in-three-days/?ref=blog.presolt.com) put combined centralized exchange holdings at about 3.67 million BTC.

Now value each one at the October 29 price of $79,704.98, multiplying the coin count by the price:

- 1.3 million BTC is worth about $103.6B.
- 2.49 million BTC is worth about $198.5B.
- 2.68 million BTC is worth about $213.6B.
- 3.67 million BTC is worth about $292.5B.

The gap between the lowest and highest estimate is roughly 2.37 million BTC, or close to $189B at today's price. That is larger than the market capitalization of any asset in our tracked list other than Bitcoin and Ether.

The daily flow numbers show the same pattern on a smaller scale. For October 27, the CryptoQuant and Santiment netflow figures differ by about 2,372 BTC. To find the size of that gap, subtract the smaller figure from the larger one and divide by the smaller figure. The result is a difference of about 19% for the same day.

None of this means the data is unreliable. It means each figure answers a slightly different question: which exchanges are covered, which addresses are counted, and over what time window.

## Direction matters more than the level

Because the levels vary so much, the most useful signal is usually the trend within one provider's series, not a comparison between providers.

Here the sources tell a more consistent story. CryptoQuant's figures show a 1.7% decline in exchange supply between September 20 and late October. [Cointribune, citing Santiment](https://www.cointribune.com/en/bitcoin-reserves-on-exchanges-continue-to-decline/?ref=blog.presolt.com), reported that exchange holdings fell from 7.1% of supply a year ago to about 6.5%, and that 24,073 BTC left exchanges on October 5, the largest daily outflow since March 1\. Glassnode found that among 11 large exchanges, six hold less Bitcoin than a year ago, and the ten exchanges other than the largest together hold about 10% less.

Different methods, same direction over months: less Bitcoin is sitting on most exchanges than a year ago.

## Proof of Reserves adds a second view

Proof of Reserves reports are published by exchanges themselves. They compare what an exchange holds with what its users have deposited. They answer a solvency question rather than a supply question, but they also feed the address lists that analytics firms use.

Glassnode noted that Gate's October 18 snapshot showed about $1.4B more in reserves than user balances. [TokenPost reported](https://www.tokenpost.com/news/business/27832?ref=blog.presolt.com) that Binance's October data showed a 100.35% reserve ratio for Bitcoin, with user Bitcoin balances up 0.91% between September 1 and October 1.

Glassnode's report also highlights a structural point. Growth at the largest exchange kept the total from falling, even as most others shrank. If that continues, the Bitcoin that remains on exchanges becomes concentrated at a few venues. A single exchange reshuffling its wallets could then move the aggregate number noticeably.

## What the data does and does not tell us

What it does tell us:

- Over months, most measures show less Bitcoin held on exchanges than a year ago.
- Large single-day outflows, such as October 5 and October 27, did happen across more than one provider's data.
- Several large exchanges publish reserve ratios above 100% for Bitcoin.

What it does not tell us:

- Why coins moved. An outflow can be a buyer moving to cold storage, an exchange moving coins to a new unlabeled wallet, or a fund shifting to a custodian. On-chain data alone cannot separate these.
- That selling or buying has happened. Inflows mean coins are available on an exchange, not that they were sold. Outflows mean coins left, not that new demand arrived.
- An exact total. With estimates ranging from about 1.3 million to 3.67 million BTC, any single "Bitcoin on exchanges" number should be read as one provider's view, not a fact about the network.
- Where prices go next. Exchange reserves describe positioning. They are not a forecast.

## Why this matters

Exchange flow headlines move quickly and often arrive without context. A "$1 billion exodus" sounds decisive, but the same day can be reported as 12,570 BTC or 14,942 BTC depending on the source, and the reason behind the move is usually unknown.

A disciplined approach is simple:

1\. Stick to one provider's series when looking at trends.

2\. Compare direction across providers rather than levels.

3\. Check whether a large move lines up with a known custody change, an exchange wallet migration, or an ETF creation before treating it as investor behavior.

4\. Pair flow data with other evidence, such as spot volume, ETF flows, and holder cohort data, before drawing conclusions.

Used this way, exchange reserves are a helpful lens on how holders are positioned. Used as a single headline number, they can mislead.

## Explore the data with Presolt

Presolt brings on-chain metrics, market data, and research together with clear methodology notes, so you can see what each number measures before you rely on it. Explore exchange flows and other on-chain indicators at [Presolt.com](https://www.presolt.com/?ref=blog.presolt.com).

---

This article is for informational and educational purposes only and does not constitute financial, investment, legal, or tax advice. Digital assets are highly volatile and can lose value. Figures are drawn from the third-party sources cited and may be revised. Past patterns do not guarantee future results. Do your own research and consult a qualified professional before making any investment decision.