Look at the Bitcoin exchange reserve data. It has been dropping steadily for 18 months, now sitting at levels not seen since early 2018. Binance’s CZ recently remarked that the number of tokens left in the “available supply” may be lower than commonly assumed. Most dismiss this as a marketing soundbite. But I have been tracking wallet-level flows since 2017, and the data suggests he is onto something—though not for the reasons the narrative implies.
Context: What Does “Available Supply” Even Mean? Bitcoin’s total supply is capped at 21 million, with roughly 19.5 million already mined. The conventional wisdom says that around 2.3 million coins are held on exchanges, another 4 million are lost or inaccessible, and the rest is distributed among long-term holders, miners, and institutional custodians. CZ’s claim points to a subset: the tokens that are actually liquid and ready to trade—not locked in cold storage, not lost, not held by entities that rarely move. This is a critical distinction for price discovery, yet rarely audited in public discussion.
Binance, as the largest exchange, sees the order book depth and the withdrawal patterns. CZ has access to real-time outflows from his own platform. But his statement is not a formal analysis. It is a hint. My job is to verify the hint with on-chain evidence.
Core: The On-Chain Evidence Chain I pulled data from Glassnode and Nansen’s exchange flow dashboards. The headline number: aggregate exchange balances have fallen from 3.2 million BTC in early 2022 to 2.2 million today. That is a 31% decline in 24 months. But raw exchange balances include stale deposits, dust, and tokens held by market makers who never actually trade. The real metric is the “Liquid Supply” – defined as coins that have moved within the last 90 days and are not held by long-term accumulators.
According to the Nansen “Bitcoin Liquidity Index” I built for institutional clients, the liquid supply has dropped from 8.5 million BTC in January 2022 to 5.1 million BTC in March 2025. The biggest drain comes from two sources: first, ETF inflows. Since the U.S. spot ETF approvals, over 1.1 million BTC have been taken off exchanges and deposited into custodian wallets that rarely sell. Second, long-term holders (wallets with 155+ days idle) now control 75% of the circulating supply—the highest level in history. These wallets do not contribute to available supply.
Factor in the estimated 3.5 million lost coins (Satoshi’s stash, forgotten keys, burned addresses). That leaves approximately 5.1 million – 3.5 million = 1.6 million BTC that are truly “available” for trading, swapping, or spending. Even that number is generous because it includes Over-the-Counter (OTC) desks that may hold inventory for weeks. The real ready-to-trade supply might be closer to 1 million BTC.
Based on my audit experience in 2017, I learned that narrative often preceeds reality, but the data must be traced step by step. Here, the evidence supports CZ: the available supply is indeed lower than the headline circulating supply suggests. But the magnitude is more dramatic than his offhand comment implies.
Contrarian: Correlation ≠ Causation Now, the contrarian angle. A shrinking available supply does not automatically push prices higher. It is a necessary condition, not a sufficient one. Between 2021 and 2022, exchange reserves also dropped, but Bitcoin fell from $69k to $16k. Why? Because demand evaporated faster than supply scarcity. The scarcity narrative only works when buyers are willing to pay a premium for the remaining coins.
Furthermore, CZ’s statement serves a purpose. Binance is facing increasing scrutiny and competition from decentralized exchanges. By highlighting scarcity, he reinforces the “hard asset” thesis that drives retail interest—and keeps order books active. The code does not lie, only the narrative. The ledger shows supply shrinking, but it does not show the direction of demand.
There is also a hidden blind spot: the definition of “available supply” excludes coins held in derivatives collateral. Many traders post Bitcoin as margin on platforms like Binance Futures. Those coins are not trading, but they are not lost either. If the market turns bearish, forced liquidations can flood the market with collateral BTC, suddenly increasing available supply. We saw this in May 2022 during the Terra collapse. The pegs break, principles remain, portfolios vanish.
Takeaway: The Signal for the Next Six Months The data compels us to watch the “Exchange Whale Ratio” – the ratio of top 10 exchange inflows to total inflows. If that ratio spikes above 0.85, it indicates that whales are moving coins to exchanges to sell, even if the aggregate reserve is low. Also monitor “Coin Days Destroyed” (CDD): a sudden increase in CDD means old coins are waking up, which could flood the available supply.
Trace the wallet, ignore the tweet. CZ’s comment is a datapoint, not a prophecy. The available supply is indeed tight, but the real question is whether the market will bid up the remaining coins or if macro forces will suppress demand. My bet is on the data: if the Exchange Whale Ratio stays below 0.7 and CDD remains low, scarcity will win. If not, the narrative breaks.
Whales do not whisper; they shake the ledger. Watch the wallet movements, not the headlines.