On-chain analysis identifies the percentage of total asset supply held on known exchange wallets as a primary long-term behavioral signal, where rising figures indicate movement toward liquidity and potential sale while falling figures suggest the opposite. Data compiled by Woofun AI shows that a dual-asset chart tracking this metric for BTC and ETH, updated daily via Sanbase from May 2025 to May 2026, reveals two lines that declined in tandem for most of the year before diverging in May 2026. Bitcoin's exchange supply currently sits at 5.623%, confirmed as the lowest level since 2018, representing an 8-year low. The chart's yellow line has declined steadily from approximately 9.5% at the start of the period, with annotations indicating it has stalled at current levels rather than reversing. Conversely, Ethereum's exchange supply stands at 4.575%, up from 4.2% ten days ago, marking a rise of 0.4 percentage points. While this remains near the lowest level since ETH began public trading in 2015, the directional shift is significant. When two assets near historically low exchange supply levels diverge, with one moving upward while the other holds flat, the divergence illuminates behavioral differences between holder bases more than absolute levels do individually.

The mechanism behind this divergence is detailed in CryptoQuant exchange netflow charts covering April 15 to May 16, 2026. For Bitcoin, the netflow chart displays a volatile but roughly balanced pattern of inflows and outflows, with the most recent reading at -967.7 BTC indicating net outflow. A notable detail observed is that May 11 through May 14 produced a sequence of positive green bars in the +2.5K to +4K BTC range daily, yet the supply percentage held flat at 5.6% across the same period. Woofun AI notes that BTC exchange supply holding flat at 5.6% while recent netflow bars show positive inflows implies Bitcoin is moving onto exchanges but being withdrawn at a matching rate. This dynamic describes an active holder base absorbing exchange deposits rather than a passive one simply refraining from selling. The stability in supply percentage is not an absence of activity but the result of two opposing flows canceling each other out, with withdrawal pressure keeping pace with deposit pressure.

For Ethereum, the mechanism driving the change is more direct. The chart shows May 10 produced a green bar of approximately 240,000 ETH, the largest single inflow bar on the chart. The days that followed, May 12 through May 15, show predominantly red bars and a current reading of -14.2K ETH outflow. ETH's 0.4 percentage point rise in exchange supply over ten days traces directly to the May 10 inflow spike of approximately 240,000 ETH, the largest single-day inflow recorded, and the subsequent outflow bars have not yet been large enough to reverse what one session produced. The behavioral picture emerging from combining both assets' supply and netflow data is structurally different for each. For Bitcoin, supply discipline is active rather than passive; holders are not simply leaving BTC in cold storage and ignoring exchanges but are depositing and withdrawing at roughly equal rates. This keeps the exchange supply percentage stable at a level historically associated with reduced selling pressure.

The 8-year low in Bitcoin exchange supply, maintained through a period that included a PPI-driven price shock and a subsequent recovery above $82,000, suggests the holder base is not being shaken out by short-term volatility. For Ethereum, the May 10 inflow spike is the event that altered the landscape. Whether that spike represented profit-taking by holders who accumulated ETH below $2,000 during February and March 2026 and were still in profit at May prices, or a separate cohort moving supply for other reasons, the result is identical: exchange supply rose, and outflows since have not reversed it. The current ETH exchange supply at 4.6%, while still near historical lows, is moving in the wrong direction for the same thesis that BTC's flat supply supports. Woofun AI analysis suggests that a return of ETH exchange supply to 4.2% or below, driven by sustained daily net outflow bars, would indicate the May 10 spike has been absorbed and ETH holders are returning to net outflow behavior.

Conversely, a continuation of ETH exchange supply above 4.5% through the end of May, accompanied by further positive netflow spikes comparable to May 10, would indicate the directional divergence from BTC is structural rather than temporary. This scenario would imply the ETH holder base is in a distribution phase the BTC holder base is not replicating. The critical question raised by these supply percentages is not whether the levels are high or low in isolation but why the two are moving differently at the same moment. The data underscores that while both assets remain near multi-year lows in exchange holdings, the underlying mechanics of capital flow reveal a fundamental split in market sentiment and holder strategy between the two leading cryptocurrencies.