Total gross activity reached $7,070.7M across buy and sell pressure combined. Net flow across all assets landed at -$179.4M, technically a slight distribution week in aggregate. But that headline number conceals a story of extraordinary internal divergence. Strip out the ETH selling and the picture inverts entirely. Add it back and it barely tips negative. The real intelligence is not in the net number — it's in where the flow was concentrated, in which direction, and on which platforms.

The week's tone: BTC accumulation, ETH distribution, stable rotation through USDC, and a market that is selectively, deliberately constructive — not universally bullish.

| Metric | Value | |---|---| | Total Buy Pressure | $3,466.8M | | Total Sell Pressure | $3,646.2M | | Net Weekly Flow | -$179.4M | | Total Events Analyzed | 2,520 | | Classified Imbalances | 452 | | Imbalance Rate | 17.9% | | BTC Net Flow | +$674.9M | | ETH Net Flow | -$700.2M |

1. +$674.9M — BTC net accumulation. Buy volume of $1,795.8M versus sell volume of $1,120.9M. This is not a contested market. Buyers dominated BTC in absolute terms. When the largest single imbalances hit 86-93% buy ratios on volumes exceeding $150M each, you are watching institutional conviction play out in real time.

2. -$700.2M — ETH net distribution. Sell volume of $1,357.3M against buy volume of only $657.1M. Average buy ratio of 42.0% means sellers had the upper hand in more than half of all measured ETH sessions. The $158.9M single-event sell at 95% ratio on Hyperliquid and Bitunix was the sharpest single institutional sell of the week across any asset.

Previous Week Comparison: Without Week 14 data for direct comparison, the internal structure of Week 15 suggests a continuation of BTC outperformance trends that have characterized Q1 2026. The aggressive buy-side concentration in Hyperliquid perpetuals specifically points to leveraged long positioning being built or extended — not spot accumulation alone. The composition of buying (futures-heavy, spot-supplemented) is consistent with institutions adding directional exposure, not just storing value.

The 52.0% average buy ratio across all BTC events understates the actual directional character of the week. When you look at the classified imbalances — the events that exceeded threshold — the ratios read 86%, 93%, 91%, 86%, 89%, 89%. These are not borderline reads. These are sessions where buyers showed up with a mandate and executed without much resistance. The 93% buy ratio event on Hyperliquid, Bybit, and Bybit Spot with $235.5M in volume is particularly notable — multi-venue coordination at that scale implies either pre-scheduled accumulation or responsive buying to a specific catalyst.

Exchange leadership: Hyperliquid (5 of 6 BTC imbalance events), Coinbase (2), Binance Futures (2), Bybit (2), OKX (2).

Interpretation: BTC is in active accumulation by multiple institutional cohorts operating across different infrastructure. The simultaneous presence of regulated (Coinbase) and unregulated (Hyperliquid) venues in coordinated events signals broad-based conviction, not a single actor. This is what institutional FOMO looks like when it's trying to be disciplined.

What can be said: the absence of notable altcoin individual events in the top 10 imbalances suggests buy pressure outside BTC was fragmented across many assets and venues rather than concentrated in a single second-layer bet. That fragmentation is characteristic of a market in early positioning, not a market in full rotation mode.

This is the week's most important distribution signal, and it cannot be characterized as noise. $1,357.3M in ETH sell volume against $657.1M in buy volume is a 2.06:1 sell-to-buy ratio. The average buy ratio of 42.0% means that across the week's measured ETH sessions, sellers held the majority position in a clear majority of events.

The defining moment: the $158.9M single-event ETH sell at 95% ratio on Hyperliquid and Bitunix. A 95% sell ratio at $158.9M in size is not a stop-loss cascade. It is not panic selling. It is a large, coordinated, pre-meditated distribution event. The venue combination — Hyperliquid perpetuals plus Bitunix — points toward Asian institutional flow, potentially crypto-native funds with Hyperliquid exposure reducing ETH delta through the derivatives stack.

Meanwhile, the $212.5M ETH buy event at 85% ratio on KuCoin, Bybit, and Binance represents the opposing side: there are buyers in ETH. But they are outnumbered, outgunned, and positioned on different venues than the sellers. The buying appears retail-heavy and derivatives-light relative to the selling, which is futures-concentrated.

Exchange leadership for ETH selling: Hyperliquid (confirmed), Bitunix (confirmed). The presence of Bitunix — a relatively newer venue with strong Asia-Pacific flow — suggests this selling originated from a cohort that is relatively newer to the institutional stack.

Interpretation: ETH is being distributed by sophisticated actors who hold large futures positions. The counterpart buying is real but insufficient to absorb it. This does not necessarily mean ETH is in a terminal downtrend — but it does mean institutional holders are reducing exposure at current prices, and the rotation appears to be into BTC (and possibly stablecoins) rather than into altcoins.

The remaining $1,764.4M in sell pressure beyond ETH and the USDC event is distributed across the altcoin complex. This represents a significant pool of distribution activity — roughly 48% of total sell pressure. The fragmentation pattern mirrors the buy side, suggesting broad-based altcoin churn rather than concentrated exits from specific tier-2 or tier-3 assets.

The six BTC accumulation events cluster into three distinct phases based on their venue compositions and size:

Phase 1 — Institutional Spot Entry (Est. Monday-Tuesday): The $292.6M Coinbase/Hyperliquid/Binance Futures event and the $191.6M Hyperliquid/Binance/Bitunix event read as early-week positioning. Coinbase's presence suggests US regulated entities. The Binance Futures component suggests simultaneous hedge-adjusted spot buying — an institution buying spot while managing delta through futures. This is not speculative behavior. This is allocation management.

Phase 2 — Perpetuals Conviction (Est. Wednesday-Thursday): The $235.5M event at 93% buy on Hyperliquid/Bybit/Bybit Spot is the week's single most concentrated BTC buy event by ratio. 93% means virtually no meaningful counter-pressure. Combined with the $160.4M Hyperliquid/OKX event at 86%, this mid-week cluster represents the peak of buying conviction — likely triggered by either a price break through a key level or a macro catalyst that cleared a decision threshold.

Comparison to Recent Weeks: Without Week 14 data for direct numerical comparison, the qualitative comparison is straightforward: the venue diversification across this week's BTC accumulation — Coinbase, Hyperliquid, Binance, Bybit, OKX, Bitunix — suggests a broadening of participants rather than a concentration. More venues means more actors. More actors means consensus, not a single whale blowout.

What This Positioning Means: Institutions buying BTC in futures and spot simultaneously, across both US-regulated (Coinbase) and DeFi-native (Hyperliquid) venues, with no significant sell-side pushback in classified imbalances, means one thing: there is no institutional seller willing to size up against this flow at current prices. When supply is absent at the imbalance level, price discovery moves in the direction of the aggressor. The aggressor this week was unambiguously the buyer.

The asymmetry here is stark. The largest ETH buy of the week ($212.5M at 85%) is smaller than the largest ETH sell ($158.9M at 95%), and the sell carries a higher conviction ratio. But more telling than any single event is the aggregate: $700.2M more ETH was sold than bought across all measured sessions.

Early Week: The $212.5M buy event on KuCoin, Bybit, and Binance likely represents continued dip-buying by retail-adjacent institutional players who are still constructive on ETH. KuCoin's dominance in this event — a venue with heavy retail institutional overlap — supports this characterization.

Mid-to-Late Week: The $158.9M sell at 95% on Hyperliquid and Bitunix likely occurred mid-to-late week as ETH failed to sustain momentum after the early buy. The 95% sell ratio is a capitulation-adjacent read: whoever sold this block was not testing the market. They were exiting.

ETH vs BTC Divergence — The Defining Theme of Week 15:

The BTC-ETH divergence this week is not subtle. BTC: +$674.9M net. ETH: -$700.2M net. The spread is roughly $1.375 billion. These flows are inverses of each other, which is not coincidence. This is rotation. Capital is leaving ETH and entering BTC. The vehicle is institutional. The scale is meaningful.

The key question for Week 16: does ETH find institutional support at lower levels, or does the distribution continue? The 42% buy ratio average suggests we are not yet at a level where buyers are overwhelmed by conviction on the upside. Until that ratio recovers to 50%+, ETH remains structurally disadvantaged in the institutional flow picture.

The BTC accumulation this week was multi-jurisdictional. Coinbase (US-primary) appeared in two of the six buy events. Hyperliquid (globally distributed but with strong Asia-Pacific and crypto-native institutional base) appeared in five. Bybit (primarily Asia-Pacific) appeared in two. OKX (globally distributed, strong Asia-Pacific) appeared in two. This geographic breadth of buying is significant. BTC accumulation driven only by US institutional flow can reverse when US macro shifts. BTC accumulation shared between US and APAC institutional cohorts is more durable because the two cohorts respond to different macro drivers.

Pattern 5: ETH Selling is Futures-First

The structural setup going into Week 16 is asymmetric. BTC has institutional accumulation momentum with no meaningful sell-side imbalance response. ETH has institutional distribution momentum that has not yet found a floor at the imbalance classification level. Until one of those two dynamics changes — until BTC buy-side pressure decelerates or ETH sell-side pressure reverses — the path of least resistance remains: BTC stable-to-higher, ETH stable-to-lower.

2. ETH buy ratio recovery — ETH's 42% average buy ratio this week is the signal to watch. A recovery toward 50%+ in Week 16 would indicate either: buyers are gaining confidence at lower levels, or the distribution campaign is completed. Either is bullish for ETH. A continuation below 42% would suggest further structured selling.

Week 15's BTC accumulation pattern — US regulated venues alongside DeFi perpetuals — is consistent with a market anticipating either a macro tailwind (rate expectations, dollar weakness, risk-on regime) or an asset-specific catalyst (ETF inflow acceleration, corporate treasury announcements, regulatory clarity). The coordination implied by multi-venue simultaneous events suggests actors with advance knowledge of their own demand schedule, which points toward pre-planned allocation rather than reactive buying.

The net negative overall flow (-$179.4M) means the market is not irrationally exuberant. There is real selling happening — it's concentrated in ETH and distributed across altcoins — and that selling is partially absorbing the BTC buying energy at the macro level. This is a healthy structure for a trending market: enough distribution elsewhere to keep prices honest while the primary accumulation target (BTC) is being stacked.

2. ETH selling accelerating to weekly volumes above $2B would signal capitulatory distribution — potentially a buying opportunity or potentially a structural break, depending on the macro context at that time.

None of these conditions are present at Week 15 close. The baseline going into Week 16 is cautiously constructive for BTC, structurally cautious for ETH, and attentive to altcoin rotation signals in the residual flow data.

Ethereum was distributed with equal conviction by a different cohort of institutional actors. The 95% sell ratio event, the 42% average buy ratio, the $700.2M net negative flow — these numbers describe a market where sophisticated holders have decided that current ETH prices are prices worth selling, not holding. The buyers are real, but they are losing.

The total market ended Week 15 with a net -$179.4M — barely negative — which is the statistical expression of a market in active rotation rather than in directional crisis. The money is not leaving crypto. It is moving within crypto. From ETH to BTC. From altcoins to stablecoins and back. From distributed long-tail risk toward concentrated, high-conviction single-asset exposure.

The job next week is simple in principle, difficult in practice: track whether the BTC accumulation extends or completes, whether ETH distribution continues or resolves, and whether the residual $3B+ in altcoin flow begins to concentrate around identifiable targets.