One Fund Took 81 Cents of Every Dollar — and Shed $4.88B Over the Prior Three Months

U.S. spot Bitcoin ETFs pulled in $853.54 million in net inflows across five trading days from August 3 through August 7 — the strongest weekly showing since mid-April. Not one session printed negative. The daily sequence ran $170.1 million, $211.5 million, $244.4 million, approximately $128.8 million, and roughly $98.85 million.

BlackRock's iShares Bitcoin Trust accounted for $693 million of that total. That is 81 cents of every dollar entering the category landing in a single fund.

The concentration was even sharper at the start of the run. Over the Monday-through-Wednesday stretch, IBIT alone captured $479 million of $626 million in total inflows — roughly 76% — including a single-day haul of $197 million on August 5 that represented nearly 80% of that session's entire category intake. Fidelity's FBTC added $19.6 million across those three days, ARK 21Shares' ARKB brought in $9.2 million and Bitwise's BITB added $8.7 million. VanEck's HODL went the other way with $14.7 million of outflows on August 5 and $32.8 million on Thursday.

The forecast here rests on a structural point that almost every flow commentary misses. IBIT is not the largest participant in the Bitcoin ETF market. It is the Bitcoin ETF market, and that concentration operates identically in both directions.

Consider the trailing data. IBIT's five-day net flows stand at $539.33 million and its one-month figure at $925.63 million. Its three-month net flow is negative $4.88 billion. Its six-month figure is negative $824.88 million. Its one-year figure is positive $3.66 billion.

Set the three-month number against the category. U.S. spot Bitcoin ETFs recorded $5.4 billion of net outflows during the first half of 2026 — their first negative half-year since launching in early 2024. IBIT shed $4.88 billion over three months on its own.

One fund produced nearly all the bleeding and is now producing nearly all the recovery. Everything below develops what that means for reading this data.

The Category Now Holds 6.10% of All Bitcoin

Scale first, because the structural position matters more than any weekly figure.

The five-day streak pushed cumulative net inflows across all U.S. spot Bitcoin ETFs to $52.18 billion since launch. Total net assets in the category climbed to $79.50 billion, representing roughly 6.10% of Bitcoin's entire market capitalisation.

Six percent of a monetary asset held in regulated wrappers with daily creation and redemption mechanics is a genuinely new market structure. It did not exist before January 2024.

The mechanical significance is that these flows are real rather than notional. ETF inflows require authorised participants to buy actual Bitcoin on the open market to back new shares, so a run like last week's represents genuine demand hitting spot liquidity rather than speculative positioning on a derivatives desk. That distinction is why analysts treat ETF flow data as one of the cleaner real-time gauges of institutional sentiment.

It also means the reverse is true and equally mechanical. When shares are redeemed, authorised participants deliver Bitcoin back and it enters circulating supply. A $5.4 billion first-half outflow was $5.4 billion of actual coins pushed onto the market, and the price reflected it — Bitcoin fell 33% in the first half of the year, dropping below $60,000 by the end of June.

That is the causal chain the flow data captures, and it is the reason to take it seriously as a forecasting input rather than as sentiment colour.

The composition of the $52.18 billion cumulative figure carries one important caveat. Grayscale's older GBTC product has shed a cumulative $27.47 billion since its ETF conversion in early 2024. So the net figure across the complex conceals an enormous internal rotation: roughly $80 billion of gross inflows into the newer low-cost products against $27.47 billion of redemptions from the legacy trust.

That rotation is largely complete, which changes the forward interpretation. For the first two years, negative GBTC flows offset positive IBIT flows and the net number understated genuine new demand. With GBTC's conversion-related exodus mostly finished, current net figures are closer to a clean read on incremental allocation — and the clean read for 2026 has been negative.

The Year-to-Date Deficit Is $4.5 Billion and That Is the Real Number

One strong week does not undo a bad year, and the arithmetic deserves stating plainly.

Bitcoin ETFs remain roughly $4.5 billion in the red on a year-to-date basis. That deficit explains much of the selling pressure that dragged the market lower earlier in 2026, and it is the context that keeps the current rebound in perspective. Even after the biggest weekly haul since mid-April, the category has substantial ground to make up before turning net-positive for the year.

Run the recovery maths. At $853.54 million per week, closing a $4.5 billion deficit takes roughly five and a half more weeks of identical performance. The category has not managed two consecutive weeks at that pace at any point in 2026.

The July precedent is instructive about durability. A seven-session inflow streak during that month briefly restored confidence, collecting nearly $1 billion — then snapped abruptly on July 24 with a single-day outflow of $225.18 million. The week immediately preceding the current run recorded $61.5 million of net outflows. So the pattern all year has been three to seven positive sessions followed by an abrupt reversal, and the current five-day streak sits exactly within that historical shape.

The threshold analysts identify as meaningful is higher than what has been delivered. Sustained weekly inflows above $1 billion are viewed as necessary for a significant price rally. The category hit that level once in 2026 — the July streak — and it reversed within days.

That gives a clean benchmark for the forecast. Last week's $853.54 million is 85% of the level required to move price materially, achieved in a week when Bitcoin gained roughly 2% and stayed inside a $62,000 to $67,000 range. The price response confirms the threshold: sub-$1 billion weeks stabilise rather than propel.

The comparison against the bull-market period is starker still. During the advance that carried Bitcoin from roughly $75,000 in April 2025 to a record $126,000 by October 2025, weekly inflows exceeded $1 billion on multiple occasions. That is the flow regime associated with re-rating, and the current pace is not it.

For anyone modelling a return toward the record, the flow requirement is roughly double what August has produced.

IBIT's Three-Month Outflow of $4.88 Billion Reframes the Whole Dataset

This is the single most useful figure in the entire Bitcoin ETF complex and it is almost never cited.

IBIT's trailing net flow data reads: five days positive $539.33 million, one month positive $925.63 million, three months negative $4.88 billion, six months negative $824.88 million, one year positive $3.66 billion, three years positive $60.96 billion.

Read the three-month and six-month figures together. IBIT lost $4.88 billion over three months but only $824.88 million over six. That means the fund was strongly positive in the earlier three months of that window and catastrophically negative in the most recent three. The damage was concentrated in the second quarter and early third quarter of 2026.

Now compare with the category. The entire complex recorded $5.4 billion of net outflows in the first half. IBIT accounts for $4.88 billion of outflow over its trailing three months. Depending on exact period alignment, one fund represents something close to the entirety of the category's bleeding.

The implication is significant and it runs against how flow data is normally used. When commentary describes "institutional demand returning to Bitcoin," the underlying observation is that BlackRock's distribution channel turned positive. That is a narrower signal than the phrasing suggests — it reflects the allocation decisions of clients accessing Bitcoin through one relationship, on one platform, with one set of model portfolios and one set of advisory recommendations.

IBIT has represented between 70% and 80% of the daily inflow total within the broader U.S. spot Bitcoin ETF sector since its January 2024 launch, and last week's 81% continued that pattern. So this is not an anomaly. It is the structure.

Two consequences follow for the forecast.

First, the flow indicator has lost breadth information. A week of $853.54 million led 81% by one product tells you far less about the depth of institutional adoption than the same total distributed evenly across six issuers. Breadth is what makes a bid durable, and this bid has none.

Second, the reversal risk is concentrated. If IBIT's channel turns — through a model portfolio rebalance, a platform-level allocation change or a shift in advisory guidance — the outflow arrives at the same 81% concentration with no diversified base of smaller issuers to absorb it. That is precisely what the negative $4.88 billion three-month figure documents.

Watching Bitcoin ETF flows now means watching IBIT. Everything else is noise around it.

HODL's Outflows Show Capital Is Rotating, Not Just Arriving

The divergence inside last week's numbers reveals a selection process that the headline total conceals.

While IBIT took $693 million and FBTC, ARKB and BITB added modest amounts, VanEck's HODL recorded $14.7 million of outflows on August 5 and $32.8 million on Thursday. On a day when IBIT captured nearly 80% of total inflows, one competitor was losing assets.

That divergence reveals a structural selection characteristic: capital withdrawing from specific strategies or smaller-scale products and rotating into leading products with greater scale advantages. Investors are showing an extreme preference for low-cost, high-liquidity flagship products.

Understand what that means for interpreting the $853.54 million.

Part of the weekly total is not new money entering Bitcoin. It is existing Bitcoin exposure moving from one wrapper to another. When a holder redeems HODL shares and buys IBIT shares, the category records an outflow at VanEck and an inflow at BlackRock, and the aggregate net figure understates the rotation while overstating the genuine new demand at the leader.

That rotation has been running since launch and it explains the persistence of the 70% to 80% IBIT share. Scale begets liquidity, liquidity begets tighter spreads, tighter spreads attract institutional order flow, and institutional order flow reinforces scale. It is a competitive dynamic with a single stable outcome, and the smaller products are being consolidated out of relevance.

The three-day breakdown quantifies how lopsided it has become. IBIT $479 million, FBTC $19.6 million, ARKB $9.2 million, BITB $8.7 million. The three non-BlackRock products combined took $37.5 million against IBIT's $479 million — a ratio of nearly 13 to 1.

For the forecast, the practical read is that a portion of any weekly headline figure represents wrapper migration rather than incremental Bitcoin demand. That makes the effective new-money number lower than $853.54 million, which reinforces why the price response was a 2% weekly gain rather than a breakout.

It also means the smaller issuers are approaching a viability question. A product losing assets while the category gains them, in a business where scale determines cost competitiveness, has a limited runway.

The Buying Happened Through a $111 Million Security Breach

The timing of this streak is the most informative thing about it, and it reveals who the buyer actually is.

The Coldcard hardware wallet exploit emerged on July 30. Galaxy Research estimated 1,719 Bitcoin worth approximately $111 million had been stolen, warning total losses could exceed $130 million. Bloomberg Intelligence observed that several Bitcoin funds recorded inflows every single day from the moment the exploit surfaced.

The five-day streak ran August 3 through August 7 — entirely inside that news cycle.

A nine-figure self-custody failure is historically the category of headline that produces immediate ETF redemptions, on the reasoning that a compromise of Bitcoin's core custody proposition undermines the investment case. Instead the money kept arriving.

The explanation is structural rather than sentimental. Institutional allocators accessing Bitcoin through IBIT and FBTC have no exposure to a hardware wallet vulnerability. Their coins sit with regulated custodians under institutional controls, audited and insured. A Coldcard failure is a headline about somebody else's operational risk, not a repricing event for their position.

The ETF wrapper has functionally decoupled institutional demand from retail self-custody risk, and last week was the clearest demonstration to date.

The same indifference extended across other crypto-specific news. The BIP-110 soft fork attempt failed over the weekend with 2.53% miner support against a 55% activation threshold, its minority chain stalling at two blocks. The Senate left Washington without voting on the CLARITY Act, pushing crypto market-structure legislation to September. Neither event registered in the flow data.

That indifference is a double-edged property and it defines the forecast risk. Buying that ignores a $111 million exploit, a contested protocol fork and a legislative delay is buying that responds only to macro. Mandate-driven allocation is stable against narrative shocks and highly responsive to rate shocks.

Which is why Wednesday matters more than anything happening inside crypto.

The sentiment backdrop underscores the divergence. The Crypto Fear & Greed Index remains in Extreme Fear territory, indicating investor confidence has not recovered. Institutions bought $853.54 million while retail sentiment sat at extreme fear — a divergence that has historically preceded recoveries and has, so far this year, preceded nothing.

Wednesday's CPI Is What the Flow Streak Is Actually Trading

Every strand converges on one macro release, and the flow data has already shown its sensitivity to it.

July U.S. CPI publishes Wednesday, August 12 at 8:30 a.m. ET, with PPI Thursday and retail sales Friday. Consensus expects the headline annual rate stepping down to 3.4% from 3.5% in June and 4.2% in May, with core at 2.5%.

The rate context is what drives allocation into a non-yielding asset. July nonfarm payrolls contracted by 23,000 against a consensus near 80,000, June was revised down to 20,000 from 57,000, and revisions removed a combined 103,000 jobs. September Federal Reserve hike odds fell to roughly 44% to 46% from about two-thirds a week earlier. The policy rate sits at 3.75% with officials divided on whether to raise as they monitor the Middle East oil shock.

Weak U.S. jobs data released during the streak eased Federal Reserve rate-hike concerns and lifted risk appetite, and the flow acceleration through Wednesday — $170.1 million, then $211.5 million, then $244.4 million — maps directly onto that repricing. Then flows decelerated to $128.8 million and $98.85 million as yields stabilised.

That pattern is the tell. The escalation phase coincided with falling hike probability and the deceleration coincided with the ten-year climbing back to 4.666%.

The scenarios follow.

At or below 3.4% headline with core at 2.5%: hike odds compress further, the ten-year retreats, and the streak has room to extend toward the $1 billion weekly threshold that analysts identify as necessary for a price rally. A favourable print would reinforce the narrative that the Fed has room to pause, potentially fuelling further allocation.

At 3.6% or higher: hike odds return toward two-thirds with Brent near $85 and the strait shut into a sixth month. A hotter-than-expected number would test the newfound momentum directly, and the July 24 precedent — a $225.18 million single-day outflow ending a seven-session streak — shows how abruptly these runs reverse.

The disinflation arithmetic argues for caution. June's headline reached 3.5% on a negative 0.4% monthly print driven by energy base effects that do not repeat with September WTI at $80.90.

Market attention has pivoted entirely to the CPI release as the next major variable that could sway sentiment and influence flow trajectory in the weeks ahead. That is an accurate description of where the marginal decision sits.

The Ether Complex Is Running the Same Pattern With Better Breadth

Cross-asset flow comparison sharpens the read on what is happening, and Ethereum's version is instructive.

Spot Ether ETFs recorded four consecutive days of net inflows during the same stretch, totalling $244.9 million for the week and extending a positive run to five weeks — the longest of 2026. On August 4, when Bitcoin ETFs took $211.5 million, Ethereum ETFs added $53.1 million. Thursday's $92.2 million was the largest single day of the Ether period.

The Ether side follows an almost identical pattern to Bitcoin's, and the monthly progression shows a genuine regime change: July delivered over $365 million after June shed $528 million and May shed $540 million.

Two differences matter for interpretation.

Ethereum's streak is longer. Five consecutive positive weeks against Bitcoin's five consecutive positive days is a materially more durable trend, and it has survived multiple weekly turns rather than sitting inside a single one.

Ethereum's structural bid is deeper. The staking ratio has climbed to 34.23% of supply — roughly 41 million coins with a staked market capitalisation above $77 billion — while exchange balances sit at multi-year lows. Bitcoin has no equivalent supply lock beyond the ETF holdings themselves.

Against that, Ethereum trades 61% below its August 2025 record of $4,951.66 while Bitcoin sits roughly 48% below its $126,000 October 2025 high. Better flow, worse price — because Ethereum's median mainnet fees fell more than 99% to under $0.02, gutting the burn that made the asset deflationary.

Further down the risk curve the flow picture collapses entirely. Solana and XRP ETF products logged effectively zero net change on August 7. All six U.S. Solana ETFs recorded zero net flow across five consecutive sessions from July 29 through August 4. XRP ETF inflows have fallen under 1% of daily spot volume despite $1.53 billion in assets.

That ordering — Bitcoin strongest, then Ethereum, then nothing — tells you the rotation is a flight to the simplest monetary asset rather than a broad risk-on move in digital assets. Institutions are separating broad crypto risk from targeted altcoin exposure, and last week they chose the top of the stack.

For the forecast, that concentration is bullish for Bitcoin's relative position and bearish for any read that treats this as a sector-wide recovery.

What Franklin Templeton Is Actually Pointing At

The regulatory catalyst cited as driving renewed appetite deserves precise treatment because its timing has already slipped.

Part of the renewed appetite appears regulatory, with Franklin Templeton pointing to the prospect of federal crypto market-structure rules as a potential turning point — a shift that could eventually open bank balance-sheet liquidity to the asset class for the first time.

That framing identifies the largest untapped demand pool in the entire thesis and it deserves to be understood mechanically.

Currently, banks cannot hold Bitcoin on balance sheet in meaningful size because capital treatment and regulatory permission are unsettled. Every dollar in the $79.50 billion ETF complex came from asset managers, wealth platforms, family offices, corporates and retail — not from commercial bank balance sheets. Bank treasuries manage assets measured in the tens of trillions globally, and a rule set that permitted even fractional allocation would dwarf the entire existing ETF category.

That is why the market-structure legislation is treated as a structural rather than a cyclical catalyst.

The complication is that it has not arrived. The Senate left Washington for its August work period without voting on the Digital Asset Market Clarity Act, pushing the vote to September. So the catalyst driving part of last week's appetite is a prospect rather than an event, and the specific legislative vehicle just slipped by a month.

The September calendar concentration is worth flagging. The Clarity Act vote and the Federal Reserve's September 16 decision fall in the same month, which makes that period the genuine pivot for the category rather than August.

The forecast implication for flows is that the current streak is running on rate expectations with a regulatory option attached, and the option has been pushed out. If Wednesday's CPI disappoints, there is no legislative catalyst available to offset it before September.

There is a further consideration on how such rules would land. Bank balance-sheet access would arrive through custody and prime brokerage relationships rather than through retail ETF wrappers, which could redirect institutional flow away from IBIT rather than into it. The 81% concentration exists partly because the ETF is currently the only compliant access route for many allocators.

That is a 2027 consideration. For now it is the strongest argument in the bull case and it is not in the price.