Bitcoin ETF Inflows and IBIT — The Fund Lost 3,511 BTC Last Week Against a 3,170 BTC Category Decline

US spot Bitcoin exchange-traded funds recorded approximately $49.7 million in net outflows on July 28, extending a streak of negative flows to four consecutive trading days.

BlackRock's iShares Bitcoin Trust drove essentially all of it. IBIT saw net outflows of $54.8 million on the day — one of the larger single-day redemptions for a fund that spent most of its first two years as the industry's dominant flow magnet. Grayscale's Bitcoin Mini Trust recorded a modest net inflow of $5.1 million, partially offsetting the pressure.

The rest of the complex did nothing at all. Most products recorded zero flow, which is its own signal. A category where eleven of twelve products register no primary-market activity on a given session is a category where authorised participants have stopped transacting because there is no demand on either side.

The prior session was similar in shape but smaller. On July 27, outflows were led by IBIT at $8.8 million and Fidelity's FBTC at $2.8 million.

Across the trailing seven days through July 28, the complex shed 3,170 BTC — roughly $200.23 million at prevailing prices. One tracker's July 28 daily read showed a marginal positive of 20 BTC, or about $1.23 million, which illustrates how different providers' cut-off conventions produce different daily signs on the same underlying activity. The weekly direction is not in dispute.

The heaviest damage came earlier in the week. July 23 delivered $225 million in net outflows and July 24 delivered $240 million — a combined $465.26 million across two sessions that ended a seven-day positive run.

Bitcoin itself trades around $64,328, up modestly on the day, with the Federal Reserve announcing at 2:00 p.m. ET and a press conference at 2:30. Futures have been pricing roughly a one-third probability of a rate hike.

For a product complex whose flows now explain a meaningful share of Bitcoin's weekly price action, that decision is the single most important input into whether the four-day outflow streak becomes a fifth or breaks.

IBIT Lost More Bitcoin Last Week Than the Entire Category

The most revealing statistic in the recent data is an arithmetic impossibility that turns out to be true.

IBIT lost 3,511 BTC over the week through July 24. The entire category's net decline over the same period was 3,170 BTC. One fund's outflows exceeded the sector's total losses, which means every other product combined was a net buyer while the largest one sold.

The offsetting flows were small but directionally clear. Fidelity's FBTC added 109 BTC and ARK 21Shares' ARKB added 77 BTC. Grayscale's Bitcoin funds shed 10 BTC and Bitwise's BITB lost 27 BTC.

That distribution matters enormously for interpretation. A category-wide institutional exodus would show outflows across every issuer. What the data shows instead is one very large holder base redeeming while smaller, newer products continue accumulating.

The concentration works the same way in reverse and has all year. On July 23, IBIT led outflows with $202 million out of a $225 million category total — 90% of the day's redemptions from a single fund. In early July, when the complex snapped a ten-day outflow streak, IBIT's return with $209.4 million on July 6 was described as the key institutional signal separating genuine re-entry from opportunistic dip-buying, precisely because IBIT is large enough that its direction determines the category.

That is the structural feature anyone reading Bitcoin ETF flow data needs to internalise: this is not a sector with twelve independent signals. It is one signal with eleven rounding errors attached.

When IBIT buys, the sector prints green. When IBIT sells, nothing else is large enough to offset it. Which means "Bitcoin ETF flows" is functionally a proxy for the positioning decisions of a single allocation desk at the world's largest asset manager, and interpreting those decisions as broad institutional sentiment overstates what the data can support.

The 3.3% Recovery: $8.2 Billion Out, $270 Million Back

The scale of the damage sustained since May puts every subsequent inflow figure in proportion, and the proportion is unflattering.

An eight-week outflow streak beginning in mid-May pulled more than $8.2 billion from the category — the longest such streak in the products' history, against a prior record of five weeks. Inside that period, a thirteen-day consecutive outflow run from May 15 to June 3 drained $4.4 billion and flipped 2026's cumulative flows negative for the first time since the January 2024 launch.

June alone was the worst month on record at roughly $4.5 billion.

Bitcoin ETFs have recovered just 3.3% of the $8.2 billion that left by mid-July.

That single figure is the honest summary of the July repair narrative. Three consecutive positive weeks — $197.4 million, $75.67 million and $33.79 million, totalling roughly $307 million — sound constructive until measured against what preceded them. The five-day net flow figure of positive $154.5 million recovers roughly 8.4% of one month's outflows.

The individual session numbers from the May collapse illustrate what the category is climbing out of. May 27 saw $733.5 million leave, with IBIT accounting for $528 million. May 18 delivered $649 million, with IBIT at $448 million. May 13 lost $630.4 million. May 19 saw $331.2 million exit, of which IBIT was $326 million. May 26 lost $334.1 million.

Five sessions inside three weeks removing more than $2.6 billion, with IBIT responsible for the overwhelming majority in each case.

Year-to-date, the complex remains down approximately $4.76 billion to $4.84 billion in net outflows depending on the measurement source.

One analyst framing captured the current state precisely: a return to a bull trend would require renewed demand and improved market conditions. Neither has arrived. What has arrived is the cessation of active bleeding, which is progress without being a turn.

IBIT's Cost Basis Is $83,691 and Bitcoin Trades at $64,328

The single most consequential number in the entire Bitcoin ETF complex sits in a regulatory filing rather than a flow tracker, and it explains the redemption pattern better than any sentiment analysis.

IBIT's quarterly report shows the trust held 783,744 Bitcoin at March 31, 2026, at an aggregate cost of $65,593,608,745. That works out to an average acquisition price of approximately $83,691 per coin.

Bitcoin trades at $64,328.

At current prices, that position is worth roughly $50.4 billion against a $65.6 billion cost basis — an unrealised loss of approximately $15.2 billion across the fund's aggregate holder base.

That figure reframes the flow data entirely. The redemptions are not a sentiment story. They are the mechanical consequence of a very large cohort of investors sitting on losses averaging 23% and choosing, session by session, to reduce exposure. Every rally toward the mid-$60s brings a tranche of holders closer to a break-even exit, and every one of those exits requires the custodian to sell spot Bitcoin.

That is the structural reason rallies in this market have been capped and why the outflow pattern has been so persistent despite Bitcoin stabilising. The selling is not driven by a view on Bitcoin's future. It is driven by the arithmetic of a position underwater.

The comparison with the prior period makes the deterioration visible. At December 31, 2025, the trust held 770,792 Bitcoin at a cost of $64,696,243,884 with a fair value of $67,415,764,355 — a modest unrealised gain. Net assets stood at $67,401,155,244.

By March 31, 2026, net assets had fallen to $53,384,746,718 despite the coin count rising from 770,792 to 783,744.

The fund bought roughly 13,000 more Bitcoin and its net assets fell by $14 billion. That is the entire 2026 experience of this product compressed into one quarter.

783,744 Coins and a Balance Sheet That Went From Gain to Loss in One Quarter

The quarter-over-quarter comparison in IBIT's filings deserves more attention than it received, because it captures a transition that changed the fund's behaviour.

At the end of 2025, IBIT was a product whose holders were collectively in profit. Cost basis $64.70 billion, fair value $67.42 billion. Modest gain, strong momentum, continued creations.

Three months later, cost basis $65.59 billion, fair value $53.40 billion. A $12 billion swing into loss on a position that grew by 13,000 coins.

That transition is the point at which a passive accumulation vehicle becomes a redemption vehicle. Holders in profit tolerate volatility. Holders 20% underwater manage risk, rebalance, and get stopped out by portfolio-level drawdown limits that have nothing to do with a Bitcoin thesis.

The fund's own AUM trajectory tells the same story from a different angle. IBIT held above $54 billion in assets as of March 2026 with more than 777,000 Bitcoin — the largest spot Bitcoin ETF by both assets and trading activity, and the world's largest and most traded Bitcoin exchange-traded product.

The structural history is worth stating because it contextualises the scale. The trust was founded on June 8, 2023, ahead of anticipated regulatory approval. It launched for public trading on January 5, 2024, and the SEC approved the first group of spot Bitcoin ETFs five days later on January 10, 2024. Net asset value is determined using the CF Benchmarks Index, with Coinbase Custody holding the underlying coins — the same custodian used by FBTC and most major US spot Bitcoin ETFs.

Unlike earlier futures-based products, IBIT directly purchases and stores Bitcoin, with shares representing proportional claims on the reserves. That direct structure is what makes the flow mechanism physically real rather than notional.

It is also what makes the $15 billion unrealised loss a live supply overhang rather than a paper figure.

$60.6 Billion Cumulative Into One Fund Against $51.6 Billion for the Whole Complex

There is an apparent contradiction in the cumulative flow data that turns out to be the most illuminating statistic available on this sector.

IBIT's cumulative net inflows since inception stand at approximately $60.6 billion, with an independent tracker placing the figure at $60.35 billion across three, five and ten-year windows — confirming the fund has taken in that amount over its entire life.

The total cumulative net inflows for the whole US spot Bitcoin ETF category since inception sit at $51.626 billion.

One fund has attracted roughly $9 billion more than the entire sector's net total. Which means every other product combined has been a net redemption of approximately $9 billion since launch.

The explanation is Grayscale. GBTC converted from a closed-end trust to an ETF in January 2024 carrying an enormous legacy holder base that had been locked in at a discount for years, and its subsequent redemptions have been the single largest capital outflow in ETF history. Those redemptions are recorded as negative flows against the category total while IBIT's creations are recorded as positive.

Net of that structural rotation, what the ETF wrapper actually accomplished was the migration of existing Bitcoin exposure from a high-fee closed-end vehicle into a low-fee ETF, plus roughly $51.6 billion of genuinely new capital.

Sixty billion dollars of cumulative net creations into a single product in roughly two and a half years remains one of the most successful ETF launches in the history of the industry, and no amount of 2026 redemption reverses that structural achievement.

But the trailing twelve-month trajectory is unambiguous. IBIT peaked as a flow magnet and the past six months have been distribution.

The other products are considerably smaller. Fidelity's FBTC and ARK's ARKB have accumulated cumulative inflows in the billions. Morgan Stanley's MSBT, launched in April 2026, sits near $400 million cumulative — the newest entrant and one of the few consistently positive contributors in recent sessions.

The Flow Windows Tell the Whole Story: +$154.5M Over Five Days, -$3.91B Over Three Months

IBIT's flow data across different measurement windows describes the entire 2026 experience with unusual clarity, and reading them together is more informative than any single figure.

Five-day net flows: positive $154.5 million. One-month: negative $1.83 billion. Three-month: negative $3.91 billion. Six-month: negative $3.07 billion. One-year: positive $5.14 billion. Since inception: positive $60.35 billion.

Read that sequence in order and the shape becomes obvious. The fund has been enormously successful over its life, still positive over twelve months, sharply negative over six and three months, negative over one month, and marginally positive over five days.

Two details in that series are worth isolating. The three-month figure of negative $3.91 billion is larger than the six-month figure of negative $3.07 billion, which means the first three months of the six-month window were net positive — the deterioration accelerated rather than being uniform. And the one-year figure remaining positive at $5.14 billion while the six-month is negative $3.07 billion means the second half of 2025 contributed roughly $8.2 billion of creations that 2026 has been steadily giving back.

The five-day positive of $154.5 million recovers roughly 8.4% of one month's outflows. That is the honest measure of the current repair.

The category-level picture matches. Three consecutive positive weeks recovered roughly $307 million against an $8.2 billion drawdown — 3.7% of the damage.

What these windows collectively establish is that the flow reversal is real in direction and immaterial in magnitude. Direction matters for what happens next; magnitude matters for what has already happened to price. Both readings are true simultaneously, and analysts describing the period as a repair observation window rather than a confirmed recovery have the framing right.

One institutional forecast cut its twelve-month inflow projection to zero earlier this year. On the current run rate, zero is roughly what the category is delivering.

How the Mechanism Actually Works and Where the 45% Figure Comes From

Understanding what the flow numbers measure — and what they do not — matters for anyone trading on them.

The daily figures published by the major trackers measure primary-market activity: specifically, the net change in shares outstanding for each ETF resulting from authorised participant creation and redemption transactions. They do not capture secondary-market trades between individual investors on exchanges.

When investors buy in, authorised participants purchase Bitcoin on the spot market and deliver it to the custodian in exchange for newly created ETF shares. When investors redeem at scale, the custodian sells the underlying Bitcoin on the spot market to return cash to the redeeming participant.

That is why sustained outflows create systematic selling pressure independent of any individual trader's price view. It is rule-based and price-insensitive.

Research cited across 2026 coverage estimates these flows now explain roughly 45% of weekly Bitcoin price moves. Separately, one bank's research found that each $100 million in net ETF inflows correlates with a roughly 53 basis point price move.

Both figures come with an important qualification that is frequently omitted. A senior industry figure clarified in February 2026 that ETF inflow numbers do not translate one-for-one into same-day spot Bitcoin purchases — there can be a lag, and authorised participants may use Bitcoin they already hold before going to market. The 45% correlation is a statistical aggregate, not a daily guarantee.

Over multi-day runs of inflows or outflows, the mechanism operates as described. Over any single session, the relationship is loose.

The practical implication for reading today's data: a $49.7 million outflow on July 28 tells you almost nothing about Wednesday's price. Four consecutive outflow days totalling a meaningful sum tells you considerably more, because that is the window over which the mechanism actually transmits.

Which is why the streak matters more than the number, and why the fifth session — determined largely by what the Fed says at 2:30 — is the one to watch.

Outflows Arrive in Bursts, Inflows Arrive in a Drip

The asymmetry in how capital enters and exits this category is the structural feature that has defined 2026, and it is worth quantifying precisely.

July 13 produced a $424.7 million net outflow — the largest single-day redemption in three months. One bad day removed 43% of what seven subsequent good days restored.

That is the shape of this market. Outflows arrive in concentrated bursts. Inflows arrive in a drip.

The seven-session positive run that preceded the current outflow streak illustrates the drip. It opened at $227 million and $203.1 million on consecutive days, then fell to $68.99 million by the third session — a two-thirds reduction in daily pace inside 48 hours. A streak that fades in magnitude while remaining positive in direction is a fundamentally different signal to one that sustains.

Compare that decay pattern against the May sequence, where individual sessions removed $733.5 million, $649 million and $630.4 million within three weeks without any deceleration.

The mechanical explanation is holder composition. Creations come from advisers making measured allocation decisions across many client accounts, which produces steady, moderate flows. Redemptions come from institutions hitting risk limits or rebalancing, which produces concentrated, large flows on specific dates.

With $15.2 billion of unrealised loss sitting in IBIT's holder base, the population capable of producing a concentrated redemption is large and the population capable of producing steady creation is comparatively small.

For anyone modelling the flow picture forward, that asymmetry is the base case rather than an anomaly. A category that takes two weeks to accumulate $300 million and one session to lose $425 million does not mean-revert toward its inflow history — it grinds until the underwater cohort has finished exiting.

Three positive weeks followed by four negative days is exactly what that grinding looks like.

The Complex Holds 6.03% of Bitcoin's Market Cap and That Is the Ceiling Nobody Discusses

The aggregate scale of the spot ETF complex relative to Bitcoin itself is the number that frames how much further this channel can realistically drive price.

Total net asset value across US spot Bitcoin ETFs stood at $78.817 billion as of July 23, representing a net asset ratio of 6.03% of Bitcoin's total market capitalisation. An earlier July measurement from a different provider put the figure closer to $74.37 billion, with the difference reflecting timing and methodology.

Roughly 6% of Bitcoin's market value now sits inside regulated US ETF wrappers, against 6.3% at a comparable point a year earlier when the category's total net asset value was materially higher in dollar terms.

That ratio has been remarkably stable through both the accumulation and distribution phases, which tells you something important: the ETF complex has been growing and shrinking roughly in line with Bitcoin's own market capitalisation rather than independently of it.

The implication for the bull case is uncomfortable. The argument that ETF adoption structurally re-rates Bitcoin requires the ratio to expand — for a growing share of Bitcoin's float to migrate into wrappers that mechanically remove it from circulation. That has not been happening in 2026. The ratio has been flat while the absolute figures have fallen.

The counterargument is that 6% represents an early stage of institutional penetration, and that the wirehouse and retirement channels have barely begun allocating. That is genuinely true — the products are still not available on every major platform, and model portfolio adoption remains limited.

The most successful ETF launch in industry history has captured 6% of its underlying asset in two and a half years. Whether that reads as impressive or as a ceiling depends entirely on the time horizon applied.

For the current trade, the relevant point is narrower: at $78.8 billion of assets against a category that has been shedding capital for three months, the ETF channel is currently a net drag on Bitcoin's price rather than a support.

Ethereum Took $103.9 Million While Bitcoin Bled — The Rotation Is Real

The most actionable signal in the current flow data is not about Bitcoin at all. It is about where the capital is going instead.

Ethereum ETFs added 37,959 ETH over the week through July 28 — roughly $71.17 million — while Bitcoin ETFs shed 3,170 BTC. Ethereum funds notched a third straight weekly inflow through July 24, pulling in $103.9 million, more than any other spot crypto ETF that week. On July 27 they flipped to a net inflow of $11.7 million while Bitcoin funds went the other way.

The issuer detail mirrors Bitcoin's concentration exactly. BlackRock's ETHA drove 37,424 ETH of the week's 37,959 ETH total — 98.6% of the category's inflows from a single product. Grayscale's Ethereum products added 5,515 ETH while Fidelity's FETH lost 4,980 ETH.

So the same asset manager whose Bitcoin fund led the category's outflows led the Ethereum category's inflows in the same week.

That is not institutional capital leaving crypto. It is institutional capital rotating within crypto, and the direction is out of Bitcoin and into Ethereum.

The plausible driver is yield. Regulatory clearance for staking inside Ethereum ETF structures means those products can hold a productive asset and distribute income, while a Bitcoin ETF can only track a price. In an environment where the risk-free rate sits at 3.50% to 3.75% and may be heading higher, the difference between a non-yielding wrapper and a yield-bearing one becomes material to an allocator's decision.

The split points to growing institutional appetite for Ether specifically rather than for digital assets generally.

For Bitcoin, that rotation is a structural headwind that operates independently of price. It means the marginal institutional crypto dollar has an alternative destination for the first time, and that alternative is currently winning.

Solana Took Inflows Every Session and XRP Sits at $1.5 Billion Cumulative

The rotation extends beyond Ethereum, and the pattern across the altcoin ETF complex reinforces the same conclusion.

Spot Solana ETFs recorded net inflows in every single US trading session in July 2026. Cumulative assets across the XRP ETF complex sit near $1.5 billion, with Solana products near $1.1 billion.

Set that against the broader picture: across Bitcoin, Ether, Solana and XRP funds combined, roughly $4.4 billion left over a recent thirteen-session stretch. Solana went positive every day inside that window.

The magnitudes are small — Solana's daily inflows have been measured in single-digit millions and XRP's in fractions of a million — but the consistency is the signal. Modest, persistent buying in the newer products while the largest and oldest one faces systematic redemption.

Altcoin ETFs showed mixed signals in the most recent sessions, with modest inflows into Solana products and outflows from newer Hyperliquid-linked funds. That mixed reading is what a maturing product landscape looks like: differentiation between assets rather than uniform risk-on or risk-off.

The interpretation that fits all of this data is that institutional crypto allocation has moved from a phase of expansion into a phase of selection. In 2024 and 2025, the question was whether to have crypto exposure at all, and Bitcoin was the answer by default. In 2026, allocators already have exposure and are choosing between assets, which means Bitcoin has to compete on merits it did not previously need.

Yield-bearing Ethereum, higher-beta Solana and regulatory-catalyst XRP all offer something Bitcoin does not.

That is the honest bear case on the ETF channel, and it does not depend on anyone turning negative on Bitcoin. It only requires them to prefer something else at the margin, which the flow data suggests they currently do.

Morgan Stanley Entered in April and the Newcomers Are Buying What BlackRock Is Selling

One counter-trend worth isolating: the newest entrants to this market have been consistent net buyers while the incumbent has been the seller.

Morgan Stanley launched its Bitcoin product on April 8, 2026, and MSBT has accumulated roughly $400 million in cumulative net inflows since. On July 23 — the session when the category lost $225 million and IBIT alone shed $202 million — MSBT recorded the day's largest single-day inflow.

Fidelity's FBTC and ARK 21Shares' ARKB have offset some of the damage repeatedly, adding 109 BTC and 77 BTC respectively during the week when IBIT lost 3,511 BTC. Grayscale's Bitcoin Mini Trust took in $5.1 million on July 28 against IBIT's $54.8 million exit.

The pattern is consistent enough to warrant explanation. Newer products have newer holder bases, which means lower average cost bases and no accumulated unrealised loss driving redemption behaviour. A fund that started buying Bitcoin in April 2026 has an entry price in the $60,000s rather than the $80,000s, which changes everything about how its holders behave.

That is the mechanical reason to expect the flow picture to improve over time regardless of price. As the underwater 2024-2025 cohort inside IBIT works through its positions and newer cohorts accumulate at lower levels, the category's aggregate cost basis falls and the redemption pressure diminishes.

It is also the reason to be careful about reading IBIT's flows as institutional sentiment. IBIT's holders are making decisions about a 23% loss. MSBT's holders are making decisions about a position closer to flat.

The wider product landscape now spans twelve or more US spot Bitcoin ETFs, including IBIT, GBTC, Grayscale's Mini Trust, FBTC, ARKB, BITB, HODL, BRRR, BTCO, EZBC, BTCW and MSBT.

That breadth is a structural positive that the current flow data obscures entirely.

What Today's 2:00 P.M. Decision Does to the Flow Picture

The Federal Reserve announces at 2:00 p.m. ET with the target range at 3.50%–3.75% and a press conference following at 2:30. Futures have priced roughly a one-third probability of a quarter-point increase, with September carrying odds near 80%.

The chair goes into this meeting with a June projection round showing nine officials leaning toward a hike, and with no forward guidance mechanism in place to anchor expectations in either direction.

The transmission into ETF flows is straightforward and it operates with a lag of days rather than hours. Higher policy rates raise the return on Treasuries and increase the opportunity cost of holding a non-yielding asset inside a wrapper that charges a management fee. For an allocator sitting on a 23% loss in IBIT with a risk-free alternative paying close to 4%, a hawkish outcome makes the redemption decision easier.

A dovish or balanced hold does the reverse, and the July repair narrative — three positive weeks before four negative days — has room to resume.

The specific things to watch, in order: the dissent count in the statement at 2:00, the characterisation of energy-driven inflation in the press conference at 2:30, and then Thursday and Friday's flow prints, which will show whether authorised participants read the outcome as a reason to create or redeem.

The two macro tests that were flagged as determining whether July's repair window extended or collapsed were June CPI and this meeting. The first passed without derailing the recovery. The second is the larger of the two.

For a category that has recovered 3.3% of an $8.2 billion drawdown, with one fund carrying a $15.2 billion unrealised loss and a rotation into Ethereum visibly underway, this afternoon is the difference between a fifth consecutive outflow day and a fourth-week inflow streak.

Neither outcome resolves the structural picture. Both determine August.

That's TradingNEWS