Invest1 distinct publisher2 min readUpdated
Spot bitcoin ETFs pulled $606 million on August 20, and IBIT kept $503 million of it. The month's record total is mostly one issuer, and the first fund liquidation shows why.
The Investor · Invest desk

Compiled by The InvestorSomething wrong?How this is made
The four sessions ending August 20 brought in about $1.61 billion [5], which is roughly 78% of everything US spot bitcoin ETFs have collected this month [1]. IBIT's $1.09 billion of that run [7] works out to about 53% of the $2.07 billion August total that decrypt.co, citing SoSoValue, calls the best month of 2026 [6][2]. The record belongs to one fund's week more than to the category.
Thursday was also the narrow one. After IBIT's $502.99 million [2], $103.30 million was left for the other eleven products [4]. Fidelity's FBTC, the next largest taker at $64.74 million [4], collected about 13 cents for every dollar IBIT did [5]. On Wednesday IBIT's share was 55% and none of the twelve funds finished negative [3]; a day later the total was larger and VanEck's HODL was losing money [4]. Earlier in August the same pattern held, with roughly 80% of a $853.5 million five-day haul arriving through IBIT [8].
The liquidation makes the arithmetic concrete. Hashdex's DEFI charged 0.25%, exactly what IBIT charges, on $14.7 million against IBIT's $47 billion [13], an asset base about 3,200 times smaller at the same price [8]. That fee on IBIT's book is worth roughly $117.5 million a year [7]. On DEFI's stated assets it would be about $36,750 [6], and the roughly $26,000 the fund actually threw off [13] implies average assets nearer $10 million over the year [6]. Same sticker, incompatible businesses.
Worth separating from all of this: the flow print did not cause the price. Decrypt traces the move to the Treasury's decision to at least double long-end bond buybacks, then a squeeze that liquidated $3 billion in shorts [17]. Bitcoin traded near $77,000 on Friday, up about 24% on the week and still some 38% below the October 2025 high above $126,000 [15]. The chart has not turned either: the death cross has held since November 16, 2025, and with the 50-day at $63,976 the price needs roughly $5,000 more for a golden cross [16].
Across assets the money did broaden. Ether funds took almost $221 million [10], XRP went from $2.35 million to $13 million and Solana from $2.10 million to $15 million [11], narrowing the bitcoin-to-XRP intake ratio from about 220 to one to about 47 to one [9]. Inside bitcoin, none of that breadth reached the small products.
Follow any of these and your For You feed starts watching them — no settings page required.
Ranked by verification strength, evidence, and original report placement.
BlackRock's IBIT accounted for $502.99 million of the August 20 total, about 83%, up from 55% the day before.
On August 20 Fidelity's FBTC took $64.74 million, Bitwise's BITB $26.4 million, Ark and 21Shares' ARKB $12.2 million, and VanEck's HODL shed $3.59 million.
The four-day inflow run was $297.56 million Monday, $189.30 million Tuesday, then Wednesday and Thursday, coming to roughly $1.61 billion.
Across the four sessions, IBIT alone accounted for about $1.09 billion of the $1.61 billion.
DEFI charged the same 0.25% as IBIT but held $14.7 million against IBIT's $47 billion, throwing off roughly $26,000 a year in fees.
US spot Bitcoin ETFs took in $606.29 million on August 20, their largest single-day haul since May 1 and a fourth straight day of inflows, according to SoSoValue.
Evidence-backed comparisons of source perspectives and observed adoption signals. Read the methodology
Which Builder, Operator, and Investor concerns the observed source mix emphasized—not a truth score.
Evidence, demonstrated adoption, hype gap, incentives, and confidence are assessed independently, each on its own current evidence. How these are measured.
Detailed but single-publisher and vendor-dependent
Every number in the cluster comes from one Decrypt article whose flow data is attributed to a single vendor, SoSoValue. The granularity is high (per-issuer daily flows, month-to-date totals, net assets, moving averages), but there is no second outlet, no issuer statement, and the Hashdex filings are invoked without being identified. Two internal inconsistencies further cap the score: the ~$26,000 fee figure does not reconcile with 0.25% of the stated $14.7 million, and the fund's payout date is left as 'August 24 or August 28'.
Real, disclosed dollar flows plus a concrete fund wind-down
Adoption here is measurable capital movement rather than intent: a $606.29 million single day, a $1.61 billion four-day run, $2.07 billion month-to-date, $90.16 billion of net assets equal to 6.18% of Bitcoin's market cap, and $53.40 billion cumulative inflows since January 2024. The counter-signal is equally concrete — the first US spot Bitcoin ETF liquidation, showing adoption is concentrated rather than broad across issuers.
Mildly overstated framing on a well-documented base
The core assertions hold up: the 83% figure, the four-day run and IBIT's share of the month are arithmetic from the reported flows, and the piece volunteers deflationary context (death cross intact, price 38% below the record, price 'ran ahead of the flows'). The overstatement is modest and structural — a 'best month of 2026' record declared with seven sessions unfinished, a one-day concentration spike generalized into a thesis, and rally causation assigned to Treasury buybacks and a short squeeze without independent corroboration.
Crypto trade outlet reporting vendor data on interested parties
The visible incentive structure is moderate and disclosed: a crypto-native publication covering flow data supplied by a single analytics vendor it names, about issuers (BlackRock, Fidelity, Bitwise, Ark/21Shares, VanEck, Hashdex) that benefit commercially from inflow narratives. No issuer is quoted and no compensation or sponsorship is indicated in the supplied material, so exposure is inferred only from the outlet's beat and its dependence on vendor numbers, not from any disclosed arrangement.
Concrete figures, unverified single channel
Confidence is limited by breadth rather than specificity. The flow, asset and price figures are precise, internally consistent for the most part, and attributed to a named vendor, which supports the central concentration finding. But one publisher, no primary filings, an incomplete month, and two acknowledged inconsistencies (fee arithmetic, payout date) mean the durable conclusion is directional — inflows are heavily concentrated in IBIT and sub-scale funds are failing — rather than precisely quantified.
invest
Bitcoin's $72,000 Break Was Mostly Forced Covering, Not Fresh Bids1 distinct publisher
invest
Moscow Exchange Puts Perpetuals Inside a Clearing House1 distinct publisher
invest
Hyperliquid and Pyth backers ask the SEC to swap the NBBO for an onchain reference price1 distinct publisher
build
The bot returned 4.33%. Doing nothing returned 127.77%. The useful part is why.1 distinct publisher
Distinct publishers with included, body-backed reporting in this cluster.
1 article · August 21, 2026