Skip to content

Invest1 publisher3 min readPublished

Two trackers read the same day of bitcoin ETF flows about twenty times apart

SoSoValue counted $13.29m leaving bitcoin funds on September 11 and $216m arriving in ether funds. Another tracker counted $267m out of bitcoin and $46m out of ether on the same date.

The Investor · Invest desk

Illustration accompanying Two trackers read the same day of bitcoin ETF flows about twenty times apart

What happened

  • SoSoValue's tracker recorded $13.29 million of net outflows from bitcoin ETFs on September 11, a figure Crypto Briefing described as the bleeding slowing after a rough week.
  • Ethereum ETFs absorbed $216 million of fresh capital the same day, which Crypto Briefing read as institutional money rotating sharply out of bitcoin products and into ether.
  • Bitcoin ETFs had already shed roughly $450 million over the three days before the 11th, and had taken in $730.9 million in a single day on September 3.
  • Other trackers put the same date's bitcoin outflows at 3,391 BTC, valued at roughly $267 million, alongside ethereum outflows of 17,723 ETH worth about $46 million.

Compiled by The InvestorSomething wrong?How this is made

Why it matters

  • contradiction One data set makes September 11 a switch between two crypto assets and the other makes it selling on both sides, so the same day supports opposite allocation stories.
  • constraint A daily flow print cannot carry an allocation conclusion when two providers disagree by more than the inflow that is supposed to be the evidence.
  • decision An allocator sizing a bitcoin-to-ether switch off these numbers has to pick a data provider first, because NAV timing and cut-off windows decide the sign as well as the size.

One tracker put September 11's bitcoin ETF flows at $13.29m of net outflows [1]. Another counted 3,391 BTC leaving, which it valued at roughly $267m [5]. The two readings sit about twenty times apart [1]. The dollar difference between them is $253.7m, larger than the $216m of ethereum inflows the rotation story rests on [2][2].

On the ethereum side the two data sets differ in sign, not just in size: $216m in on one, 17,723 ETH out on the other at about $46m, a spread of $262m [2][6][8]. Crypto Briefing set the test for its own headline when it wrote that "If institutions were simply de-risking, you would expect outflows across both products." [12] The alternative figures cited in the same article are outflows across both products [5][6]. The publisher attributed the gap between providers to the timing of NAV calculations and data cut-off windows, which it said can produce materially different headlines from the same underlying market activity [7].

The coin-denominated numbers carry their own problem. Spread $267m across 3,391 BTC and the implied price is about $78,700 a coin [3]; spread $46m across 17,723 ETH and the implied price is about $2,595 [4]. Crypto Briefing described those conversions as being at prevailing prices and approximate [5][6], and the article does not say which price source or timestamp produced them.

Scale matters here more than direction. Against total assets of $97bn to $99bn across the major bitcoin ETF issuers in mid-September, $13.29m of outflows is about 0.014% [9][5]. The three days before the 11th averaged $150m a day of outflows, so the 11th ran under a tenth of that pace [3][6]. Crypto Briefing wrote that the reading "looks less like a recovery and more like the bleeding slowing to a trickle after a rough week" [11], which is a different claim from a rotation.

There is a value point buried in the asset figure. Bitcoin ETFs have taken in more than $55bn of cumulative net inflows since their US launch in January 2024 [8], so of the $97bn to $99bn now sitting in them, $42bn to $44bn is not cumulative net inflow [7]. BlackRock, Fidelity, Grayscale and ARK 21Shares hold the bulk of that stack, with BlackRock's iShares Bitcoin Trust consistently leading on volume [10].

In my view the September 11 prints do not establish that institutions moved money out of bitcoin products and into ether ones, which is what Crypto Briefing said happened [13]. Two readings of the same day disagree by more than the inflow that would be the evidence [2]. There are ways this comes out the other way. If issuers confirm creations matching the $216m, SoSoValue's version stands and the coin-denominated trackers are measuring a different window [7]. If ether inflows persist for a week or two against continued bitcoin outflows on one provider's methodology, the one-day dispute stops mattering. And the whipsaw itself is documented either way: $730.9m into bitcoin funds on September 3, roughly $450m out over the following days [4][3].

What to watch

  • Whether ether ETF inflows persist for a full week on one provider's methodology while bitcoin ETF flows stay negative.
  • Whether issuers' reported creations and redemptions for September 11 match SoSoValue's $13.29m and $216m figures or the coin-denominated counts.
  • Whether bitcoin ETF assets under management hold near $97bn to $99bn as the cumulative net inflow line stays above $55bn.
Loading claim ledger
Loading source directory links
Loading share composer
Loading topic controls
Loading related stories