TL;DR: Farside’s preliminary figures show US Bitcoin funds retaining a $301.2 million weekly inflow despite two days of withdrawals. Ether funds were down $220.7 million for the week; Friday’s tables still contained missing entries.

US spot Bitcoin funds gave back much of their early-week buying by Friday, leaving a smaller positive balance in the latest available data. The reversal makes the week’s total more informative than either the strong opening sessions or the withdrawals that followed.

Farside Investors’ table, checked on October 24, showed net outflows of $137.9 million for Friday after $635.6 million on Thursday. Together, those two sessions removed $773.5 million. Monday through Wednesday had attracted $1.0747 billion, leaving a provisional five-session inflow of $301.2 million.

Ether’s Friday inflow leaves the week underwater

Friday’s Bitcoin withdrawals were concentrated in BlackRock’s IBIT and Fidelity’s FBTC, at $110.3 million and $31.8 million respectively. Morgan Stanley’s MSBT recorded a $4.2 million inflow.

The Ether fund table showed a different balance. Friday’s reported inflow of $4.4 million followed a $110.6 million withdrawal on Thursday. Adding the five daily totals gives a provisional weekly outflow of $220.7 million. The small final-session gain therefore did little to offset the earlier redemptions.

Both calculations need a qualification: Farside still displayed a dash for VanEck’s Friday entry in each table when checked. The published totals should be treated as incomplete snapshots. A dash has not been converted into a confirmed zero in the underlying data, and subsequent updates could change the weekly figures.

Flows describe fund demand, with limits

The comparison suggests demand through these listed vehicles was more resilient for Bitcoin than Ethereum over this particular week. It does not establish who was buying, how long they intend to hold, or what happened across every other route into the crypto market.

The SEC’s general explanation of ETF mechanics distinguishes trading existing shares from creating or redeeming fund shares. Investors can trade with one another during a session without every transaction producing a corresponding change in the fund’s assets. Turnover and net flows answer different questions.

That distinction also matters when interpreting blockchain transfers. A transfer involving a custodian is not, on its own, a complete account of investors’ subscriptions and redemptions. Fund-flow data, share counts and custody movements need to be read in their own context.

For the next trading week, the useful question is whether fresh subscriptions can rebuild the demand lost in the last two sessions. A positive weekly balance leaves that possibility open, but the late reversal gives little support to treating institutional demand as a steady, one-directional buyer.

Written by the News Desk and edited by Samuel Rae.

Bitcoin ETF flows track recent price moves more closely than subsequent ones, according to DWF Labs research published October 22. The findings challenge a familiar shortcut: treating a large daily inflow as a forecast.

TL;DR: DWF finds Bitcoin ETF flows track past returns more closely than future returns.

The study examined 694 trading days through October 16, using aggregate U.S. spot ETF flows from Farside and Coinbase BTC-USD prices. It reported a correlation of 0.54 with the preceding five-day return, compared with 0.03 for the following five days.

The distinction is between observing demand and forecasting returns. Two series moving together does not show which caused the movement. A relationship with earlier prices may be useful for interpreting investor behavior without supplying a reliable rule for what happens next.

An ETF Purchase Can Have A Hedge Attached

Interpreting the economic motive behind a flow requires more than its direction. CME Group’s explanation of crypto basis trading describes how an investor can buy a spot ETF and sell futures at the same time when futures trade at a premium.

The position aims to capture the difference between the two prices as they converge. Its combined exposure can therefore be substantially different from buying Bitcoin outright in anticipation of a higher price. Financing, trading costs and the terms of the hedge affect the outcome.

CME also notes that several large spot Bitcoin ETFs and its futures use related reference-rate benchmarks. That alignment helps institutions pair the two sides. An observed ETF purchase can consequently belong to a trade whose full risk position is invisible in the fund-flow total.

Creation Data Is Only Part Of The Picture

The plumbing changed during the study period. In July 2025, the SEC permitted in-kind creations and redemptions for crypto exchange-traded products by authorized participants. That allows qualifying transactions to use the underlying assets instead of requiring cash. It does not give every retail shareholder a direct redemption right.

These mechanics make fund flows useful evidence of changes inside an exchange-traded wrapper, while leaving questions about investor intent and transactions elsewhere. Holdings under fund custody reveal only one part of a participant’s total position.

A near-zero forward correlation should be read narrowly. It does not prove that ETF activity can never move Bitcoin’s price, or rule out effects within a trading day. Daily observations cannot, by themselves, settle the sequence of every purchase and price move inside that day.

The practical implication is a higher bar for predictive claims. A useful forecast needs evidence that it works on subsequent observations and survives costs and changing market conditions. A large flow number, even when accurately measured, is only the starting observation for that test.

Written by the News Desk and edited by Samuel Rae.