Bitcoin ETFs lost $430 million as Fidelity reversed course

A positive day for US spot Bitcoin ETFs gave way to a $430.2 million withdrawal. Fidelity’s fund accounted for most of the swing between sessions, while BlackRock’s fund recorded outflows on both days. The individual funds show how quickly one positive headline can obscure a mixed market.

US spot Bitcoin ETFs have gone from a $74.6 million net inflow to a $430.2 million net outflow in one session, with Fidelity’s fund responsible for most of the change.

Farside Investors’ fund-level figures show that Fidelity’s FBTC took in $236.7 million on 4 November and lost $199.9 million on 5 November. BlackRock’s IBIT recorded outflows of $201.2 million and $105 million on those dates.

The two funds illustrate why a daily total can be misleading on its own. The largest swing came from a fund that had just taken in money, while a fund that was already losing money saw its withdrawals shrink.

A positive day built on one large inflow

An earlier 30-day ETF flow comparison also showed BlackRock and Fidelity moving differently. The daily tables let readers see how that split developed in November.

The 4 November total was positive because inflows more than offset redemptions elsewhere. Fidelity alone received more than three times the net industry figure, while BlackRock recorded a substantial outflow.

Removing Fidelity from the calculation leaves negative $162.1 million for the rest of the group that day. That arithmetic does not describe a hypothetical market without Fidelity. It shows what the other funds collectively recorded in the same session.

The Wednesday headline hid two markets

The difference between a headline total and what the funds underneath it were doing was already visible on 4 November. Fidelity added $236.7 million, while BlackRock lost $201.2 million. If the two are combined, their net contribution was a positive $35.5 million, leaving $39.1 million of the group’s $74.6 million inflow attributable to the remaining funds.

Look at the same day without Fidelity and the remainder was negative $162.1 million. Look at it without BlackRock and the remainder was positive $275.8 million. Both calculations use the same underlying table, and both answers are arithmetically correct. Each is useful only if the reader knows which fund was removed and why.

The calculations expose the size of the offset. The two largest opposing moves totalled $437.9 million in absolute value, almost six times the $74.6 million net total. Absolute fund-level flows are not trading volume or a measure of gross orders inside those funds. The comparison simply shows how much activity can disappear when opposite daily figures are added together.

Other funds mattered even if they did not dominate the headlines. Bitwise added $15.2 million, Ark’s ARKB added $19.9 million and Invesco’s BTCO added $4 million. The remaining funds’ positive $39.1 million comes from those reported inflows. The day was mixed, with a large withdrawal at one fund being offset by substantial subscriptions at another and smaller gains elsewhere.

How much Fidelity explains

The change in the total was negative $504.8 million, calculated by subtracting the 4 November inflow of $74.6 million from the 5 November outflow of $430.2 million expressed as a negative number. Fidelity’s change was negative $436.6 million.

Dividing the size of Fidelity’s move by the total change gives about 86.5%. That is Fidelity’s contribution to the day-to-day swing. Its share of the second day’s net withdrawals was smaller, about 46.5%, calculated as $199.9 million divided by $430.2 million.

The funds behind the change

The selected rows are from Farside and the changes subtract 4 November from 5 November. The all-fund total also includes products omitted from the selected rows. BlackRock’s positive change means the outflow became smaller. It is possible for a fund to contribute a positive change while still recording a withdrawal.

Fidelity and the funds other than BlackRock had a combined negative swing of $601 million. BlackRock’s $96.2 million improvement partly offset that figure, leaving the $504.8 million industry swing. The calculation identifies an offset that is easy to miss when each fund’s share of the latest day’s outflow is presented on its own.

Franklin and VanEck together brought in $56.5 million on Thursday. If those two reported inflows are removed from the arithmetic, the rest of the table shows $486.7 million of net withdrawals. Those subscriptions matter to the scale of the day’s result even though neither fund changed its direction.

The table also helps put the Wednesday figure in context. Two opposing moves of more than $200 million each can leave a net reading below $100 million. A reporting series that includes the largest positive and negative contributors next to the headline total gives readers a better picture of how many funds actually shared in the recovery.

What the other funds did

On 5 November, BlackRock’s outflow declined by $96.2 million compared with the previous session. Ark’s ARKB moved from a $19.9 million inflow to an $18.9 million outflow, while Grayscale’s GBTC went from zero net flow to a $49.9 million withdrawal.

Franklin and VanEck recorded inflows of $32.9 million and $23.6 million respectively on the second day. Taking Fidelity out of the 5 November total leaves a $230.3 million net outflow. The reversal extended beyond one fund, even though one fund explains most of the change.

The Thursday reversal was wider than Fidelity

The two largest funds by absolute movement on Wednesday both reported withdrawals on Thursday. Their combined outflow was $304.9 million, equivalent to approximately 70.9% of the $430.2 million net total. The number uses the combined withdrawal divided by the industry net withdrawal, so inflows elsewhere reduce the denominator.

Ark and Grayscale also recorded withdrawals. Franklin and VanEck took in money. The net outflow among funds other than Fidelity and BlackRock was $125.3 million, compared with a positive $39.1 million one day earlier. The rest of the market had reversed alongside the two largest contributors.

That matters because the obvious counterfactual can lead to the wrong answer. Remove Fidelity’s Thursday withdrawal and the total is still negative $230.3 million. Remove both Fidelity and BlackRock and the total is still negative $125.3 million. Fidelity explains most of the swing, but the sign of Thursday’s result does not depend on retaining it in the calculation.

The distinction is especially useful when making a headline. A story about Fidelity’s reversal describes the change from Wednesday to Thursday. A story about where Thursday’s money came out needs to explain the other withdrawals too. A headline can lead with the same fund in both cases while making two different claims about what happened.

How many funds actually turned negative?

The daily total identifies the direction of money across all products, but it does not identify how many funds shared in the move. Counting reported non-zero entries gives another view. On 4 November, Fidelity, Bitwise, Ark and Invesco reported inflows, while BlackRock reported the only outflow. On 5 November, BlackRock, Fidelity, Ark and Grayscale’s GBTC showed outflows, while Franklin and VanEck showed inflows. The counts treat a reported zero as neither a subscription nor a withdrawal.

The number of funds with withdrawals therefore increased from one to four. More funds reported positive flows on Wednesday than on Thursday. That supports a broader change in direction than Fidelity’s figure alone suggests. It still leaves a distinction between a day on which several issuers lose money and a day on which the largest redemptions account for most of the total.

The positive funds on Thursday matter because the group’s outflow is net of their subscriptions. Franklin and VanEck added $56.5 million between them. The combined value of the four reported withdrawals was $486.7 million. Subtracting the two inflows leaves the headline $430.2 million. These figures describe fund-level net entries, so the sum of negative rows is not a measure of every redemption order placed across the funds.

Wednesday offers the opposite example. Adding the four positive entries gives $275.8 million. BlackRock’s $201.2 million withdrawal reduced the total to $74.6 million. Seen this way, Wednesday featured a more substantial set of inflows than the headline alone would suggest, even though those gains were concentrated at Fidelity and offset in large part by BlackRock.

How the sessions compare with the late October exit

The $430.2 million withdrawal was smaller than the $643 million exit reported for 30 October. The difference was $212.8 million using those rounded headline figures. A single-session comparison helps establish scale, while the pattern across sessions shows why Wednesday’s positive reading offered limited evidence of a recovery.

Fidelity and BlackRock combined for a net inflow of $35.5 million on Wednesday. On Thursday they combined for $304.9 million of net outflows. Subtracting those combined numbers from each session’s total leaves $39.1 million of net inflows and $125.3 million of net outflows respectively for the remaining funds. Both groups moved in the same direction between sessions.

What the whole week showed before Friday

Farside recorded outflows of $398.6 million on Monday and $115.6 million on Tuesday. Adding the four reported sessions through Thursday gives $869.8 million in net outflows.

The positive Wednesday therefore interrupted a negative run without reversing the cumulative result. Friday’s final figure would determine the completed weekly total.

The week looked negative before Friday traded

The first two November sessions produced a combined $514.2 million net withdrawal. Wednesday’s $74.6 million intake recovered around 14.5% of that amount, calculated by dividing the inflow by the preceding two sessions’ outflows. The running total remained negative by $439.6 million after Wednesday.

Thursday added another $430.2 million in net withdrawals, bringing the cumulative figure to negative $869.8 million. The first four sessions can therefore be described as a negative start interrupted by one positive day. They cannot be treated as a completed weekly result before the last session is available.

The prior comparison is useful on two timescales. The previous Friday saw $643.2 million leave the funds according to earlier coverage of the 30 October figures. Thursday’s withdrawal was $213 million smaller on that basis. The four November sessions together nevertheless exceeded the size of that single October exit.

A rolling comparison also depends on the sessions included. A five-session total ending on Thursday would still include the 30 October figure. A month-to-date total would exclude it. Both can be legitimate, but treating them as interchangeable would make the latest weakness appear to come from a different set of trades than it actually did.

What the four sessions say about BlackRock and Fidelity

The daily comparison can also hide what happened earlier in the week. BlackRock recorded negative figures of $405.3 million, $95 million, $201.2 million and $105 million across 2–5 November. Together they give $806.5 million in net withdrawals. Fidelity reported positive figures of $6.7 million, $6.8 million and $236.7 million before its $199.9 million withdrawal. Its four-session balance was still positive by $50.3 million.

That reverses the emphasis a reader might take from Thursday alone. Fidelity was the largest negative contributor that day and explained most of the change from the previous session. BlackRock was the principal source of cumulative withdrawals over the four sessions. Both observations can be true because one compares adjacent days and the other adds a defined period.

The two funds together contributed negative $756.2 million through Thursday. The rest of the tracked funds accounted for negative $113.6 million after subtracting their combined balance from the $869.8 million total. As with the two-day comparison, removing the leading contributors leaves a negative number, though one much smaller than the headline.

Using the same cut-off also avoids attributing a week’s movement to a figure that arrived after publication. The relevant comparison for a story dated 6 November is what the market knew after Thursday’s US close. If a Friday figure is added later, the article’s period and calculation should change together. The four-session total describes an incomplete trading week even though it captures all sessions discussed here.

Why the flows and the Bitcoin chart can diverge

The reporting on Bitcoin’s reaction after the US midterms puts the price recovery alongside persistent ETF selling. Those two series cover different transactions.

ETF figures describe net fund activity over the reported session. Bitcoin trades around the clock on venues outside those funds, where liquidity and positioning can change even when US stock exchanges are closed.

An hourly comparison would need to align the ETF session with spot-market prices and volumes. Derivatives open interest and funding rates could help establish whether leverage expanded or contracted. The daily fund numbers alone do not identify who traded Bitcoin at a particular moment.

What a fund flow number measures

A US spot Bitcoin ETF trades on an exchange. A holder can sell a share to another investor without causing the fund to redeem that share. Secondary trading can therefore be busy on a day when the fund-level flow number is small. A net redemption concerns shares moving out of the fund’s outstanding share count through the creation and redemption process.

Authorised participants operate at that boundary. They can create or redeem shares in large blocks under fund procedures, while buyers and sellers in the secondary market normally trade individual shares. Demand or selling pressure can feed through to fund flows, but the two records cover different transactions. An ETF ticker’s trading volume is not another name for its net subscriptions.

Redemption mechanics can also vary across products and periods. The authorised participant may deal in cash or deliver and receive assets under the procedures available to the fund. The timing of any corresponding spot execution cannot be reconstructed from a daily flow total alone. Establishing that timing would require information about the fund’s basket, execution practices and the market session.

A decline in a fund’s reported assets is another distinct number. The assets can fall because Bitcoin’s price fell, because the fund had redemptions or through a combination of both. Comparing daily assets without adjusting for price can therefore mislabel an investment loss as a withdrawal. Looking at fund flows isolates a different part of the change.

Why the fund table cannot identify who was selling

A fund-level flow series can identify which product experienced a net redemption. It generally cannot identify the final holder whose decision sat behind the flow. Some market participants trade ETFs for directional exposure. Others use them in strategies that also include derivatives or another fund.

A holder could sell one product and buy another on the same day. That would leave opposing entries in a fund table. A long ETF position could also be held alongside a futures short. Closing both legs could reduce ETF exposure and futures open interest without expressing a simple view that Bitcoin was about to fall.

Those are possible explanations rather than evidence of what happened on either date. Establishing the role of basis trades would require observing futures pricing and positioning alongside fund flows. A narrowing basis or a simultaneous fall in open interest could justify a closer look. It would still be necessary to distinguish position closure from a new short or a price move that changed notional values.

The strongest conclusion supported by the table is narrower. The direction of money across products changed, and the distribution of those changes was uneven. An account of why specific investors redeemed would require position-level disclosures, comments from the participants or other records that connect the fund activity to those decisions.

Aligning the clocks before drawing a price conclusion

The earlier analysis of Bitcoin’s leverage and liquidation risk tracked derivatives positioning alongside fund flows. The timing and units need to match for that comparison to work.

Bitcoin’s market day does not start when US stock markets open. A move during Asia or Europe can change the price at which US investors begin trading. A flow reported after the US close can be compared with that session, but a 24-hour price change taken at another time may include activity the ETF figure could not have caused.

The comparison should therefore begin with a defined time window. It would establish the Bitcoin price and spot volume at the start and end of US trading, then compare that window with activity outside the session. Separating those periods gives a better chance of identifying whether an ETF-heavy story matches the sequence in the underlying market.

Leverage has its own timing problem. A falling market can force positions to close, while some traders open new positions in response to the move. Open interest measures what remains outstanding and funding describes the relative cost of holding perpetual positions. Neither is a direct measure of every liquidation or all spot demand.

That is where the two-session flow analysis can lead without overstating its result. Wednesday’s positive total relied heavily on Fidelity while BlackRock lost money. Thursday’s outflows reached multiple issuers, with a reduced withdrawal at BlackRock and a much larger change at Fidelity. A later comparison with time-aligned spot and futures data would have to preserve those differences.

How to reconcile the table with issuer disclosures

A second way to check a reported flow is to compare the fund’s shares outstanding across consecutive reporting dates. A fall in share count can corroborate a redemption even if the fund’s dollar assets rise because Bitcoin appreciated. A rise in the share count can likewise sit alongside a decline in assets when the price of the holdings falls enough.

The dollar conversion also needs to use the fund’s own valuation convention. Multiplying a share-count change by an unrelated intraday quote can produce a different result from a published dollar flow. Even with the right valuation, a fund report and a tracking table can differ in timing if one incorporates a late update.

The two November dates provide a useful example of why the reconciliation should be conducted by issuer. Fidelity’s direction reversed, while BlackRock’s did not. A check based only on a pooled asset figure could miss that distinction. Looking first at reported shares and then at the conversion to dollars helps a reader separate the direction of the flow from the amount attached to it.

When a redemption might reach the spot market

An ETF investor can sell a share on an exchange during the day. The buyer could be another investor, a market maker or another participant who later assembles shares for redemption. The transaction between the initial seller and buyer occurs in the share market. A redemption is an additional step and can happen under a different set of instructions.

The SEC permitted in-kind creations and redemptions for crypto exchange-traded products in July 2025. In an in-kind redemption, a participant can receive the underlying asset rather than only cash, depending on the applicable fund procedures. That makes it particularly important to establish the mechanism before treating a dollar redemption figure as a dollar amount of Bitcoin sold at the same time.

Even where the process involves a spot trade, the net flow does not reveal which venue handled it or when execution took place. Some activity may be anticipated by dealers before a final flow appears in a tracker. Other trades can occur for unrelated reasons in the same window. The daily table establishes the direction and scale of subscriptions and withdrawals. The market impact question needs evidence at a finer interval.

The distinction also helps when interpreting a positive day. New ETF shares can be issued without producing a readily observable, simultaneous spike in exchange volume. A dealer may source assets through a route that an exchange-based chart does not capture. Investors can therefore follow both the share creation process and the wider spot market without expecting the two series to trace identical paths.

The price comparison has to use matching time zones

A Bitcoin price captured at midnight UTC can include trading after the US session ends. A rolling 24-hour change from the following morning includes another set of hours. Both can be accurate measures of the price while answering a different question from the one addressed by a US ETF flow dated 5 November.

A better comparison would begin with the US session’s start and end and use the same windows on both days. The reader could then compare the direction of Bitcoin during that interval with the recorded fund flows and with moves before the opening bell. If the larger price move occurred before the US session, that timing would make a simple explanation based only on the day’s ETF figure less persuasive.

Derivatives measures require similar care. A dollar value for open interest can fall when Bitcoin’s price falls even if the number of outstanding contracts changes little. An analysis would need to know whether the data is expressed in dollars, coins or contracts before treating it as evidence that positions were closed. A funding rate also applies to a specific venue and interval. Comparing unlike measurements can turn ordinary differences in reporting into a story about leverage.

The fund comparison itself is more direct. Fidelity and BlackRock recorded opposite flows on Wednesday, both recorded withdrawals on Thursday, and their cumulative positions over the four reported November sessions differed again. The reversal can be stated with confidence at the fund level without using the flow data to fill gaps in the price history.

What a second rebound would have to show

The previous rebound in Bitcoin fund demand also left open a question about spot demand. An ETF rebound and a deeper spot market require separate evidence.

Another positive ETF session would carry more information if inflows appeared across several issuers and persisted across multiple days. A single large subscription could still pull the total above zero even while other funds lose money.

The same distinction applies to outflows. A widespread withdrawal and a reversal concentrated in one fund can produce similar totals while describing different activity underneath.

What would count as a wider return of ETF demand?

A useful next comparison would record both the dollar total and the number of funds with positive entries. One large inflow can lift the aggregate while leaving most issuers flat or negative. A string of positive sessions across several products would describe a different pattern, even if the daily headline figures were smaller.

That approach also puts the next Fidelity figure in perspective. A renewed inflow at FBTC could improve the total. Whether demand has widened would still depend on the other funds, especially those that reversed or reported fresh withdrawals on Thursday. The table already supplies a benchmark: Wednesday’s $74.6 million total included four positive rows and one negative row, while Thursday’s $430.2 million withdrawal included two positive rows and four negative ones.

The two figures worth keeping separate

Fidelity accounted for about 86.5% of the change between sessions. It accounted for about 46.5% of the second session’s net withdrawal. The difference comes from the $236.7 million it took in before the reversal.

For these sessions, the sign convention is particularly important at BlackRock. Its figure improved by $96.2 million even though the fund still had a $105 million withdrawal. Describing the improvement as an inflow would overstate support for the fund. Conversely, treating the second day’s outflow as evidence that BlackRock drove the aggregate deterioration would reverse the direction of its contribution.

What to Watch

• Friday’s closing flows: The completed US session figures and their effect on the four-session net withdrawal total.
• Fund participation: The number of ETFs taking in or losing money alongside the largest individual contributions to the daily result.
• Spot trading: Bitcoin prices and spot volumes during the US equity session, compared with activity before the opening bell.
• Futures positioning: Changes in open interest and funding over the same time window, with the units and venues identified.

FAQs

Disclaimer: This article is for information and educational purposes only and does not constitute financial or investment advice. Figures are based on sources available at the time of writing and may change with new disclosures. Nothing here is a recommendation to buy, sell, or hold any security or asset. Always do your own research. Information is current as of 6 November 2026. Legal questions should be discussed with qualified counsel.