Institutional Signal · Published

Institutional Bitcoin Demand Eases to 42.4 as Breadth Sinks

Are institutions buying bitcoin right now?

Yes, but less emphatically than a week ago, and through a narrower door. bryptoflows' composite institutional bitcoin demand score reads +42.4 on a -100 to +100 scale as of the October 1 close, down from +49.7 last Friday. The model still classifies that as moderate accumulation, and it is computed on all eight components — the missing-components list is empty, weight coverage is complete, and no weights were renormalised — so it is directly comparable to the +49.7, +23.8, +28.9 and +55.2 readings of the previous four Fridays.

The 7.3-point decline is not a dollar collapse. The composite's 20-session net flow actually edged up, from $2,821.8M to $2,885.8M. What deteriorated was the shape of the buying: a six-session inflow streak broke, and only 3 of 12 US spot products bought on the session the model samples.

What pulled the institutional bitcoin demand score down 7.3 points

The composite normalises eight components to a -100..100 range, weights them, and sums the results. Here is the full breakdown in points of the final 42.4, as of the October 1 close, with last Friday's contribution in brackets:

Two components gave back almost everything: persistence shed 6.7 points and breadth 4.7, for a combined 11.4-point drag. Against that, concentration added 3.7 and price confirmation 1.5. The eight deltas reconcile to the headline 7.3-point move within rounding on last week's published one-decimal figures.

The model's agreement figure — which measures how much the eight components corroborate each other, not a confidence interval around the score — fell to 52.7 from 56.7. A score drifting down on falling agreement means its inputs are pulling against each other, which is what the component table shows.

Why flow breadth collapsed while concentration improved

This is the week's genuine oddity, and worth unpicking: the two components look like they should move together, and this week they moved opposite ways.

On October 1 the channel took in a net $102.7M: gross inflows of $217.2M against gross outflows of $114.5M. Three products were positive — BlackRock's IBIT at $195.6M, Grayscale's BTC at $14.6M and Morgan Stanley's MSBT at $7.0M. Six were negative: Fidelity's FBTC at -$60.7M, Grayscale's GBTC at -$31.4M, ARKB at -$7.7M, BITB at -$6.9M, BTCO at -$4.2M and HODL at -$3.6M. Three reported exactly zero.

Breadth counts only the positive share, so three of twelve gives 25%, and its normalisation of twice the positive percentage minus 100 puts it at -50.0 — its worst reading in a month. Concentration, by contrast, measures how evenly flow is distributed across products regardless of direction. Nine of twelve funds were active on October 1, against six on September 24, so the HHI fell from 0.678 to 0.394 and the component printed positive at +32.3.

Read those together and the improvement in concentration is not the good news it looks like. The HHI fell because selling spread out, not because buying broadened. A session where one fund buys and six sell is more evenly distributed than one where a single fund buys and nobody else trades, and the component cannot tell those apart. Treat a positive concentration print alongside a deeply negative breadth print as a warning, not a confirmation.

IBIT's dominance is intact underneath all of it. Its $195.6M was 90% of all gross inflow and roughly 1.9x the session's entire net total — the channel was negative without BlackRock. Since January 2024 IBIT has absorbed $65,573.9M against a channel net of $57,666.7M, which is 114% of the net and 57.8% of all gross flow, with GBTC at -$27,896.0M on the other side. Monday's flow piece worked through the same imbalance from the fade side.

Is institutional bitcoin demand fading, or just resting?

External commentary this week says it is not fading, and on a monthly view that is defensible. US spot bitcoin ETFs took in $2,647.7M in September by bryptoflows' own count — a figure The Block reported independently as $2.65 billion on October 2, quoting Zeus Research analyst Dominick John that "ETF inflows suggest that institutional demand has not faded".

The composite does not contradict that so much as look at a shorter clock. September was strong, and the bulk of it landed in one week: September 18 through 24 delivered $2,684.3M. The five sessions since have delivered $185.7M in total, an average of $37.1M a session against $536.9M a session for that earlier block. The five-day flow acceleration reading sits at -499.7.

So both statements are true at once. The month was strong; the last five sessions were not. A monthly total is backward-looking by construction and will keep reading well for weeks after the pace changes — which is precisely the gap a daily composite exists to close.

The $2.68B block that expires in three weeks

Here is the most useful forward-looking number in this article, and it is a correction to the comfortable picture last Friday's signal piece painted.

That piece reported that only 4.3% of the composite's support sat in sessions about to expire, and concluded the score was no longer vulnerable to a mechanical unwind. That was accurate for the window as it stood — but the vulnerability did not disappear, it moved.

The current window runs September 3 to October 1. Its support is extraordinarily concentrated in time: the newest ten sessions hold $2,870.0M of the $2,885.8M total, or 99.5%. The oldest ten hold $15.8M. Thirteen of the twenty sessions are net positive.

Breaking the roll-off into five-session blocks shows three distinct phases ahead:

That third phase, roughly three weeks out, is the real test. The near-term picture is mild and the middle stretch is mechanically helpful, which means the score may well hold or drift up over the next fortnight for reasons that have nothing to do with new demand. Reading that as strengthening institutional bitcoin buying would be a mistake. The honest framing is that this indicator has a known, datable cliff in front of it, and the pace needed to clear it is roughly fourteen times the current one.

Strategy added 1,665 BTC and treasury accumulation still fell

Strategy's September 28 filing disclosed an acquisition of 1,665 BTC at an average of $85,681, for $142.7M, lifting holdings to 847,666 BTC at an average cost of $75,437. That is a purchase, as were all acquisitions in the model's window — there have been no disposals, and the model reads direction on every filing.

Yet treasury accumulation fell from +3.0 points to +1.4. The reason is window composition, not selling: the four filings the component reads now hold 2,615 BTC, because August 31's 4,603 BTC purchase has rolled out and two of the four filings in the window disclosed no purchase at all. Treasury capacity also slipped fractionally, from a ratio of 0.621 to 0.603, purely because the position it measures grew with price while undrawn financing stayed at $44.09B. Thursday's treasury piece covers the stock-funded financing mix, and live figures sit on the Strategy treasury dashboard.

What would move the bitcoin demand indicator next week

In rough order of how much each could shift the score:

What this composite cannot see

All eight components were available this run and weight coverage was complete. The model never substitutes zero for a missing input — it omits the component and renormalises the rest — which is why the count matters: a score built on five components is a different object from one built on eight, and readers comparing Fridays should check that first.

Two data notes belong on the record. The archived daily-candle dataset currently reports as unavailable in the API's own source listing, while the live price feed is current; price confirmation ran off the live feed and its +11.60% window reproduces exactly against closes from September 12 to October 2. That window also runs through October 2 while all flow data ends at the October 1 close, so the composite mixes as-of dates by one session.

Three structural limits bear restating. The composite reads twelve US spot ETF products and one corporate balance sheet, so it measures one visible channel rather than institutional demand in general — direct custody, offshore vehicles and private mandates are invisible to it. Its two heaviest components are 20-session trailing windows, so it is built to lag and will confirm a turn well after the turn, in both directions. And as this week shows, two components can move opposite ways for the same underlying reason, so the component table matters more than the headline number.

The composite is a directional proxy for institutional bid strength. It is not a price forecast, there is no mechanical link between the score and future returns, and nothing here is investment advice. A reading of +42.4 says the measurable institutional channel absorbed materially more supply than it released over the last twenty sessions, on support that is heavily concentrated in one week of September — not that bitcoin will rise.

All flow, absorption, treasury and composite figures are read directly from bryptoflows' own API, current through the October 1 close and Strategy's September 28 filing. The September total and analyst quote come from The Block article linked above, retrieved October 2, 2026. Current readings stay live on the institutional signal, overview and ETF flow dashboards.