The headlines treat a big ETF day like proof that “institutions are buying.” Sometimes creations line up with real demand. Sometimes they are just the product tape doing what products do. As Senior Trading and Operations Associate at Stillman Digital — and first hire on that trading desk — with 10+ years in crypto and digital asset markets and $30B+ personally traded across institutional digital-asset markets, my filter is blunt: is this durable size that settles, or is it a creation print wearing an institutional costume?
What the Farside print actually shows
Farside’s Bitcoin ETF Flow table (US$m), checked on September 23, 2026, lists a 999.0 total for September 21 and a 714.7 total for September 22. On the 21st the largest lines were IBIT at 381.4, ARKB at 289.1, and FBTC at 238.8, with smaller positives in BITB, MSBT, GBTC, and Grayscale’s BTC trust line. On the 22nd the largest lines were IBIT at 350.3 and FBTC at 257.4, with MSBT at 99.0. Those are dated table numbers. I am not inventing a fresher session, and the September 23 row on that snapshot is incomplete or zero — so I am not forecasting today.
Those figures measure net creations and redemptions across listed U.S. spot Bitcoin ETF products. They are useful. They are loud. They are still a product-flow print.
Product flow isn’t a desk fill
A creation print tells you shares were created against underlying Bitcoin exposure in the ETF complex. It does not by itself answer the questions an OTC digital-asset book cares about: who wants inventory, on what credit, with what settlement path, and whether that size is still there when the headline cools.
Those two can move in the same week. They are not the same claim.
I hear the market call every large ETF day “institutions arrived.” Sometimes institutions did use the wrapper. Sometimes the wrapper’s mechanics and the tape’s storytelling got ahead of settlement reality. If you cannot say which, you do not have a thesis. You have a screenshot of someone else’s table.
This is the same ownership problem I wrote about when liquidation prints cleared levels: mechanical flow can move price without proving durable demand that settles. See Liquidation prints aren’t a desk bid. Different surface. Same line — who owns the risk after the print.
What a desk waits for after a creation headline
After a near-$1B ETF day, the useful questions are boring on purpose:
- Is there still real size when the product-flow narrative fades?
- Does the tape clear without needing the next creation print to explain the bid?
- Are you looking at fund mechanics, or at demand that can settle into inventory?
I will not pretend those answers live in a single Farside total. I will say this: if your only evidence for “desks are buying” is that ETFs printed creations, you have not proven OTC absorption. You have proven the product complex took in net capital that day.
Concentration matters too. When a few tickers dominate the session — as IBIT, ARKB, and FBTC did on the 21st per Farside — you are often reading wrapper preference and issuer share, not a uniform institutional stampede across every venue desks use.
Continuity — the filter does not change with the product
Charts, liquidations, and ETF creations are weather. The filter is ownership of risk after the move.
I grade tools the same way I grade tape: analysis and briefing can help; anything that wants you to confuse a product print for a settled desk fill fails. When to lean into exposure and when to stand down stays human — same doctrine as open and close staying human on the desk. See Execute or vanity: how an OTC operator grades AI and Jev grades decisions. It still doesn’t get the button. — different surfaces, same ownership line. Note: “OTC” there is activity framing and prior post titles, not my job title.
Treat the $999M and $715M sessions as product-flow news. Ask the settlement question next.
Key takeaways
- Attribute the Sep 21 999.0 and Sep 22 714.7 U.S. spot Bitcoin ETF net inflow totals to Farside Investors — dated table, US$m.
- ETF inflow prints measure creations/redemptions; they are not automatically durable OTC desk demand.
- Product flow can print large without proving size that settles on credit and inventory terms a desk recognizes.
- After a creation headline, ask what remains when the product narrative cools.
- I am not claiming a personal fill, client, or Stillman desk anecdote on these sessions. The filter is ownership of risk after the print.