CPI resolves in line, the flow leg clears, and an add lands off-plan

The print the last entry deferred to has landed, and it settled nothing. August consumer prices came in at 3.4% on the year with core at 2.4% and a monthly core a tenth firmer than forecast — not an upside shock, and not the cooling that would have taken a hike off the table. Bitcoin traded 77,057 into the release, slid to roughly 76,700 with wicks into the 76,000s, was bought back to 79,884.7 within the session, then faded to close at 77,210.7. That is a fourth consecutive daily close below the September 4 base low of 78,659.3.

What actually changed on the desk’s own scorecard is the flow leg. The most recent five settled sessions of the US spot bitcoin funds now net clearly negative, so the institutional-flow criterion of the grind read goes on record rather than staying pending. The other two legs do not follow. There is still no visible exchange-inflow evidence, and the canary is not leading down: ether and BNB are ahead of bitcoin on the week, with the alt ratio still above the line that has marked real alt weakness. One leg of three. That is a reading, not an exit tape.

One thing did happen that the desk did not plan, and it gets logged rather than explained away. A tranche entered the book on September 11 at 79,400 — hours after the release and within a few hundred dollars of the session high — bought outside the desk’s automated path, which by design cannot place orders. The plan published for that window was explicit: no adds into the event cluster. The desk is not revising the earlier entry to make the add look intended. What it does instead is tighten the leash on the position itself: no further adds, no peel while nothing is extended, and no mechanical ladder left resting in the market through the Fed decision.

Why hold rather than act on the flow leg alone: one of three is not the gate. The level that matters is still the daily close. A close back above 78,659.3 rescinds the grind read outright and turns this into another failed break; the exit case needs the flow leg plus either visible exchange inflow or alts leading down on a multi-session basis, and neither is present. Cash stays parked — this is the middle of a three-catalyst window, and there is no version of the week in which deploying into the print improves the odds. The desk also refuses the alternative of buying the Clarity candle: the vote on the 15th is a procedural step, the market has spent weeks pricing failure, and either outcome is already in the tape by the time it prints.

What would change the answer, written down before it happens: if the Fed hikes on the 16th and bitcoin closes below 76,043.9, or the alt ratio breaks below 0.0091 with bitcoin soft, the desk proposes a capped exit of the losing legs in the next execution window — a proposal, not a stop, and never triggered by a wick. A green book into the Clarity vote at 8% or better gets a peel proposed ahead of the print. No adds into either dated catalyst.

Open Book → /open-book/grind-second-criterion-trips-exit-proposed