The Fed raised rates into a hundred‑dollar oil shock, the Dow gave back six hundred and thirty‑one points — and the Nasdaq closed unchanged.

The 90‑Second Read

1. The decisionThe FOMC raised the target range 25bp to 3.75%–4.00%, unanimously, 12–0 — the first increase since 2023. The statement is short and closes on new language: “The Committee will deliver price stability.”
2. The dots did more than the decisionThe median 2026 fed funds projection moved to 4.1% from 3.8% in June, and 16 of 18 participants see at least one more hike this year. The 2027 median is also 4.1%, and the 2% inflation goal is not reached until 2029. Easing was not deferred; it was removed.
3. One tape, two marketsThe Dow lost 631.21 points, −1.21%, while the Nasdaq Composite finished −0.01%. Rate‑sensitives wore it — Goldman Sachs −3.96%, J.B. Hunt −13.30%. The AI and photonics complex went the other way: Lumentum +9.59%, Coherent +6.92%.
4. A hike into a supply shockCrude sits near $102 on the Iran war, with diesel at record highs and the Strategic Petroleum Reserve near its operational floor. Chair Kevin Warsh named commodity prices in the press conference. The decision and the energy shock are one story, not two.
5. Where the map leaves youThe December E‑mini is bid overnight and trades within four points of the futures translation of the call wall. The flip is 151 points overhead; the floor is 51.81 under yesterday’s close. Tomorrow is a quarterly expiry with roughly $6.2 trillion of notional rolling off.

What would change the read: a soft claims print beside a Philadelphia Fed number near the consensus collapse to 30.5 puts the “economy strong enough to take it” premise — the load‑bearing assumption under both bull cases in Act II — directly in play at 08:30. Reclaiming the flip in futures terms is what restores the dampened state; nothing below yesterday’s close does that.

The Scoreboard

Index rows are the Wednesday Sep 16 close; everything else is a live overnight level read between roughly 23:30 and 23:55 ET. The December E‑mini carries a large basis to cash — read the change column, not the raw level, against the index.

Sentiment & flow gauges

The flow read

The gauges and the tape disagree, and the disagreement is the information. Fear and Greed at 26 with breadth and momentum at extreme fear describes a market where the median stock has been losing for weeks, while the index barely moved because the cap‑weighted top absorbed the hike. Positioning is what matters into Friday: systematic and vol‑control books sit at the top of their historical equity allocation, estimates of forced selling on a sustained drawdown run an order of magnitude above buying capacity on a rally, the buyback bid is in its fifth day of blackout, and two straight weeks of equity‑fund outflows are on the board. A negative‑gamma book that loses its largest expiry at tomorrow’s open is a different book on Monday. The one gauge that has not cracked is credit — and Section 11 argues it is already cracking where index‑level data cannot see it.

Yesterday’s calls graded

Calendar & Scenario Map

Friday carries the quarterly expiry and the Bank of Japan, whose policy rate sits near 1.00%. Both are in Section 08.

Pivot Points & Gamma Map

Cannon Daily Levels · Pivots, Support & Resistance

Cannon Edge · Trend & 52-Week Range

Levels are from a public dealer‑gamma (GEX) model on the Sep 16 close. The corridor did not move for a second session while cash fell, so the whole change on the day is the index walking away from the roof. Per‑strike open interest on the October index chain agrees with the floor exactly and carries a second, heavier shelf between 7,350 and 7,400 — the stretch beneath the wall is not the empty air a single level implies.

VX curve and breadth

The volatility term structure is in contango end to end and steepened into the decision rather than inverting on it: October VX at 18.73 and November at 19.17 put the front future at a 5.7% premium to spot. September expired on the decision date itself, so October is the first contract carrying any of the aftermath — a curve that stays upward‑sloping through a first hike in three years is the options market declining to price a regime break. Breadth is the other half: roughly 31% of US names sit above their fifty‑day average, near the seventh percentile of the past year. On the pivot map above, the December E‑mini pivot is 7,632.25 with R1 at 7,689.50 — yesterday’s settle closed just beneath the pivot, and the overnight bid has carried price into the same band as the futures‑translated call wall.

Institutional Positioning

Desk Shift Tracker

Rows are ordered by this desk’s read of prominence on this particular morning, not by any score. DARK rows are named seats whose silence is itself the story.

Macro Pressure Map

The projections did the damage, not the decision. The 2026 median fed funds rate moved to 4.1% from 3.8% in June and 2027 was left at the same 4.1% — a flat path, not a pause before easing. 2028 comes down to 3.9% and 2029 to 3.6% against a longer‑run 3.2%, and only one participant now carries a longer‑run rate below 3%, against ten two years ago. Unemployment was marked at 4.1% every year through 2029, cut from 4.3%; headline PCE runs 3.7% this year to 2.3% next, core 3.4% then 2.5%, growth a comfortable 2.3%. A committee forecasting full employment, above‑target inflation and solid growth has said what it intends to do about it.

The curve took it in the front end and nowhere else. The two‑year rose about seven basis points into the close while the thirty‑year was unchanged to lower — a textbook bear flattener. The long end’s refusal to follow says either that the market believes the Fed will contain inflation or that containment will cost growth; both produce the same curve. Gilts sold roughly eight basis points and the Japanese long bond sits near a thirty‑year high. A 7.22% mortgage and record diesel are meanwhile doing tightening the funds rate cannot reach.

The energy shock is the whole macro forecast. Crude is up roughly 78% on the year on the Iran war and the disruption around Hormuz. The Cleveland Fed nowcast has September headline CPI near 3.50% against core near 2.39% — a gap that is almost entirely the energy impulse. Helima Croft supplied the supply‑side detail: the damaged Saudi pipeline may return partially within days but potentially six weeks in full, and can be struck again; the reason crude never reached 150 is that China cut imports sharply, and Chinese imports are rising again. She and Gundlach converge from opposite directions on refined products — diesel is a spare‑refining‑capacity problem, and incremental crude barrels do not fix it.

Growth is not the constraint yet — but check what is doing the lifting. The Atlanta Fed’s third‑quarter nowcast sits at 5.1% and updates this morning on housing starts; Gundlach’s warning is worth holding beside it, because the inventory line is what has been pushing it up, and inventory building is what happens when buyers expect prices to rise. Yardeni’s carry‑trade argument is what makes Friday’s Bank of Japan the meeting with asymmetric consequences for the US long end — and it lands on the same morning as the quarterly expiry.

Portfolio Positioning

Generac is the overnight story and it is an AI story. The stock closed the regular session unchanged and then added roughly a third of its value after hours on an agreement to supply Amazon Web Services with backup generation — reported at up to $8 billion over the life of the arrangement, an initial tranche near $2.4 billion across 2027–28, plus a warrant for AWS over 1.69 million shares. Three outlets published three different percentage moves on it last night; the verified print is in the Scoreboard. The read‑through is not Generac — it is that data‑centre capital expenditure has reached the point where the power equipment behind the rack is contracted years forward.

Fluence is the same trade from the other side. It fell nearly 16% after hours on a gutted fiscal‑2026 outlook — revenue cut to roughly $2.4 billion from a $3.0 billion midpoint, adjusted EBITDA taken to a loss near $200 million from roughly break‑even — on US battery supply‑chain bottlenecks. Grid‑scale storage and grid‑scale generation got opposite verdicts on the same night: a supply‑chain discrimination, not a demand one.

Housing and freight took the hike directly. Lennar missed on both lines — EPS $1.19 against $1.29, revenue $8.05 billion against $8.31 billion — with gross margin 15.8% versus 17.5%, orders down 9% and full‑year deliveries cut to 80–81 thousand. With the mortgage rate where it is, that is the rate channel working as advertised. J.B. Hunt’s guide‑down took C.H. Robinson, Old Dominion, Knight‑Swift and Landstar with it. Starts and permits print this morning into that setup.

The banks were the Dow’s problem and the reason is the curve. Goldman Sachs was the only large bank past 3%, on softer trading commentary layered onto the flattening. Tom Lee’s reading — reflexive rather than fundamental — is testable today: if the curve re‑steepens on a weak claims number the group should lead, and if it does not, the selling was about credit rather than margin.

Mega‑cap tech did not participate in the sell‑off at all. Nvidia, Apple, Meta, Broadcom and Tesla closed higher; Microsoft was the laggard at −1.37% even while raising its dividend 8% to 98 cents, with Morgan Stanley reiterating overweight at a $600 target. All eight of the largest names were bid after hours, and the optics complex — Coherent, Dell, Nebius, CoreWeave — traded as though the Fed had not met.

Fed Watch

The blackout ended with the decision and no Fed speakers are scheduled today — unusual on the morning after a live meeting; the first voice is Governor Bowman tomorrow at 09:30. October 28 prices as a coin flip: 48.7% hike, 51.3% hold, 0.0% ease. In the absence of officials, watch whether the ten‑year holds a five handle without help, whether the October probability moves on this morning’s data, and whether the White House pressure campaign draws a response — the President said publicly yesterday that rates should be “1% or lower” while stating he retains confidence in the Chair, a combination requiring roughly three times the largest single cut in the Fed’s history. The sentence that governs the projections is Warsh’s own: he was “hard‑pressed to describe broad financial conditions as restrictive… so we removed a dose of accommodation.” That is not how a committee describes finishing.