Fed Hike Day: Don't Trade the Hike, Trade the Dot Plot

A 25bp hike is 93% priced. The market-moving risk is Warsh ratifying 3–4 more hikes while oil sits above $100 — and today's energy washout is the entry, not the exit.

1. The hike is priced — the dot plot is the event

The decision itself is theater at this point: CME FedWatch puts a 25bp hike to 3.75%–4% at 93% odds, up from 36% a month ago, and Rosenberg expects no dissents. That repricing was driven by Warsh's Jackson Hole warning, a warm August CPI, and oil back above $100 — Morgan Stanley flipped from zero hikes to two (September + December) on exactly that mix.

What isn't priced is the path: futures now imply three to five hikes by next September, centered on four (to 4.5%–4.75%), with 77% odds of a second hike before year-end. The June SEP had the funds rate ending 2026 at 3.8% — one hike — and 3.6% in 2027. If today's SEP and dot plot ratify the market's four-hike path, that is a hawkish surprise even with a fully expected hike. If Warsh soft-pedals 2027, beaten tech rips. CNBC's live blog and Schwab's open note both flag the projections as the real risk — correctly, in my opinion.

History says Fed days punish late positioning: the S&P has fallen all five decision days this year by 1.5% on average (Bespoke), and DataTrek shows the Nasdaq fell one month after five of the last six first hikes. HSBC's counter — normalization hikes (1997, 2016) saw stocks recover in 3–6 months — is the base case, but only if oil cooperates. That is the whole ballgame.

2. Oil's 3% dip solves nothing — the bypass is still broken

Today's pullback — WTI below $102 (−3.7%), Brent $105.35 (−3.1%) on a reported U.S. crude build and word the damaged Saudi East-West pipeline "restarts in days" — reads like relief. I read it as noise against a still-broken map, and in my opinion the market is over-discounting a headline.

The facts: the East-West pipeline (the main Hormuz bypass) remains offline after drone strikes; the Houthis now hold the Bab el-Mandeb island workaround that Saudi exports were rerouted through; Hormuz tanker traffic runs "well below prewar levels" with crossings in single digits; and U.S. reserves sit at 40-year lows and are still being drained. Brent spiked to ~$109–110 and WTI to $106 this week; European refiners had September Saudi cargoes canceled. A pipeline that "restarts in days" into a strait controlled by Houthis is not a supply fix. Fortune's breakdown of the feedback loop and CNBC's oil update carry the details.

Consequence: a Fed hike does not produce barrels. Tightening into a supply shock squeezes demand to meet constrained supply — textbook stagflation pressure. That is why Rosenberg's line ("never a fan of tightening into a supply shock") matters more than the hike odds. Retail sales were strong (+1.2% headline, +1.4% control group, restaurants/bars up a seventh straight month; Atlanta GDPNow 4.4%), so Warsh has cover to hike — but hiking with $105 oil and 5% 10Y risks the 1994 analog DataTrek flags (−9.9% Nasdaq six months after the first hike) rather than the 1999 melt-up (+50%).

3. 5% is the regime line — mortgages, deficits, and AI capex all reprice there

The 10Y touched 5.041% Tuesday, closed above 5% for the first time since 2007, and hovers ~4.96% this morning. It started the year at 4.15% and has added 100bp+ since the Iran war began in late February. The 30-year mortgage is 6.76% vs 6.15% in January. The MOVE bond-vol index sits near 84 vs 65 in June, with a −60% 30-day correlation to the S&P — wider discount-rate bands make every equity multiple fragile. CNN's 5% explainer frames the consumer pass-through well.

Three downstream repricings, each with a level:

• Equities: BofA now sees 7,400 year-end (~2% below Tuesday's 7,585 close) and only 7,800 in 12 months; HSBC holds 8,100 but expects an initial dip. Six down in seven sessions to six-week lows with weakening breadth is what a 5% world looks like — earnings must do the work, and they haven't yet.
• AI credit: BofA's survey has 53% calling semis the most crowded trade and 42% flagging AI capex as the likely systemic credit-event source. Sharma's FT warning — 5% "decisively breached" chokes hyperscaler bond + equity financing — is already visible: Axon fell ~10% on a $1B debt raise; debt-financed AI buildouts drove August hedge-fund positioning. NVDA bounced ~2% today to $216 on Huang/Meta pushing back on AI-slowdown calls, but that is sentiment, not funding costs.
• Fiscal: Capital Economics' point stands — with debt above 100% of GDP, rising yields feed deficit worry which feeds yields. Nominal GDP still outpaces debt-service growth, so no doom loop yet — but every week above 5% narrows that margin.

4. The trade: buy the Fed-day energy washout, wait on everything else

Energy is the only S&P sector down hard today (−2.9%) while tech leads (+0.9%) into the decision — a classic pre-FOMC rotation that inverts if the dot plot is hawkish. Diamondback at −9% on no company news is forced de-risking, not a thesis break. My view: fading a supply-driven oil dip into a demand-driven hike is backwards. The hike is the reason to own the barrel, not sell it.

Fits open book: XLE, BTC ($75,765 drifting toward $73.5k weekly kill — no adds pre-FOMC), ETH ($2,391 near $2,350 daily kill — tighten, no adds), QCOM/PLTR/CEG/AEP/VST/HPE/CRWD unchanged. No contradictions introduced.

5. Bull / base / bear into 2 p.m. ET — and what proves me wrong

Base 55%: hawkish hike. 25bp + dots ratify ~2 hikes this year and higher 2027. 10Y retests 5.04%, S&P chops toward 7,400–7,490, energy outperforms, tech fades. Action: fill XLE adds, hold T-bills.

Bull 25%: dovish hike. 25bp but Warsh stresses supply-shock limits, dots show one-and-done. 10Y drops through 4.85%, Nasdaq +2–3%, beaten semis (INTC/SKHY on the Ohio memory-talk, NVDA) squeeze. Action: trim into strength, do not chase — oil still above $100 caps the multiple.

Bear 20%: hawkish shock. 25bp + 4-hike dots + hot SEP inflation (PCE well above June's 3.6%/3.3%), or a split vote that reads as loss of control. 10Y >5.10%, S&P breaks 7,400 toward Goldman's severe-shock 5,400 tail, credit spreads gap. Action: stop adding anything but T-bills; XLE kill at Brent $88.

What would prove me wrong: (1) Brent daily close below $88 with Hormuz crossings normalizing and East-West verifiably flowing — supply fixed, energy thesis dead; (2) dots show only one hike through 2027 AND 10Y closes below 4.50% — duration regime flips bullish; (3) August CPI/PCE revisions erase the inflation overshoot — hike rationale collapses and today's call inverts. Any one triggers a written update.

What to do

• Before 2 p.m.: nothing pro-cyclical. Stage XLE bids $63–66; keep duration in bills.
• After the statement: read dots first, statement second, Warsh third. Two or more 2026 hikes = stay defensive; one-and-done = fade the tech spike, don't join it.
• Into week-end: watch Brent $88, 10Y 5.04%, S&P 7,490, BOJ Sept 18. Retail strength (+1.2%) says the consumer can fund one hike — not four with $105 oil.

Appendix

A. Data snapshot · B. Model & load-bearing assumption · C. Sources

A. Data snapshot (Sept 16, 2026, intraday unless noted)

B. Model & load-bearing assumption

Framework: hike-into-supply-shock = higher real rates + intact oil bid = energy earnings up, broad multiples down. Load-bearing input: Brent stays above ~$95 while Hormuz/bypass flows stay impaired. If that input is wrong (verifiable pipeline + strait normalization and Brent <$88 daily), the conclusion flips: cut XLE, add duration, and upgrade beaten tech — because the inflation constraint dissolves and Warsh can pause.

C. Sources (linked, working)

• CNBC — Fed meeting live updates, Sept 16, 2026 (odds, Rosenberg, HSBC, DataTrek, LaVorgna)
• Charles Schwab — Fed rate decision awaited with hike odds above 90%, Sept 16, 2026
• Investopedia — Stock market today Sept 16, 2026 (retail sales, BofA 7,400, sector moves)
• CNN — 10-year hits 5%, Sept 14, 2026
• Fortune — Oil jolts yields past 5%, debt feedback loop, Sept 14, 2026
• CNBC — Oil falls on Saudi pipeline restart hopes, Sept 16, 2026