A Second Night of Strikes, Hormuz Transit Nearly Halted — Yet Oil Fell 2%: The Market Voted 'Another Blink' on a Real Escalation

This report is based on intraday data as of 12:26 PM ET and does not reflect closing prices. Markets may have moved since publication.

Reviewing Yesterday’s Call

Yesterday (7/8) we wrote that “the tail the market had priced at zero fired today” — the ceasefire was declared over, WTI jumped 8%, VIX rose 13.6%, and precious metals fell in the war-driven risk-off, which we read as Phase 1. Our three referees were: does tonight’s threatened strike actually land, does Iran’s retaliation draw blood, and does the far curve rise or fall? Today the answer is clear: the strike did land — a harder second night (~90 targets, larger than the first) — and Hormuz transit “almost ground to a halt” (~14 ships). And yet the market’s vote ran in exactly the opposite direction from the escalation: WTI fell 2.1%, VIX collapsed back to 16, and semis ripped across the board. The escalation was real; the market re-filed it as “another blink.”

Today’s Core Judgment

The escalation is objectively harder — a second night of ~90 targets, Hormuz transit nearly halted, IRGC again striking Kuwait and Bahrain. But the market voted the other way: WTI -2.1%, VIX back to 16, semis surging. When escalation becomes predictable, symmetric, intercepted, and casualty-free, it gets re-priced as “the new normal of junk time” rather than “war re-ignition” — the market trades the second derivative (will tomorrow be worse than today?), not the first (how bad is today?). Silver’s +4.2% is just an oversold mean-reversion; with RSI still at 30 and DXY RSI still at 61, this is still Phase 1, not the Fed-capitulation Phase 2.

Macro & Geopolitical Deep Read

The most important lesson today is that a harder escalation plus a worse physical reality failed to buy a higher oil price. Overnight the U.S. launched a second night of strikes on ~90 targets (Bushehr, the port of Chabahar, Golestan transport arteries, facilities around Iranshahr airport), larger than the night before; Iranian state sources said Hormuz transit had “almost ground to a halt,” with Kpler showing ~14 merchant transits on 7/8, far below the ~34/day average that followed the interim deal; IRGC retaliated again before dawn, with Kuwait reporting the interception of 3 ballistic missiles, 1 cruise missile and 10 drones, and Bahrain reporting multiple interceptions. By every objective measure: harder than yesterday, fewer transits, broader retaliation. And yet WTI fell 2.1% and VIX collapsed to 16.06 (-5.0%). This is a live sample of our “market trades the second derivative” framework — when the second night of strikes is still “symmetric, intercepted, no casualties,” what the market reads is not an escalation spiral but “this is the new normal” (a market analyst quoted by ZeroHedge: “This is the new normal — an uneasy equilibrium”). The memory of seven prior blinks, plus a dark fleet that keeps physical flow from ever hitting zero, together re-filed this salvo as junk time.

But record this divergence honestly: oil (down) and physical transit (14 ships, nearly halted) pointed in opposite directions. The front month is trading “war fatigue / blink mode,” while the physical barrel count is tightening. The one indicator that reconciles the split is still the far curve: if the front month falls but Dec 2027/2028 Brent holds or rises toward $80+, the market is treating this acute disruption as transient while treating the structural premium as real; if the far curve is pushed back to $65 alongside, that is a genuine bet on a full return to pre-war. There is also a quiet counter-vote to oil’s “blink”: EASA issued a notice telling airlines to avoid Iranian/Iraqi airspace through 8/31 — aviation and insurance are treating this as a multi-week event measured in months, which does not square with an oil market that “moved on” in a single day. The physical and insurance clocks are more honest than the front month.

The FOMC minutes (6/16-17) landed, giving yesterday’s cliffhanger a textual verdict. The minutes show “a few” officials saw a case to hike in June but ultimately backed a hold; inflation was described as “well above” target with risks still to the upside; and if inflation stays elevated due to AI demand, the Middle East or tariffs, nearly all who discussed it saw a need for some tightening — explicitly naming “strong demand for AI infrastructure” as a persistent source of upward pressure on tech-product and electricity prices. But Warsh himself still declined to submit a personal dot and kept his statement as short as possible. The market’s read (FXStreet, 7/9): the minutes showed no further hawkish shift, so dollar momentum actually eased, with USD/JPY slipping from 162.66 to 162.32 and DXY -0.1%. This continues the throughline that “the market has priced a hike path Warsh never promised”: the minutes are an inflation-first hawkish hold, not the three-hike path BofA has penciled in. And by labeling AI infrastructure and electricity prices as a structural inflation source, the minutes are the same coin as today’s “semis and power surge while the long end stays pinned at 5%” — AI capex is both a growth engine and an inflation engine, and that is precisely the structural reason the long end cannot come down.

The data likewise supports the idea that the floor pressing on precious metals is still in place. Initial jobless claims came in at 215,000, below the Reuters survey’s 218,000 — the labor market remains resilient, which gives the Fed no excuse to cut, which keeps spreads from narrowing, which keeps the dollar/real-rate foundation intact. On the other side, June existing-home sales printed a 4.09M SAAR, down 2.4% MoM and below consensus — rate-sensitive demand is softening under high rates. Resilient jobs on one side, a cold housing market on the other — which is exactly why silver’s +4.2% today only counts as an oversold bounce: the trigger that drags PM from Phase 1 into Phase 2 is spread compression (2Y falling + DXY turning), not another war headline, and certainly not one day of dollar softness. Vehicle pain ≠ thesis break.

Rates Read

Treasuries were strikingly quiet through this geopolitical whipsaw: 30Y 5.06%, 10Y 4.55%, 2Y 4.19%, near-zero intraday change, RSIs nesting in the neutral-to-firm 60-67 zone. That “quiet,” set against oil’s violence and the semis surge, is itself information — the 30Y sitting calmly above 5.0% for a sixth straight day, with no drama, is “new normal confirmed”: the long end is being offset by three forces (the energy pulse, sticky services/tariff prices, and AI-driven electricity inflation), pinned along the 4.8-5.0% line. The real event is today’s 1:00 PM ET $22B 30-year reopening (5.000% coupon, pre-auction WI ~5.09%, about 7bp richer than last month, whose tail was just 1.2bp) — the result comes after our data snapshot and is a real-time supply stress test for fiscal dominance; yesterday’s $39B 10-year reopening, with a 2.59 bid-to-cover (above average), offered some reassurance first. Zoom out and the long end that is truly moving in one direction is in Tokyo: JGB 30Y at 3.998% (RSI 80.85) is pressing right against the 4% handle, with the 10Y at 2.856% (RSI 79.08) — the BOJ’s hawkish path lifting Japan’s super-long end in a straight line. The U.S. long end pinned at 5.0% and Japan’s reaching for 4.0% are two sides of the same fiscal-dominance coin.

Sector Spotlight

AI / Semis: leadership flips for the second time in 8 days = rotation confirmed, not an exodus. Today’s surge is in the compute/infrastructure leg — MU +7.11%, AMD +6.80%, ARM +10.78% (RSI still just 32.5), AVGO +3.21% (5D +11.29%), DELL +4.75% (5D +14.75%), SMH +3.5% — while the platform/efficiency leg and the megacaps lagged: PLTR -4.14%, GOOG -1.92%, MSFT -0.94%, NVDA -0.91%. This is the mirror image of the 7/1 rotation (when META/PLTR ripped and MU/TSM were slaughtered) — two reversals in 8 days = the “compute vs efficiency” fault line oscillating back and forth, an internal rotation rather than capital fleeing. The catalyst is concrete and solid: Reuters reported on 7/9 that Meta’s in-house AI chip, Iris, has finished ~6 weeks of bug testing and is slated for production in 2026-09, designed with AVGO and manufactured by TSMC; Meta plans to deploy 7 GW of compute in 2026, doubling in 2027 — another hard print of custom-silicon demand feeding straight into the AVGO/TSMC supply chain. Worth flagging separately: ARM rose nearly 11% while its RSI is still only 32.5 — the hardest-hit names bounce hardest, which itself shows how deep the oversold hole was, and gives the rebound a mean-reversion character.

Traditional Energy: giving back part of yesterday’s war premium. WTI -2.1% to $71.99 (RSI back to 38.66), Occidental -2.47%, ExxonMobil -2.14%, Chevron -0.95%. That oil fell back amid a second night of strikes and near-halted transit is the sector’s confirmation of the “blink vote” — even an objectively harder escalation failed to hold yesterday’s war bid. It contrasts with yesterday’s +8% “re-pricing”: moving on in a single day says the market still treats this as an event pulse, not a structural re-rating.

Precious Metals: an oversold bounce, but still Phase 1. Silver +4.2% (RSI 30.78, still deeply oversold), gold +1.5% (RSI 36.49), the gold/silver ratio -2.6% (RSI 70.44). Silver outperforming gold (a falling ratio) = this is a risk-on oversold bounce, not a fearful safe-haven bid. Layered on today’s mild dollar softness (DXY -0.1%), it makes a textbook mean-reversion — but with DXY RSI still at 61, the strong-dollar regime is not broken and the trigger (spread compression) has not been pulled. Yesterday’s -5.8% flush and today’s +4.2% bounce are two breaths of the same Phase 1 vehicle, not the launch of Phase 2.

Copper & Cyclicals: the other face of risk-on. Copper +3.0% (5D +2.9%) firmed with the return of risk appetite, backed by China large-caps’ +4.4% 5-day demand read. Copper, the silver bounce, the semis surge, and VIX collapsing — the day’s signals all line up as “yesterday’s risk-off fully retraced.”

Power / Utilities: AI electricity demand gets an endorsement from the minutes. NRG +3.78%, VST +2.75%, CEG +2.00% (RSI 38). Today’s FOMC minutes explicitly listed electricity prices as an AI-driven inflation source — a rare validation, at the level of monetary-policy text, of the “Merit Order + AI power scarcity” logic.

Uranium: BofA cuts Cameco’s target but keeps a Buy. BofA on 7/9 lowered its Cameco (CCJ) price target from $143 to $140 while maintaining a Buy, noting uranium spot is about 23% below its 2026 average forecast but still calling Cameco its top uranium pick; Sprott uranium spot +0.6%. Soft spot vs strong franchise fundamentals = a “show me” for the nuclear leg — watch whether spot can catch up to the build-out narrative.

What to Watch & Framework Ahead

Tonight / coming days: whether a third night of strikes materializes. Oil is already voting “blink” — no third night, or another symmetric-and-intercepted exchange = junk time confirmed, and the oil pulse may keep giving back; if Iran’s retaliation on Kuwait/Bahrain draws blood and triggers a tit-for-tat, the conflict jumps a level and only then does the far curve genuinely rise. This is the fastest tell for “regime change vs another blink.”

7/11 Islamabad talks: Trump called the MoU “over,” but Iran’s foreign ministry has not formally withdrawn. Whether this round proceeds as planned is the dividing line between “verbally void” and “materially broken.”

7/17 Iran oil license wind-down expiry: the general license revoked on 7/7 permits already-authorized transactions only through 7/17. After that, the compliant channel for the ~63 million barrels of in-transit Iranian crude narrows — a concrete date for a re-tightening of the supply side.

Far Brent curve (Dec 2027/2028): the only indicator that reconciles the “oil down vs only 14 transits” split. A rise toward $80+ = the market accepting structural, treating the acute disruption as transient; a push back toward $65 = a bet on a full return to pre-war. It leads both the front month and the single-day transit count.

30Y auction result (today, 1:00 PM ET) + next EIA report (~7/15, API after the close 7/14): the former’s bid-to-cover and tail = a real-time fiscal-dominance stress test; on the latter, watch whether the SPR draw rate slows (a sub-5M/week week = a rationing signal = the borrowed “future oil” coming due) and the direction of commercial inventories.

7/24 Section 122 global 10% tariff expiry (15 days out) + USTR Section 301 hearings (7/7-7/9, ongoing): renewal, escalation, or a switch to Section 301 are all possible, entering the window where the market will start to price it.

7/29 FOMC + Day-60 fee showdown (~mid-August, aligned with the 8/21 sanctions-waiver expiry): watch the former for the trajectory of July CPI/PCE and whether “inflation first” softens; the latter remains the sole unresolved binary variable, and the second-night escalation has pushed that tail one step further to the fore.

Late-July earnings season (a structural slow variable): the “efficiency vs compute” debate, plus open-source/Chinese models as “good-enough and several times cheaper” cost competitors, is the question the AI-capex leg propping up U.S. GDP growth must answer. Vistra Q2 is set for 8/7; Meta’s Iris chip production timeline is set for 2026-09 (a forward anchor for the AVGO / TSMC supply chain).

Risk Disclaimer

This article is public market commentary and personal research notes. It does not constitute investment advice.