The market tested every line we drew. The rules held.

1. The scoreboard: four wires, four verdicts

We set these ex ante last week, with responses pre-committed. Here is what fired, what held, and what we did about it — graded the way we'd grade anyone else.

Trip-wire #1: The Warsh minutes — fired

The wire: minutes show a June hike was seriously argued, or the balance-sheet task force leans toward accelerated QT. The minutes' own words: “A few participants commented that… there was a case for raising the target range for the federal funds rate” — they backed the unanimous hold, but the hike argument is now on the record, alongside a committee split almost exactly in half on year-end rates and one sentence with no hedge on it at all: “The Committee will deliver price stability.” The easing bias wasn't dropped in June; the minutes show it was struck deliberately. The QT half of the wire did not fire — balance-sheet operations were untouched, and the task forces got members on Thursday (Rajan, Stein, and Dynan drew the balance-sheet mandate; Marc Andreessen, of all people, drew productivity), not mandates.¹ Verdict: fired on the hike leg. Response executed as pre-committed: defensive into Tuesday's CPI, no duration adds, no dip-buying the rate-sensitive longs.

Trip-wire #2: The long end — fired on the level, not the mechanism

The wire: a 10-year close above ~4.50% or the 30-year through 5.00% on weak auction demand. The level broke — the 10-year closed above 4.50% Tuesday through Friday (ending ~4.56–4.57%, its highest close since May 22), and the 30-year held above 5% from Tuesday on. But the mechanism we named never showed: all three auctions stopped through. The 10-year's 0.6bp stop-through was its biggest since September 2025, with 81.5% indirect takedown — third-highest on record; the 30-year stopped at 5.058%, the highest 30-year auction yield since 2007, and still came through the screws for the first time since March.⁴ What actually drove the break was Tuesday's oil-and-war repricing plus a hike-floating Fed — and here we grade our own wire-writing: we specified the wrong mechanism. The level was the information; demanding a particular villain was rule-lawyering in advance. We executed the response anyway — overall duration to below-benchmark — because every channel that took yields there (war oil, a Fed floating hikes, a Monetary Policy Report that says inflation “stepped up further”) is at least as hostile to duration as the term-premium channel we drew. And the larger concession, owed since June: Schwab was right about the long end, and we were not. Their below-benchmark duration call is now our stance, a month late.

Trip-wire #3: The $900 line — pierced intraday, held on the rule

Tuesday at the lows, Micron traded $891.66 — through the line. It closed at $938.38. Our rule, written weeks ago and restated every week since: a close below $900 and we are out in full. Closes, not ticks — that design was deliberate, precisely so that a panic wick in a crowded name couldn't take the position away from us. It grazed the line again Wednesday morning ($900.41 low) and held. Then Thursday Micron rallied 4.5% on its $3B U.S. supply-chain announcement and the SK hynix book coverage, and by Friday's close the stock had finished the week up — a $144 intraday round trip from $891.66 to a $1,035.50 Thursday high.³ Verdict: the rule held, and the rule was right. A tick-based stop would have sold the exact low of the week. We take no victory lap on the position — the reckoning note's third cut stands, and so does the line — but the process note matters: pre-committed rules exist for the mornings when $58B profit prints get sold.

Trip-wire #4: Monetization spreads — quiet, and expensive

No second hyperscaler moved to resell compute this week, and Nvidia's vendor-financing program added nothing new. But the market spent the week pricing in the first mover: Meta rose 14.5% — its best week since early 2024, roughly $219B of market cap — substantially on enthusiasm for the compute-and-models pivot we flagged as this wire's origin.⁵ That is the uncomfortable version of quiet: when the tape pays the first mover that richly, imitation gets more likely, not less. The wire carries, unchanged.

2. Memory: the $58 billion quarter nobody would buy — and the $26.5B everyone would

Walk the sequence, because it is the cleanest demonstration all year of a market arguing about the future while ignoring the present. Monday, memory bounced with names attached — UBS took its 3Q DRAM forecast to +32% from +17%, BofA called the crash a “healthy reset,” and consensus expected Samsung to print the biggest quarter in its history. Tuesday, Samsung delivered exactly that: ₩89.4 trillion of preliminary operating profit, roughly $58 billion — up more than eighteen-fold year over year, the largest quarterly operating profit ever reported by any company, past Nvidia's record. Samsung fell about 7%. Micron fell 4.7% to $938.38 after touching $891.66 — and one wire headline called it plainly: memory entered a bear market.²

That is the tell we marked our book to on Wednesday — the third cut in two weeks, $1,500 in June to ~$1,300 on July 2 to ~$1,100, each with its trigger narrated in the standalone reckoning note, where single-name numbers live. A multiple cut, not an earnings cut: when record numbers get sold, the market is repricing the cycle's durability, and pretending otherwise would be marketing, not research. The Street still hasn't blinked — not one downgrade all week, with the mean target sitting ~50% above the stock — which is its own warning: a de-rating without sell-side capitulation is a de-rating that isn't finished being argued.²

Then Friday delivered the verdict we'd circled for two weeks. SK hynix priced 177.9 million ADRs at $149 on Thursday night — $26.5 billion, the largest US share sale ever by a foreign company, past Alibaba's $25B from 2014 — off an order book roughly seven times oversubscribed. It opened at $170 and closed its first session at $168.01, up 12.8%. And here is the part that matters for the thesis: the proceeds are earmarked for new Korean fabs, EUV lithography scanners, and a first U.S. fab in Indiana — the market knowingly handed the memory industry $26.5B to build more supply, and bought the rest of the sector anyway. SanDisk finished the week +9.8% (Goldman doubled its target), Western Digital +8.1%, the SOX +2.7%, and Micron — the name with the most to lose from a listed Korean competitor — closed the week green at $979.30.³

Two honest cautions before anyone declares the unwind over. First, the scarcity-premium erosion we flagged is now permanent — Micron is no longer the only US-listed way to own memory, and Monday begins SKHY's first regular-trading week, complete with a fresh set of 2× leveraged ETFs bolting Korean-retail-style machinery onto the US line. Second, the price signal and the stock signal still disagree: TrendForce's moderating-but-rising 3Q contract forecasts got a louder echo this week from module-maker ADATA, which sees Q3 DRAM up 20–30% — while the reckoning's core finding stands, that the marginal upside accrues to the uncapped names, not the one that sold its ceilings. That is why the crash-week buying went into the DRAM basket and ASML rather than more Micron — and ASML reports Wednesday.³

3. The minutes, the war, and $119B of duration

The first written record of the Warsh Fed reads like the man: short, unhedged, and hawkish at the margin. Beyond the June-hike sentence, the minutes show a committee split nearly in half on year-end rates — “many” seeing the funds rate at or below the current range, “many other” participants above it — with upside inflation risks judged “elevated” while employment risks “moderated a bit.” The drivers they named should sound familiar to readers of this series: tariffs, Hormuz supply disruptions, and — notably — AI-infrastructure demand itself as a source of upward pressure on technology prices and electricity. And one sentence stands alone, no qualifier attached: “The Committee will deliver price stability.” Thursday, Warsh named his five task forces — Rajan, Stein, and Dynan get the balance sheet; Mankiw and Sargent the inflation framework; Mervyn King communications; and Marc Andreessen, of all people, productivity — advisers, not mandates, for now.¹

Then there is the war. Tuesday, Iran attacked three commercial vessels in the Strait of Hormuz; the U.S. struck back at more than eighty targets and revoked Iran's oil-sanctions waiver. Wednesday, at the NATO summit in Ankara, the President declared the ceasefire over and called further negotiation a waste of time — though officials signaled talks may yet resume through Pakistani and Qatari mediation. WTI settled Wednesday at $73.52, up 4.4% on the day, and finished the week at $71.41, +4% — even with OPEC+ adding its fifth straight monthly quota increase.⁶ The June disinflation we've been riding — the oil crash that will flatter Tuesday's CPI print — is now running in reverse, and the Fed's own Monetary Policy Report, published Friday, pre-graded the situation: inflation “stepped up further this spring,” running roughly twice the target.¹

Against all of that, the $119B auction slate was the week's quiet stunner. Every leg stopped through: the 3-year by 0.6bp; the 10-year by 0.6bp — its biggest stop-through since last September, on the third-highest indirect takedown ever recorded; and the 30-year at 5.058%, the highest 30-year auction yield since 2007, still stopping through for the first time since March.⁴ Read those two facts together and the message is precise: there is no buyer's strike in Treasuries — there is a price. The world will fund the U.S. government enthusiastically at 5% for thirty years. That is not a crisis; it is a repricing — and it is exactly why we stopped arguing with it and took duration to below-benchmark. Futures put a July 29 hike near one-in-four and lean toward a first move by the fall meetings.⁴

4. The tape: a record Monday, a red Dow, and a very narrow week

The index scoreboard flatters the week. The S&P rose 1.2% to 7,575 — about half a percent from its record — and the Nasdaq 1.7%; but the Dow, which set its 21st record close of the year Monday (its first-ever finish above 53,000), gave all of it back and ended the week down, including a 577-point Wednesday on the minutes and the war. The Russell fell too, and the equal-weight S&P was roughly flat: strip out Meta's +14.5% ($219B) and Nvidia's +8% ($391B) and most of the market went sideways-to-down. Sector-wise it was energy (+3.3%, the war bid) and technology (+2.9%) over everything else — last week's rotation into defensives reversed as fast as it appeared.⁵ Under the AI hood, the split-within-the-split kept paying: Broadcom +11% on its new Apple deal — $30B for 15 billion US-made chips — while Intel fell 8.7% on 18A yield doubts and the indignity of AMD out-earning it in data center for the first time.⁵

Two more process notes. SpaceX joined the Nasdaq-100 Tuesday, and the inclusion week went exactly the way our initiation warned mechanical bids can: index trackers bought Monday's close, sellers used the liquidity, and SPCX fell 10.3% to an all-time closing low of $145.30 — the tape keeps walking toward our fair-value work, not away from it. And the week's three earnings reports (Levi's beat and raised; Pepsi beat on revenue; Delta beat on both lines and reaffirmed) all met the same grudging tape — every one of them closed down on its reaction day. A market that sells good news in staples, airlines, and denim while paying $219B for a compute pivot is a market with exactly one conviction.⁵ VIX at 15 says nobody is hedging that concentration.

5. Quantum check-in: the basket did its job

A second straight down week, and the sharpest one yet: IonQ −12.7%, D-Wave −10.8%, Rigetti −7.8%, Quantum Computing Inc. −4.3%, with Tuesday's risk-off taking the whole complex down 6–8% in a session on no company news at all. Here is the number we built the sleeve for: the QTUM basket fell 0.35% on the week. The diversified fund — ballasted by the incumbents we were told diluted the “pure” exposure — was flat while every pure-play fell mid-single to low-double digits; the pure-play-only fund, QTUP, lost 9.9%. That is not luck; it is the construction choice from our June 29 note doing exactly what it was designed to do.⁷

The week's filings tell the sector's story in miniature, again. Quantum Computing Inc. followed last week's authorized-share doubling with an S-8 registering 13.5 million more plan shares — the dilution machine, now in its second consecutive weekly filing. D-Wave's 8-K trumpeted a genuine honor (an IDC “Leaders” designation, with usage up 314% year over year) that changes no revenue line. Quantinuum popped 11.6% Monday on its post-IPO coverage wave — four banks, targets up to $100 — and then fell four straight days. And through all of it, not one downgrade anywhere in the group: a de-rating without sell-side capitulation, the same pathology we just watched in memory. Sleeve unchanged — venture-sized, basket-first, nothing added on momentum. Next dated catalyst: the QC-ADDS technical specs, ~September 20.⁷

6. The week ahead: CPI meets the banks, and the toolmakers report

Tuesday morning is the most concentrated macro morning of the summer. At 8:30 ET, June CPI — consensus sees the headline flat-to-negative on the month (roughly 3.8–3.9% year over year, down from May's 4.2%) with core near +0.2–0.3% — because June's basket contains the postwar oil crash. Ninety minutes later, Warsh delivers his first semiannual testimony to House Financial Services (Senate Banking follows Wednesday), grading that print live, under oath, with a hike on his own committee's table. And in between, JPMorgan, Citi, Wells Fargo, Goldman, and Bank of America all report Q2 before the open — five megabanks, one morning. The trap in the CPI number: it describes a world that ended Wednesday, July 8. The war premium in oil belongs to July's basket. A soft Tuesday print is a rear-view mirror, and we'd expect Warsh to say roughly that from the witness chair.⁸

Wednesday belongs to the toolmaker: ASML reports Q2 before the open — the first print of the newest position in the book, initiated into the memory crash on the logic that the lithography monopoly gets paid whichever way the memory-cycle argument resolves. The line we're reading is bookings, specifically memory/EUV orders — SK hynix just raised $26.5B with EUV scanners named in the use of proceeds, and ASML's order book is where that intention becomes a number. Thursday pre-market, TSMC — the cleanest read on AI capex before the hyperscalers themselves report — then Netflix after the close, with PPI, the Beige Book, retail sales, and Warsh's Senate round filling the gaps. Friday: housing starts (May's ran at a six-year low — the hike regime's most visible casualty), industrial production, Michigan sentiment, and monthly options expiration.⁸

Then the room goes quiet: FOMC blackout begins Saturday July 18 ahead of the July 28–29 meeting, making Waller (Monday), Cook (Thursday), and Jefferson (Thursday night) the last scheduled Fed voices. On the horizon that referees everything we own: Alphabet opens the hyperscaler capex prints July 22, Microsoft and Meta July 29, Amazon and Apple July 30 — and the policy calendar stacks up behind them (USTR's Section 301 decision due July 20; the 10% global tariff surcharge expires July 24 with no extension moving; Iran talks in whatever form they resume). Memory gets its own sub-plot Monday: SKHY begins regular-way trading, leveraged ETFs in tow.⁸

7. The book & trip-wires

Positioning into the week: duration formally at below-benchmark — the response Trip-wire #2 pre-committed, executed, with the Schwab concession attached. The Micron runner — the reduced residual position — stays behind the unmoved close-below-$900 line; the DRAM basket and ASML carry the memory thesis where the upside actually lives; the quantum sleeve is unchanged; and cash stays heavy by design — the late-July capex prints remain the deployment gate, and nothing this week changed that. Four wires for the week.

Sources & footnotes

1. FOMC minutes, June 16–17, 2026 meeting (released Wednesday July 8, 2:00 PM ET; federalreserve.gov, quoted verbatim): “A few participants commented that, in light of these developments, there was a case for raising the target range for the federal funds rate, but those participants indicated that they supported maintaining the current target range at this meeting”; vote unanimous 12–0 to hold at 3.50–3.75% (fourth straight hold). Year-end split: “Many participants” saw the appropriate rate within or slightly below the current range at year-end; “Many other participants” above it (June SEP: 9 of 18 submitters above, 8 unchanged, 1 below; Warsh submitted no dot). Conditional firming: in scenarios where inflation stays elevated, “almost all of these participants indicated that some policy firming would likely be warranted.” Unhedged sentence: “The Committee will deliver price stability.” Easing-bias language struck deliberately (“Most participants emphasized that they preferred not to repeat the language…”); majority favored a shorter statement. Balance sheet: operations unchanged, ample-reserves language intact — no QT-acceleration lean; the Chairman “described plans to establish five independent task forces.” Task-force members named Thursday July 9 (~4:47 PM ET): Communications — Fisher, Fraga, Mervyn King; Balance Sheet — Dynan, Rajan, Stein; Data — McMillon, Chetty, Murphy; Productivity & Jobs — Andreessen, C. Jones, Sharma; Inflation Framework — Mankiw, Sargent (rosters per CNBC/Bloomberg/Yahoo, minor variation across carries). Waller (Rome, Monday July 6): forward guidance “valuable” at times — a gentle counterpoint to Warsh's no-guidance doctrine. Fed Monetary Policy Report (Friday July 10): inflation “stepped up further this spring,” PCE running roughly twice the 2% target as of May; GDP ~2.1% pace in early 2026; housing “stagnant.” Fed pricing at week's end: July 28–29 hike odds roughly one-in-four (feeds ranged ~22–34% across CME FedWatch snapshots and aggregators); futures lean toward a first hike by the September/October meetings (CNBC's week-ahead cited ~61% for September). Sources: federalreserve.gov (minutes, MPR, Board calendar); Bloomberg, Reuters, AP, CNBC, Fortune minutes coverage; CME FedWatch as carried. ↩
2. Samsung Electronics preliminary Q2 2026 guidance (Tuesday July 7): operating profit ₩89.4T (~$58B at ~₩1,530/USD) on revenue ₩171T — a record for a third straight quarter, up ~18× year over year, above the ~₩87.3T consensus carried by Reuters/CNBC, and the largest quarterly operating profit on record, past Nvidia's ~$53.5B; Samsung shares fell ~7% on the print (capex/AI-durability worries), dragging the global complex (SOX −4.65% Tuesday). Micron fell 4.71% Tuesday to $938.38 (intraday low $891.66) and closed $948.80 Wednesday. Michael Burry's Micron short near $1,050 (“a destroyer of capital,” via Substack) was disclosed the same week. Street state through the week: no MU downgrades; mean target ~$1,486 (~50%+ above the stock) with no Sells — the “stale target” problem. Our third base-case revision ($1,500 → ~$1,300 → ~$1,100) and the full re-underwrite: the July 8 reckoning note. Sources: Samsung preliminary guidance coverage (Reuters, CNBC); verified closes; the reckoning note's sourced inputs. ↩
3. SK hynix Nasdaq debut: priced Thursday July 9 at $149/ADR × 177.9M ADRs = $26.5B — the largest-ever US share sale by a foreign company (above Alibaba's $25B, 2014), below the ~$29B ceiling in the original plan; order book ~7× oversubscribed (~$171.5B; Reuters: 500+ institutions placed orders; ~1,000 in the roadshow per Seoul coverage); cornerstone investors (Baillie Gifford, Coatue, Situational Awareness) indicated up to $7B and were allocated ~$5B. Friday July 10: opened $170 (+14.1%), closed $168.01 (+12.8%) under when-issued ticker SKHYV; regular-way trading as SKHY begins Monday July 13 (Nasdaq), with Direxion 2× long/short ETFs (SKUU/SKDD) launching alongside. Use of proceeds: Korean fab expansion (Yongin, Cheongju), EUV lithography scanners, and a first US fab in Indiana. Memory week (closes verified): MU $984.75 / $938.38 / $948.80 / $991.64 / $979.30 (weekly +0.38%; intraday range $891.66–$1,035.50; the $900 line pierced intraday Tuesday, grazed Wednesday at $900.41, never breached on a close); Thursday's +4.52% came on Micron's up-to-$3B US supply-chain investment (including $500M strategic financing for GlobalWafers America's Sherman, TX 300mm fab plus a 10-year wafer supply agreement) and the SKHY book coverage. SNDK +9.8% to $1,915.92 (Goldman target $2,200 from $1,200); WDC +8.1% to $582.59; SOX +2.70%. Monday's analyst wave: UBS 3Q DRAM contract forecast to +32% QoQ (from +17%); BofA reiterate Buy/$1,550; Citi 90-day upside watch. ADATA (July 8): sees Q3 DRAM +20–30%, NAND +35–40%; TrendForce (July 3, trade-press carries): 3Q26 DRAM ~+13–18%, NAND ~+10–15%, gains moderating. ASML week: +1.6% to $1,797.32 (initiation-day close $1,825.07; −4.3% Tuesday on Samsung-capex worries); Q2 report confirmed Wednesday July 15 pre-market. Sources: listing coverage (Bloomberg, Reuters, CNBC, Seoul press); Nasdaq Trader notice; verified exchange closes; company PR (Micron/GlobalWafers, July 9); analyst notes as reported. ↩
4. The $119B auction slate (all stop-throughs): $58B 3-year (Tue July 7) stopped 4.179%, 0.6bp through, bid-to-cover 2.60; $39B 10-year reopening (Wed July 8) stopped 4.580%, 0.6bp through — largest since September 2025 — bid-to-cover 2.593 (highest since Sept 2025), indirect takedown 81.5% (third-highest on record); $22B 30-year reopening (Thu July 9) stopped 5.058% — the highest 30-year auction yield since 2007 — 0.3bp through, the first 30-year stop-through since March, bid-to-cover 2.44. Daily CMT closes (H.15/Treasury; Friday varies ~1bp by feed): 2yr 4.13 / 4.19 / 4.21 / 4.16 / ~4.21–4.22; 10yr 4.48 / 4.55 / 4.56 / 4.54 / ~4.56–4.57 (highest close since May 22); 30yr 4.99 / 5.05 / 5.06–5.07 / 5.05–5.06 / ~5.06–5.07. The first closes above 4.50%/5.00% came Tuesday — before the 10- and 30-year auctions — on the war-oil repricing and pre-supply concession, not on auction failures; weekly changes ~+8bp across the curve. Trip-wire basis note: the published wire required weak auction demand; demand was strong — the level fired, the mechanism did not, and the response (below-benchmark duration) was executed on the level. Fed pricing per fn1. Sources: TreasuryDirect/FiscalData auction results; H.15; CNBC rates coverage; auction wrap coverage. ↩
5. Equity tape, July 6–10 (closes verified): S&P 500 7,575.39 (+1.23% wk; ~0.45% below the 7,609.78 record close of June 2); Nasdaq 26,281.61 (+1.74%; ~3% below its record); Dow 52,637.01 (−0.50% wk — the only major average down — after a record 53,055.91 Monday, its 21st record close of 2026 and first above 53,000, then −576.76 Wednesday on the minutes and Iran); Russell 2000 2,977.81 (−0.61%); S&P equal-weight roughly flat (~−0.2%, per weekly recaps). Sector leaders: Energy +3.31%, Technology +2.93%; Materials, Health Care, Staples lower. Meta +14.5% (~$219B added; best week since early 2024) on compute-pivot enthusiasm plus Muse model releases; Nvidia +8% to $210.96 (~$391B); Broadcom +11% to $399.97 — Apple–Broadcom multiyear deal announced Wednesday: $30B+ for 15B+ US-made chips, $1.5B Fort Collins fab expansion; Intel −8.7% (18A yield doubts; AMD's $5.8B Q1 data-center revenue passed Intel's $5.1B for the first time). SpaceX (SPCX): Nasdaq-100 inclusion effective Tuesday July 7; index trackers bought Monday's close (188.8M shares traded); the stock fell 10.3% on the week to an all-time closing low of $145.30 — see our initiation. VIX 15.03 Friday (weekly high 16.90 Wednesday). Earnings micro-reads: Levi's (Wed AMC) beat and raised (adj EPS $0.28 vs $0.24; revenue $1.56B), guide-midpoint nit sent it −2.2% Thursday, +2.0% Friday; PepsiCo (Thu) revenue beat/EPS 1¢ miss, −3.8%; Delta (Fri) beat both lines ($1.56 vs $1.48 LSEG; $17.67B) and reaffirmed FY, still −1.8% on fuel-cost worries. Sources: verified index/stock closes; CNBC/AP/Reuters wraps; company releases. ↩
6. War, oil, and the macro week: Tuesday July 7 — Iranian attacks on three commercial vessels in the Strait of Hormuz; U.S. strikes on 80+ Iranian targets; Iranian oil-sanctions waiver revoked. Wednesday July 8 — at the NATO summit (Ankara), the President declared the ceasefire over and further negotiation a “waste of time”; officials signaled talks may resume via Pakistani/Qatari mediation (as reported). WTI: ~$68.60 Monday → $73.52 Wednesday settle (+4.4%; Brent $78.02) → $71.41 Friday; ~+4.0% on the week. OPEC+ (Sunday July 5): fifth consecutive monthly quota increase, +188K bpd for August (seven-country group; UAE exited the alliance May 1). Gold: roughly −1.2–1.8% on the week to the ~$4,100–4,130 area (Friday quotes conflict across feeds); DXY ~flat near 101. Week's data: ISM Services 54.0 (prices 67.7 — still hot; employment 51.2, first expansion in four months); May trade deficit $77.6B, widest in over a year, on record $128.0B capital-goods imports (the AI buildout in the import data); initial claims 215K (below consensus); May consumer credit ~flat vs +$17.1B expected; wholesale inventories +0.3%. Policy horizon: the 10% Section 122 surcharge expires July 24 with no extension moving in Congress; USTR's Section 301 replacement decision (12.5% duties proposed on 46 countries) due July 20. Sources: CNBC/Reuters/AP war and oil coverage; EIA; OPEC; ISM; BEA/Census; DOL; Fed G.19; trade-law alerts. ↩
7. Quantum week (closes verified): IONQ $42.86 (−12.7% wk; ~$16.0B market cap; ~−25% trailing month), RGTI $16.54 (−7.8%), QBTS $20.09 (−10.8%), QUBT $8.66 (−4.3%); QTUM $154.46 (−0.35%) vs QTUP (pure-play-only) $18.45 (−9.9%) — the basket-construction contrast. Tuesday July 7: complex-wide −6.7% to −7.9% risk-off, no company-specific catalyst. Quantinuum (QNT) $70.45 (−5.5% wk): +11.6% Monday on the post-quiet-period coverage wave (Needham Buy $100; BofA Buy; JPMorgan OW $97; Morgan Stanley EW $78), then four straight down days; Capital Group disclosed 13.9% via 13G (July 6). Filings: QUBT S-8 (July 8/9) registering 13,544,946 additional plan shares (~$118.5M) — second consecutive week of dilution machinery after the 260M→460M authorization; D-Wave 8-K (July 7): named an IDC MarketScape quantum “Leader” (Advantage2 usage +314% y/y); IonQ and Rigetti filed nothing (EDGAR, clean week). No downgrades or target cuts anywhere in the group — de-rating without capitulation. Next dated catalyst: QC-ADDS 90-day technical specifications, ~September 20. Sources: verified closes (cross-checked vs exchange data); SEC EDGAR; analyst notes as reported. ↩
8. Week of July 13–17: June CPI Tuesday July 14, 8:30 AM ET (consensus: headline ~−0.2% to 0.0% m/m, ~3.8–3.9% y/y from 4.2%; core ~+0.2–0.3% m/m, ~2.9% y/y — FactSet/BMO ranges); Q2 bank earnings Tuesday pre-market — JPMorgan (consensus ~$5.5–5.7 EPS on ~$49–51B revenue, per previews), Citi, Wells Fargo, Goldman (~$14.01 on ~$16.0B), Bank of America (~$1.12–1.13 on ~$30.7–30.8B) — Morgan Stanley Wednesday; Chair Warsh's debut semiannual testimony: House Financial Services Tuesday July 14, 10:00 AM ET; Senate Banking Wednesday July 15, 10:00 AM. Wednesday: PPI 8:30 (consensus ~0.0 to −0.2% m/m), Beige Book 2:00 PM, ASML Q2 pre-market (company guide €8.4–9.0B revenue), J&J pre-market, United Airlines after close; China Q2 GDP overnight (~Wed ET, date varies by source). Thursday: retail sales 8:30, claims, Philly Fed; TSMC Q2 pre-market (consensus ~$3.81/ADR on ~$40B, +33% y/y); UnitedHealth pre-market; Netflix ~4:01 PM ET. Friday: housing starts and permits (May starts 1.177M SAAR, a six-year low), industrial production, import prices, U. Michigan sentiment; monthly options expiration. Treasury: bills only (next coupons: 20-year July 22, 10-year TIPS July 23). Fed speakers pre-blackout (Board calendar): Waller Monday (NYABE), Cook Thursday 1:00 PM (Exchequer Club), Vice Chair Jefferson Thursday 7:00 PM (Stanford); FOMC blackout begins Saturday July 18 (meeting July 28–29, presser July 29 2:00 PM). Horizon: Alphabet earnings/capex July 22; Microsoft and Meta July 29; Amazon and Apple July 30; USTR Section 301 decision due July 20; Section 122 expiry July 24; SKHY regular-way trading from Monday July 13 (SKUU/SKDD 2× ETFs launching). Sources: BLS/Census/Fed schedules (primary where fetchable); company IR notices; consensus figures per named previews — treat as ranges. ↩

Nothing on this page is investment advice. We work in scenarios and trip-wires, not price targets — everything above is for thought and process, not for trading. Forward-looking statements are scenarios, not promises. See disclaimer.