| Index | Close | Chg | %Chg | Note | | S&P 500 | 7,437.63 | +121.48 | +1.66% | Range 7,370.98–7,448.75; recovered 108% of Wednesday’s 112.63-pt loss | | Nasdaq Composite | 25,122.18 | +679.24 | +2.78% | Snapped a five-day losing streak; range 24,813.84–25,171.44 | | Nasdaq 100 | 28,106.35 | +914.04 | +3.36% | Bloomberg 4 p.m. mark +3.4%; exits the technical correction — now 8.6% below the 30,762.20 record vs. 11% Wednesday | | Dow Jones Industrials | 52,208.06 | +613.92 | +1.19% | Recovered 53% of Wednesday’s 1,153.18-pt loss; range 51,655.52–52,266.45 | | Russell 2000 | 2,946.10 | +39.79 | +1.37% | Lagged the S&P by 29 bp — the squeeze was in mega-cap tech, not in beta | | VIX | 17.09 | −3.57 | −17.28% | Investing.com 16:14:46 print 17.08 (−17.33%); intraday 17.00–20.08; largest one-day VIX decline of the episode | | PHLX Semiconductor (SOX) | 11,303.0 | +855.5 | +8.19% | Biggest one-day advance since April 2025 (Bloomberg); range 10,963.9–11,406.8; still 22.9% below the 52-wk high of 14,655.3 and −21% month-to-date, the worst month since 2008 | | UST 10Y (official par) | 4.68% | +1 bp | — | Bloomberg 4 p.m. real-time mark: −1 bp to 4.67% — reconciled in §6 | | UST 2Y (official par) | 4.23% | +1 bp | — | Bloomberg real-time: −4 bp to 4.24% | | UST 30Y (official par) | 5.21% | +1 bp | — | A new 19-year closing high — the long end did not join the equity relief | | WTI front month | $84.23 | — | −0.27% | TradingEconomics board; Bloomberg 4 p.m.: −0.9% to $83.67 — basis note in §10 | | Brent front month | $89.36 | — | −1.52% | Back below $90 as Hormuz transits picked up (Bloomberg) | | Gold (spot, TradingEconomics) | $4,105.47 | — | +0.97% | Bloomberg spot 4 p.m. $4,116.73 (+1.2%); Dec futures $4,160.60 (+1.55%) on the Bloomberg board | | DXY | 99.986 | — | −0.89% | First sub-100 print of the episode; Bloomberg Dollar Spot Index −0.9%; yen +2.5% on confirmed MOF intervention |
Sources: CNBC market live blog and Bloomberg Markets Wrap (“Chip Stocks Post Biggest Advance Since April 2025”) for index closes and narrative; Bloomberg /markets quote board; WSJ Markets (“Microsoft Posts Biggest One-Day Market-Cap Gain for Any U.S. Company”; “Citadel Buys Situational Awareness’s Stock Portfolio”); Investing.com Major Indices, SOX, NDX, Dow-component and trending-stock boards (all single-name closes stamped 15:59:59); U.S. Treasury Text View; CME FedWatch and Investing.com Fed Rate Monitor; BEA/BLS via CNBC; TradingEconomics commodities, currencies and world-indices boards; Finviz Groups (rendered); NY Fed July and August calendars; Earnings Whispers day pages. Full source list in the companion US_CrossAsset_Daily_2026-07-30_DataNotes.txt.
The tape in one paragraph. This was a forced-covering rally with a very specific fingerprint, and the bond market refused to sign it. The proximate catalyst was Microsoft +15.51% to $451.10, adding roughly $450 billion of market value — the largest one-day market-cap gain by any U.S. company ever (Bloomberg, WSJ) — on fiscal-Q4 EPS $4.74 vs. $4.24, revenue $90.01bn vs. $87.62bn, Azure +43% constant currency against a 40.2% StreetAccount bar, and FY2026 Azure revenue past $100bn for the first time; CFO Amy Hood then guided capex higher for FY2027 citing “demand signals across our portfolio,” and the market paid for it. But the mechanism of the day was liquidation, not conviction: Leopold Aschenbrenner’s Situational Awareness — $45bn at its early-July peak — was forced to unwind its entire public book, and Citadel bought the stock portfolio (WSJ, CNBC). The names that ripped are the fund’s disclosed Q1 longs and the mirror image of its software shorts: SanDisk +25.99%, IREN +30.54%, Nebius +27.13%, Bloom Energy +26.42%, CoreWeave +21.51%, Micron +18.36%, Lam Research +17.98%, SK Hynix ADR +17.52%, AMD +13.00%, Marvell +12.18% — while the short leg unwound the other way, Adobe −5.91% one day after +5.69%, Salesforce −4.06%, Fortinet closing +0.62% after trading as high as $169.26 having been +12% after hours. Nvidia managed only +2.65% against a SOX up 8.19% — the index leader was the worst-performing large semi on the tape, which is exactly what a positioning squeeze looks like and exactly what a fundamental re-rating does not. JPMorgan’s Nikolaos Panigirtzoglou supplied the supporting evidence: investor deleveraging in tech, semis and memory “has advanced faster than we had previously anticipated,” leaving “more limited room for any further deleveraging.” CNBC’s Jim Cramer called it “a clearing event.” Bloomberg’s strategist Michael Ball was blunter: the rebound “looks more like a positioning squeeze than the start of a durable risk-on move.” The macro cooperated but did not lead: Q2 GDP grew 1.5% versus 1.8% consensus (from 2.1%), the June PCE deflator fell 0.1% m/m with the annual rate at 3.7%, core PCE +0.1% m/m and 3.3% y/y against 0.2%/3.3% forecasts, and initial claims rose 9,000 to 197,000. The Bank of England held at 3.75% on a 6–3 vote, euro-zone Q2 GDP surprised at +0.4% versus +0.2%, and Japan carried out a yen-buying intervention with the U.S. conducting a rate check (Nikkei), sending USD/JPY down 2.30% to 159.63, the yen’s biggest gain since December 2022. The tell that should govern positioning: the official par 30-year rose another basis point to 5.21%, a fresh 19-year closing high, and the 20-year to 5.22%, on a day the Nasdaq 100 rose 3.36% and the VIX fell 17.28%. Equity vol collapsed; the term premium did not. After the bell the capex referendum continued: Amazon +3.90% into the close and higher again after hours on AWS +37% to $42.2bn — the fastest in eighteen quarters — total revenue past $200bn for the first time and an AWS operating margin of 39.4%, with Andy Jassy guiding 2026 capex to $220bn, while Apple −1.41% in the session and down a further 3–4% after hours despite EPS $2.02 on $109.4bn because Services came in at $30.7bn against a $31.3bn bar and Greater China at $18.8bn against $19.6bn (§4). And Meta −7.98% to $538.88 finished as the day’s most conspicuous loser, dragging Communication Services to −2.26%, the worst of eleven Finviz groups on a +1.66% index day. |
| 2 · Market Hot Spots (ranked by tradability) |
1. Nvidia +2.65% while the SOX rose 8.19% mechanically identifies this as a forced-liquidation squeeze, not a re-rating. The dispersion is the evidence: SanDisk +25.99% to $1,279.96, Micron +18.36% to $874.66, Lam Research +17.98% to $297.72, SK Hynix ADR +17.52% to $149.00, AMD +13.00% to $485.39, Marvell +12.18%, TSMC +7.63%, Arm +7.40%, Broadcom +4.73%, Nvidia +2.65%. In a genuine demand re-rating the liquid megacap leads and the beta follows. Here the most-shorted, most-deleveraged names led by 10–23 points and the most liquid name lagged everything. The proximate flow is named and dated: Situational Awareness was forced to liquidate its entire public equity book and Citadel bought the portfolio (WSJ), with Bank of America, Goldman Sachs and JPMorgan Chase marketing baskets of the firm’s longs and shorts ahead of Thursday’s open (CNBC). Expression: fade single-name strength in the highest-squeeze names against long NVDA or SOXX — the pair converges once the basket is placed. Invalidation: a second, unrelated fund unwind, or an AI-capex datapoint (Amazon’s $220bn 2026 guide is one) that lifts the whole complex together rather than dispersively. |
2. JPMorgan put a date on the end of the deleveraging — the most important sell-side call of the week. Nikolaos Panigirtzoglou, in a Wednesday note published into Thursday’s tape: “We find that investors deleveraging in the tech and semiconductor space, including memory stocks, has advanced faster than we had previously anticipated. As a result, we now see more limited room for any further deleveraging,” adding that the April/May hedge-fund leverage build has been “largely unwound.” That is a flow call, not a fundamental call, and it is falsifiable within days. If it is right, the SOX’s −21% July — on track for the worst month since 2008 (Bloomberg) — has removed the mechanical seller and the next move belongs to the AMD (Tue 8/4 4:15 PM) and Western Digital/SanDisk (Wed 8/5) prints. If it is wrong, Thursday was a bounce in a downtrend, and the invalidation is a lower low in the SOX below 10,445.4, Wednesday’s session low. |
3. UBS initiated SK Hynix at Buy, $204 target — the cleanest bull case on memory in print. Analyst Nicolas Gaudois: “At current valuation, we do not believe SK Hynix shares fully reflect structurally higher memory profitability,” with DRAM demand growth forecast to reach 36% in 2027 from 22% in 2026 on agentic-AI workloads, plus long-term agreements, advancing 2027 HBM negotiations and capacity expansion. The $204 target implied 61% upside from Wednesday’s $126.79 close; after Thursday’s +17.52% to $149.00 the implied upside is 36.9%. Note the sequencing — this initiation landed after SK Hynix posted record Q2 revenue of ₩79.32tn and operating profit of ₩60.54tn that still missed LSEG SmartEstimates. The bull case now requires the market to stop marking memory to the quarterly print and start marking it to the 2027 contract book. |
4. The bond market did not ratify the equity rally, and that divergence is the most underweighted risk into next week. Official par: 30Y +1 bp to 5.21% (a fresh 19-year closing high), 20Y +1 bp to 5.22%, 10Y +1 bp to 4.68%, 2Y +1 bp to 4.23% — a near-parallel one-basis-point shift, with 1M +6 bp, on a day the NDX rose 3.36% and the VIX fell 17.28%. Long-end yields rose through a soft GDP print, an in-line core PCE and a 17% collapse in equity vol. Bloomberg’s headline for the week — “Bond Yields at 19-Year High Send Warsh Credibility Warning” — is the frame. The practical consequence: the equity market has repriced the flow problem and has not repriced the discount-rate problem. Watch 30Y 5.35% as the level at which the multiple, not the sector mix, reprices. |
5. Communication Services −2.26% was the worst sector on a +1.66% index day, and it is one stock. Meta −7.98% to $538.88 on 38.1m shares, after Wednesday’s EPS $6.18 vs. $7.17 (a $2.4bn legal charge) and a Q3 revenue guide of $61–64bn against a $63.15bn bar — and, disclosed Thursday, almost $700 billion of already-committed future spending through long- and short-term agreements on AI data centres and cloud (Bloomberg). That number is why the stock could not participate in an AI melt-up: the market is no longer pricing Meta’s AI spend as optionality, it is pricing it as a liability schedule. The pair that expresses the day’s actual rule — long MSFT / short META — worked by 23.5 points in a single session: convert the spend and you get paid; commit the spend and you get charged for it. |
6. Goldman Sachs +4.50% and JPMorgan +1.80% reversed Wednesday’s most anomalous print, and the reversal is informative. Wednesday, GS −5.09% and JPM −3.50% into a 15-bp steepening was the session’s contradiction — banks selling off on a steeper curve. Thursday they led the Dow’s financials back with the curve essentially unchanged. That sequencing says Wednesday’s bank selloff was de-risking flow tied to the same fund liquidation (GS and JPM were named prime brokers on the unwind), not a credit-tail repricing. Financials +1.38%, second-best of eleven groups. If banks hold these gains through next week’s ISM/payrolls sequence with 2s30s still near +98 bp, the “policy-error-with-a-credit-tail” read from Wednesday is falsified. |
7. The BoJ trade fired before the meeting: confirmed intervention, a U.S. rate check, and the biggest yen gain since December 2022. USD/JPY −2.30% to 159.634 (TradingEconomics; Bloomberg 4 p.m. 159.38, yen +2.5%), with the Nikkei reporting that Japan carried out yen-buying intervention and that the U.S. conducted a rate check — an unusually explicit two-government signal. The BoJ is expected to hold at 1% on Friday, so this is a pre-emptive defence of the level, not a policy pivot. The second-order read: a 2.3% yen move on a day the Nasdaq 100 rose 3.36% is not the classic carry-unwind pairing, which means the funding-currency channel was not the driver of the AI drawdown — a genuinely useful negative result. |
8. Bloom Energy +26.42% re-opens the power-for-AI trade Baird shut on Tuesday — but on flow, not thesis. BE posted adjusted EPS $0.78 versus $0.41 expected on record revenue of $1.07bn versus $827m and closed −1.89% on Wednesday; Thursday it gained 26.42% to $207.02. The same pattern ran through Vertiv, Generac and Nebius (+27.13%), and Caterpillar itself rebounded +3.35% to $808.94. What changed was not the permitting backlash Baird flagged when it cut CAT to Neutral with a $900 target — the zoning, water and interconnect constraints are unchanged — but the disappearance of the forced seller. Treat the complex as flow-cleared, not thesis-cleared; CAT’s Tuesday 8/4 6:30 a.m. print is where the thesis gets adjudicated. |
9. ICE agreed to buy MarketAxess for $167 a share in cash — the fixed-income market-structure trade of the year. A 33% premium to the 7/29 close, roughly $5.7–6.0bn enterprise value. ICE CEO Jeff Sprecher: acquiring MarketAxess “is the natural next step in that journey,” uniting ICE’s market infrastructure and data analytics with MarketAxess’s network of roughly 2,100 institutional investors and broker-dealers across more than 90 countries; close expected in H1 2027, subject to shareholder and regulatory approval. BofA Securities advised ICE, J.P. Morgan advised MarketAxess; WSJ prices the deal at $6 billion. Read across to Tradeweb and Cboe — and note it lands in the week the bond market is delivering 19-year-high yields. |
10. The fades and the non-participants are the honest scoreboard. Fortinet closed +0.62% at $154.17 after trading to $169.26 intraday, having been up more than 12% after hours Wednesday on a beat and a raised outlook — a full round trip in a green tape. Adobe −5.91% gave back Wednesday’s +5.69% and then some. Robinhood −3.60%, Qualcomm −2.62%, Crocs −10%+ despite a beat and a raised FY forecast (weak Q2 margins, light Q3 bar; not an S&P 500 constituent), Norwegian Cruise −7%+ after cutting guidance. Inside the Dow the defensives were sold to fund the squeeze: Walmart −2.73%, J&J −3.66%, Travelers −3.36%, Verizon −2.34%, Disney −2.38%, Nike −2.20%, IBM −2.08%, P&G −1.47%. That is the signature of a funding rotation, not new money. |
| 3 · Sector Performance — July 30, 2026 (Finviz classification, rendered) |
| Sector | 1-Day | 1-Week | YTD | | Technology | +5.55% | −1.04% | +17.16% | | Basic Materials | +2.21% | +1.82% | +9.38% | | Financial | +1.38% | +2.14% | +7.29% | | Consumer Cyclical | +1.19% | +2.54% | −7.43% | | Industrials | +1.17% | −2.67% | +10.27% | | Energy | +0.99% | −0.45% | +32.03% | | Utilities | +0.28% | −3.21% | +4.33% | | Healthcare | −1.15% | +1.24% | +6.00% | | Real Estate | −1.16% | +0.84% | +12.03% | | Consumer Defensive | −1.89% | +3.19% | +8.87% | | Communication Services | −2.26% | +1.29% | −4.70% |
Four red out of eleven on a +1.66% index day — that is the funding-rotation signature, and it is the most diagnostic feature of the table. Technology’s +5.55% is the largest single-day sector gain of the year and, per CNBC, the S&P 500 information-technology sector’s biggest one-day advance since April 2025, dragging the group back into the green on the week and to +16% for 2026 on the official S&P classification. But the money to buy it came out of the defensives: Consumer Defensive −1.89% after a +3.19% week, Healthcare −1.15%, Real Estate −1.16% — three groups that had absorbed the rotation during the five-day tech slide and gave it straight back. Communication Services −2.26% is a single-stock artefact: Meta −7.98% on 38.1m shares, and the group is now the only sector besides Consumer Cyclical with a negative YTD (−4.70%). Note what did not happen: Utilities managed only +0.28% and Real Estate fell on a day the AI complex ripped 8% — the market is still discriminating between the compute layer and the power/permitting layer that Baird’s Caterpillar downgrade repriced on Wednesday. Industrials’ +1.17% recovered barely a third of Wednesday’s −3.40% and the group is still −2.67% on the week. Energy +0.99% with crude lower (WTI −0.27%, Brent −1.52%) is the refining margin at work — Bloomberg’s “US Refiners See Billions in Profits From Global Fuel Crunch” and Shell’s $9.84bn adjusted Q2, its best since Q2 2022 versus an $8.79bn LSEG bar. Energy remains +32.03% YTD, 14.87 points clear of Technology, still the widest sector spread of the year.
Source: Finviz Groups via rendered page; Finviz buckets are not official GICS/S&P sector indices, which is why CNBC’s “+16% in 2026” for S&P information technology and Finviz’s +17.16% for “Technology” differ. Reconciliation: ten of eleven groups reconcile against the prior report’s YTD compounded by Thursday’s 1-day move — Energy 1.3075 × 1.0099 = +32.04% vs. 32.03% shown; Technology 1.1102 × 1.0555 = +17.18% vs. 17.16%; Industrials 1.0899 × 1.0117 = +10.27% vs. 10.27%; Financial 1.0583 × 1.0138 = +7.29% vs. 7.29%; Consumer Cyclical 0.9149 × 1.0119 = −7.42% vs. −7.43%; Utilities 1.0403 × 1.0028 = +4.32% vs. 4.33%. Maximum deviation across those ten: 0.02 pt. The exception is Communication Services: 0.9776 × 0.9774 = −4.45% versus −4.70% shown, a 0.25-pt gap — consistent with a constituent/weight refresh in a group whose 1-day move is dominated by one name.
| 4 · Movers & Single-Name Catalysts |
All levels and percentages are Investing.com closing prints stamped 15:59:59 ET unless otherwise noted.
Upside
| Name | Close | %Chg | Catalyst | | Microsoft (MSFT) | $451.10 | +15.51% | The largest one-day market-cap gain by any U.S. company on record, roughly $450bn (Bloomberg, WSJ). FQ4 EPS $4.74 vs. $4.24 on $90.01bn vs. $87.62bn; Azure +43% cc vs. 40.2%; FY26 Azure past $100bn; capex + finance leases $41bn, +69% y/y. CFO Amy Hood guided further FY27 capex growth on “demand signals across our portfolio”; building useful lives extended to 25 years from 15, implying roughly $175bn of capex. Piper Sandler PT to $550 from $540 (Overweight), +21.9% to the close; Evercore ISI to $528 from $525, +17.0% to the close (both struck off Wednesday’s $390.54) | | IREN Ltd (IREN) | $38.26 | +30.54% | Highest-beta name on the squeeze list; not an S&P 500 constituent | | Nebius (NBIS) | $188.43 | +27.13% | Reverses Wednesday’s −12.65%; not an S&P 500 constituent | | Bloom Energy (BE) | $207.02 | +26.42% | Adj. EPS $0.78 vs. $0.41 on record revenue $1.07bn vs. $827m — a print the market ignored Wednesday (closed −1.89%). Not an S&P 500 constituent | | SanDisk (SNDK) | $1,279.96 | +25.99% | Largest S&P 500 gainer of the session; range $1,124.00–$1,285.48. Reports Wed 8/5 4:05 PM | | CoreWeave (CRWV) | $73.90 | +21.51% | Named by Bloomberg among Situational Awareness holdings that soared more than 20%; not an S&P 500 constituent | | Micron (MU) | $874.66 | +18.36% | Reverses Wednesday’s −10.07% close-on-the-low at $737.88; range $789.00–$882.50 | | Lam Research (LRCX) | $297.72 | +17.98% | Delayed payment for Wednesday’s beat-and-raise (FQ1-27 guide $8.1bn ±$400m, $2.15 ±$0.15) after closing −7.04% on the day of the print; record quarterly revenue and earnings | | SK Hynix ADR (SKHY) | $149.00 | +17.52% | UBS initiated Buy, $204 target, +36.9% to the close; not an S&P 500 constituent | | AMD (AMD) | $485.39 | +13.00% | Range $457.68–$498.20; reports Tue 8/4 4:15 PM — now the most important semi print of next week | | Marvell (MRVL) | $183.30 | +12.18% | — | | TSMC ADR (TSM) | $403.27 | +7.63% | — | | Arm (ARM) | $241.54 | +7.40% | Range $232.67–$268.48 — closed 10.0% off its high; not an S&P 500 constituent | | Broadcom (AVGO) | $387.84 | +4.73% | — | | Goldman Sachs (GS) | $1,024.86 | +4.50% | Best Dow name; reverses Wednesday’s −5.09% | | Amazon (AMZN) | $235.50 | +3.90% | Higher again after hours — see “After the close” | | Tesla (TSLA) | $308.85 | +3.53% | — | | Caterpillar (CAT) | $808.94 | +3.35% | Bounce from Baird’s Neutral/$900 downgrade; reports Tue 8/4 6:30 AM | | Boeing (BA) | $220.90 | +3.22% | — | | Nvidia (NVDA) | $195.04 | +2.65% | 123.7m shares — the largest volume on the tape and the smallest percentage gain of any major semi | | Mastercard (MA) | $577.35 | +2.49% | Reported BMO Thursday | | American Express (AXP) | $337.52 | +1.82% | — | | JPMorgan (JPM) | $350.93 | +1.80% | Reverses Wednesday’s −3.50% | | Starbucks (SBUX) | $105.85 | +1.64% | — | | Cisco (CSCO) | $113.56 | +0.96% | — | | Chevron (CVX) | $192.52 | +0.34% | Reports Fri 7/31 6:15 AM | | MarketAxess (MKTX) | — | — | Agreed to be acquired by ICE for $167 per share in cash, a 33% premium; ~$5.7bn EV (ICE) / $6bn (WSJ); close expected H1 2027 |
Downside
| Name | Close | %Chg | Catalyst | | Meta Platforms (META) | $538.88 | −7.98% | 38.1m shares; range $524.49–$539.88 — closed at the session high, the only constructive detail. Thursday’s disclosure: almost $700bn of future spending already committed through long- and short-term AI/data-centre and cloud agreements (Bloomberg) | | Adobe (ADBE) | $247.87 | −5.91% | Mirror image of Wednesday’s +5.69% — Adobe was a named short in the Situational Awareness book, and the unwind ran both directions | | Salesforce (CRM) | $180.74 | −4.06% | — | | Johnson & Johnson (JNJ) | $255.82 | −3.66% | Defensive sold to fund the squeeze | | Robinhood (HOOD) | $86.61 | −3.60% | Reported Wednesday AMC | | Travelers (TRV) | $375.95 | −3.36% | — | | Walmart (WMT) | $111.10 | −2.73% | — | | Qualcomm (QCOM) | $151.60 | −2.62% | CEO Cristiano Amon: smartphone demand “remains strong” but rising device costs are “keeping the market down” | | Walt Disney (DIS) | $96.14 | −2.38% | Reports Wed 8/5, “Before Open” — a new addition to the calendar (§12) | | Verizon (VZ) | $46.12 | −2.34% | — | | Nike (NKE) | $42.27 | −2.20% | — | | IBM | $221.74 | −2.08% | — | | Procter & Gamble (PG) | $143.95 | −1.47% | — | | Home Depot (HD) | $333.35 | −1.45% | — | | Apple (AAPL) | $333.43 | −1.41% | Down a further 3–4% after hours — see “After the close” | | McDonald’s (MCD) | $268.44 | −1.13% | — | | 3M (MMM) | $176.09 | −1.05% | — | | Visa (V) | $366.27 | −0.67% | — | | Coca-Cola (KO) | $88.52 | −0.63% | — | | Netflix (NFLX) | $73.17 | −0.62% | — | | Merck (MRK) | $129.79 | −0.44% | — | | Crocs (CROX) | — | −10%+ | Beat fiscal Q2 and raised the FY forecast, but Q2 margins missed and Q3 estimates were light; +16% YTD. Not an S&P 500 constituent | | Norwegian Cruise (NCLH) | — | −7%+ | Beat but cut guidance. CEO John Chidsey: “it’s mostly on us, not the macro”; problems are “self-inflicted”; Q1 2027 flagged as a drag on the year | | Jersey Mike’s (JMKE) | $21.63 | −5.96% | IPO priced at $23, opened at $21; 43.5m shares raising ~$1bn at a $7.3bn valuation | | Fortinet (FTNT) | $154.17 | +0.62% | Listed here as the day’s most instructive fade: +12% after hours Wednesday, traded to $169.26 intraday, closed 8.9% off its high and essentially flat |
After the close. Amazon (AMZN) — Q2 net sales $200.6bn (+20% y/y) vs. $196.47bn expected, the first $200bn quarter; AWS revenue $42.2bn (+37% y/y), the fastest in eighteen quarters, vs. $40.54bn expected, a $169bn annualised run-rate; AWS operating income $16.6bn (from $10.2bn) at a 39.4% margin; total operating income $27.5bn, +43%. CEO Andy Jassy guided 2026 capital spending to $220bn. Bloomberg: cloud revenue accelerated for a fifth straight quarter, “easing investors’ concerns that the company won’t produce a return on its spending.” Shares jumped after hours. — Apple (AAPL) — FQ3 EPS $2.02 vs. $1.89 on revenue $109.4bn vs. $108.8bn; iPhone $54.2bn vs. $53.5bn (from $44.5bn a year ago); but Services $30.7bn vs. $31.3bn and Greater China $18.8bn vs. $19.6bn (from $15.4bn). Shares fell roughly 3–4% in extended trading. — Reddit (RDDT) projected strong sales but no new AI deals; Coinbase (COIN) posted a loss as revenue fell for another quarter; Oracle (ORCL) said it will expand its Google Cloud partnership to bring Gemini AI tools to business customers (all Bloomberg). |
| 6 · U.S. Treasury Yields — Official Par Curve (Treasury.gov, 3:30 PM ET) |
| Maturity | 7/30/26 | 7/29/26 | Δ 1-Day (bp) | 7/23/26 | Δ 1-Wk (bp) | | 1M | 3.79 | 3.73 | +6 | 3.82 | −3 | | 1.5M | 3.80 | 3.80 | 0 | 3.90 | −10 | | 2M | 3.84 | 3.83 | +1 | 3.95 | −11 | | 3M | 3.82 | 3.83 | −1 | 3.95 | −13 | | 4M | 3.92 | 3.91 | +1 | 4.04 | −12 | | 6M | 3.98 | 3.97 | +1 | 4.09 | −11 | | 1Y | 4.04 | 4.04 | 0 | 4.15 | −11 | | 2Y | 4.23 | 4.22 | +1 | 4.37 | −14 | | 3Y | 4.30 | 4.29 | +1 | 4.40 | −10 | | 5Y | 4.38 | 4.37 | +1 | 4.46 | −8 | | 7Y | 4.52 | 4.51 | +1 | 4.58 | −6 | | 10Y | 4.68 | 4.67 | +1 | 4.71 | −3 | | 20Y | 5.22 | 5.21 | +1 | 5.20 | +2 | | 30Y | 5.21 | 5.20 | +1 | 5.17 | +4 |
Curve: 2s10s +45 bp (0 d/d; +11 w/w); 3M10Y +86 bp (+2 d/d; +10 w/w); 2s30s +98 bp (0 d/d; +18 w/w).
Read — a near-perfect parallel one-basis-point shift, and on this tape “parallel” is the hawkish outcome. Twelve of fourteen tenors rose exactly 1 bp; the 3M fell 1 bp and the 1M rose 6 bp (a month-end bill-supply effect at the very front, not a policy signal). No curve segment moved: 2s10s unchanged at +45 bp, 2s30s unchanged at +98 bp, 3M10Y +2 bp to +86 bp. The diagnostic matters because of what the day contained. The bond market received (i) a Q2 GDP miss at 1.5% versus 1.8%, (ii) a negative monthly PCE deflator (−0.1%) with core at 0.1% versus 0.2% expected, (iii) a 17.28% collapse in the VIX and (iv) a 3.36% Nasdaq 100 rally — and priced all four at one basis point across the entire curve, leaving the 30-year at a fresh 19-year closing high of 5.21% and the 20-year at 5.22%. A soft-growth, soft-inflation print that cannot pull the long end down is the definition of a term-premium-anchored curve: the compensation being demanded is for policy credibility and fiscal supply, not for the next two data points. On the week the picture is a very large bull-steepening: the entire front and belly rallied hard (2Y −14 bp, 3M −13, 4M −12, 2M/6M/1Y −11, 1.5M/3Y −10, 5Y −8, 7Y −6, 10Y −3) while the 20Y rose 2 bp and the 30Y rose 4 bp — 2s30s +18 bp on the week, 3M10Y +10 bp, 2s10s +11 bp. In one week the market removed roughly 14 bp of two-year policy expectation (consistent with §8: CME cumulative September hike odds fell from 82.3% to 63.4%) and simultaneously added 4 bp to the 30-year. That is not an easing trade; it is a relocation of tightening from the policy rate to the term premium, and it is now two weeks old. |
Vendor reconciliation. Bloomberg’s 4 p.m. New York marks had the 10Y −1 bp at 4.67%, the 2Y −4 bp at 4.24% and the 30Y +1 bp at 5.21%, versus official 3:30 p.m. par closes of 4.68% (+1), 4.23% (+1) and 5.21% (+1). The levels agree to within one basis point at every tenor; the apparent 5-bp gap at the 2-year is a prior-day baseline artefact — CNBC reported the 2Y at 4.283% in Asian trading Thursday morning, so Bloomberg measures from an overnight high while Treasury measures 3:30 p.m. to 3:30 p.m. Official par figures are used throughout. Weekly change is computed against Thursday 7/23 to match the weekday.
| 7 · U.S. Macroeconomic Calendar |
Current week (Jul 27–31) — released
| Date (ET) | Release | Actual | Consensus / Prior | Take | | Mon 7/27 8:30 | Advance Durable Goods, June | +0.3% m/m to $334.8bn (ex-transport +0.6%) | vs. +2.1% (Dow Jones); −4.0% May | Clear miss; computers & electronics +3.1% vs. transportation −13.5% | | Mon 7/27 10:30 | Dallas Fed Manufacturing, July | General activity +1.3 | prior 0.0 | Accelerating with uncertainty falling | | Tue 7/28 10:00 | Conference Board Consumer Confidence, July | 90.8 | vs. 92.0 cons.; 92.2 June (rev. up) | Soft but backward-looking | | Tue 7/28 10:00 | Richmond Fed Manufacturing, July | Composite +5 | vs. 10 expected | Headline miss; shipments 8 from 3 | | Wed 7/29 10:30 | EIA Weekly Petroleum Status Report | Crude stocks −7.2 mb to 404.5 mb, 7% below the 5-yr average | prior week also a large draw | The physical corroboration behind Wednesday’s crude spike; partially unwound Thursday as Hormuz transits resumed | | Wed 7/29 2:00 PM | FOMC + Warsh press conference | Held 3.50–3.75%, fifth straight meeting; 9–3 (Hammack, Kashkari, Logan dissenting for +25 bp); IORB 3.65% effective 7/30 | Hold was the base case | The credibility event that produced the bear-steepener (§6, §8) | | Thu 7/30 8:30 | GDP, 1st release, Q2 | +1.5% annualised | vs. +1.8% cons.; +2.1% Q1 | Miss. Bloomberg: “solid consumer spending and business investment signaled underlying strength even as growth moderated.” A 1.5% handle with 3.3% core PCE is the mix the three dissenters like least | | Thu 7/30 8:30 | Personal Income & PCE Deflator, June | Headline −0.1% m/m, +3.7% y/y; core +0.1% m/m, +3.3% y/y | headline in line; core vs. +0.2% / +3.3% | The most important print of the week, and it was benign — core undershot by a tenth on the month and matched on the year. It removed the upside tail in front of a Fed that had just refused to pre-commit, and it still did not move the long end (§6) | | Thu 7/30 8:30 | Initial Claims (wk ended 7/25) | 197,000, +9,000 from the prior week’s revised level | prior ~188k revised | Off the multi-decade low but still historically very tight; not enough to change the labour read before 8/7 payrolls |
Remaining this week
| Date (ET) | Release | Sensitivity | | Fri 7/31 8:30 | Employment Cost Index, Q2 | High — the single most important remaining print of the month. With core PCE in line and GDP soft, ECI is the last wage read before the August 7 payroll. A hot ECI is the one datapoint that revives the September hike case the front end just spent a week dismantling (§8); a soft ECI leaves 63.4% cumulative hike odds looking rich | | Fri 7/31 10:00 | Michigan Consumer Survey (Final) | Medium — the inflation-expectations sub-index. The survey window straddles a crude round trip in both directions (Brent $84 → $90 → $89), so read the revision rather than the level | | Fri 7/31 10:00 | NY Fed Multivariate Core Trend Inflation | Low/Medium — but genuinely useful this month as a cross-check on whether Thursday’s benign core PCE was signal or composition | | Fri 7/31 12:45 | NY Fed Staff Nowcast | Low |
Next week (Aug 3–7) — jobs week (NY Fed official August calendar, re-verified this session)
| Date (ET) | Release | Sensitivity | | Mon 8/3 10:00 | ISM Manufacturing, July; Construction Spending | High — first hard read on the post-oil-spike cycle; prices-paid is the tell now that Brent has round-tripped | | Tue 8/4 8:30 / 10:00 | Advance International Trade in Goods; Trade Balance; JOLTS, June; Manufacturing Shipments & Orders | Medium/High — tariff pass-through appears in trade first; JOLTS is the tightness cross-check on a 197k claims print | | Wed 8/5 8:15 / 9:00 / 10:00 | ADP National Employment, July; NY Fed Labor Market Tightness Index; ISM Non-Manufacturing, July | High — services prices-paid is the oil/tariff transmission line into core inflation | | Thu 8/6 8:30 / 10:00 / 11:30 | Initial Claims; Productivity & Costs (Preliminary), Q2; Global Supply Chain Pressure Index; Wholesale Trade | High — unit labour costs confirm or refute Friday’s ECI; GSCPI is unusually relevant with Hormuz traffic still abnormal | | Fri 8/7 8:30 | Employment Situation, July (payrolls) | Very High — the first payroll report after a 9–3 hold, and the largest scheduled risk of the next month | | Fri 8/7 11:00 / 12:45 | Survey of Consumer Expectations; NY Fed Staff Nowcast | Medium |
Look-ahead framing. Thursday resolved the inflation tail benignly and the growth tail slightly negatively, and the market’s response was to reprice equity positioning by 3.36% on the NDX and rates by one basis point. That asymmetry is the whole story. The front end has already done its repricing — CME cumulative September hike odds have fallen from 82.3% a week ago to 63.4% and 2Y yields are 14 bp lower on the week — so the marginal information in Friday’s ECI and next Friday’s payrolls now lands on a curve that is long disinflation at the front and short credibility at the back. Practically: a soft ECI buys the 2-year very little and the 30-year nothing; a hot ECI hurts the 2-year a lot and the 30-year more. The hierarchy into next week is ISM Manufacturing prices-paid (Mon) → ADP and ISM Services prices-paid (Wed) → unit labour costs (Thu) → payrolls (Fri 8/7), with AMD (Tue 4:15 PM) and Disney, Lilly, Uber and CVS (Wed BMO) as the equity-side catalysts. Where a verified consensus was not obtainable, the sensitivity column describes market sensitivity rather than inventing an expectation figure. |
| 8 · Fed Funds Futures & Rate Path (CME FedWatch / Investing.com Fed Rate Monitor) |
Current target range: 3.50–3.75% (held 9–3 on Wednesday 7/29; Hammack, Kashkari and Logan dissented for +25 bp; IORB 3.65% effective 7/30). Chair: Kevin Warsh. Next FOMC: Wednesday, September 16, 2026, 2:00 PM ET. Investing.com cards timestamped Jul 30, 2026 05:45 PM EDT; CME FedWatch headline stamped 30 Jul 2026 04:57:54 CT.
CME FedWatch headline — September 16, 2026 meeting
| Target rate (bps) | NOW | 1 DAY (29 Jul) | 1 WEEK (23 Jul) | 1 MONTH (30 Jun) | | Ease (below 350) | 0.0% | 0.0% | 0.0% | 0.0% | | 350–375 — hold (current) | 36.6% | 41.7% | 17.6% | 32.2% | | 375–400 (+25 bp) | 63.4% | 57.1% | 57.3% | 50.6% | | 400–425 (+50 bp) | 0.0% | 1.2% | 25.0% | 17.2% | | Cumulative hike (≥375) | 63.4% | 58.3% | 82.3% | 67.8% |
Column sums: 100.0 / 100.0 / 99.9 / 100.0 — the 1-week column sums to 99.9 on CME’s own rounding.
Reconciling CME against Investing.com — in-text, not in a footnote. For the same September meeting, CME puts the hold at 36.6% and the +25 bp at 63.4%; Investing.com puts them at 35.9% and 64.1% — a 0.7-point gap in each direction. Two causes, both benign: (i) the timestamps are twelve minutes apart (CME 4:57:54 p.m. CT = 5:57:54 p.m. ET; Investing.com 5:45 p.m. ET), which straddles the fed funds futures settlement window; and (ii) CME publishes from the ZQ contract mid at its snapshot instant while Investing.com’s calculator uses its own 30-day fed funds futures price (its September card shows a future price of 96.290). The gap is smaller than one CME rounding increment applied twice and changes no conclusion. The larger and more instructive gap is in the “previous day” columns: CME’s 1-DAY (29 Jul) hold of 41.7% versus Investing.com’s previous-day hold of 37.9%. Investing.com’s previous-day column is a fixed daily snapshot taken at a set hour, not the prior evening’s settlement; CME’s is an end-of-29-July basis. Yesterday’s report quoted CME cumulative September hike odds of 57.4%; CME’s own 1-DAY column now implies 58.3% for the same day — a 0.9-point restatement from exactly this snapshot effect. Use one vendor’s columns consistently within a comparison and never mix them. |
(1) 2026 meeting distributions — Investing.com current / [prev-day] / [prev-week]
| Meeting | 3.50–3.75 (hold) | 3.75–4.00 (+25) | 4.00–4.25 (+50) | 4.25–4.50 (+75) | 4.50–4.75 (+100) | Cumulative hike | | Sep 16, 2026 | 35.9 / [37.9] / [19.5] | 64.1 / [62.1] / [55.8] | 0.0 / [0.0] / [24.7] | 0.0 / [0.0] / [0.0] | 0.0 / [0.0] / [0.0] | 64.1 / [62.1] / [80.5] | | Oct 28, 2026 | 25.1 / [26.6] / [13.2] | 55.7 / [54.9] / [44.1] | 19.1 / [18.6] / [34.7] | 0.0 / [0.0] / [8.0] | 0.0 / [0.0] / [0.0] | 74.8 / [73.5] / [86.8] | | Dec 9, 2026 | 15.4 / [14.8] / [8.1] | 43.8 / [42.4] / [32.2] | 33.4 / [34.6] / [38.3] | 7.5 / [8.2] / [18.3] | 0.0 / [0.0] / [3.1] | 84.7 / [85.2] / [91.9] |
Row sums: September 100.0 / 100.0 / 100.0; October 99.9 / 100.1 / 100.0; December 100.1 / 100.0 / 100.0. No easing is priced at any 2026 meeting in any column — 0.0% below 3.50% throughout.
Multi-day momentum. On Investing.com’s card the September +25 bp bucket sits at 64.1%, up from 62.1% a day earlier and up from 55.8% a week earlier — but that single bucket is misleading, because a week ago the market also carried a second hike: the 7/23 card priced 24.7% at +50 bp, for a cumulative 80.5%. On CME’s cleaner cumulative basis the arc is 82.3% a week ago → 67.8% a month ago → 58.3% a day ago → 63.4% now: September hike odds are up 5.1 points on the day and down 18.9 points on the week. The December cumulative tells the same story with more resolution: ≥+25 bp by December 84.7% now versus 91.9% a week ago; ≥+50 bp 40.9% versus 59.7%; ≥+75 bp 7.5% versus 21.4%. The market has not removed the tightening cycle — it has removed the second and third hikes and pushed the first one out.
(2) 2027 meeting path — modal range, probability, cumulative above/below 3.50–3.75%
| Meeting | Modal range | Prob. | Cumulative above 3.75% | Cumulative below 3.50% | Hold | | Jan 27, 2027 | 3.75–4.00 | 38.4% | 87.6% | 0.0% | 12.4% | | Mar 17, 2027 | 4.00–4.25 | 36.2% | 91.0% | 0.0% | 9.1% | | Apr 28, 2027 | 4.00–4.25 | 35.6% | 91.8% | 0.0% | 8.1% | | Jun 9, 2027 | 4.00–4.25 | 35.6% | 91.8% | 0.0% | 8.1% | | Jul 28, 2027 | 4.00–4.25 | 35.6% | 91.8% | 0.0% | 8.1% | | Sep 15, 2027 | 4.00–4.25 | 33.1% | — (withheld) | 1.4% | 11.6% | | Oct 27, 2027 | 4.00–4.25 | 32.0% | 84.8% | 2.2% | 12.9% | | Dec 8, 2027 | 3.75–4.00 | 30.6% | 80.0% | 4.2% | 15.7% |
Row sums: January 100.0; March 100.1; April 99.9; June 99.9; July 99.9; October 99.9; December 99.9. The September 2027 above/below aggregate is withheld because that card’s upper tail was truncated in the retrieved text; the modal range, its probability, the hold and the below-3.50% figure were retrieved cleanly and are shown. The 2027 path peaks in the middle and fades at the ends: the modal range moves from 3.75–4.00% in January to 4.00–4.25% from March through October and back to 3.75–4.00% in December, with cumulative above-range probability topping out at 91.8% in April–July and decaying to 80.0% by December 2027, while the first genuine easing probability appears late and small (1.4% by September 2027, 4.2% by December).
(3a) Year-end 2026 ladder — December 9, 2026 meeting (current / [prev-day] / [prev-week])
| Outcome | Range | Probability | | −75 bp (2.75–3.00) | cut | 0.0% / [0.0%] / [0.0%] | | −50 bp (3.00–3.25) | cut | 0.0% / [0.0%] / [0.0%] | | −25 bp (3.25–3.50) | cut | 0.0% / [0.0%] / [0.0%] | | Hold (3.50–3.75) | unchanged | 15.4% / [14.8%] / [8.1%] | | +25 bp (3.75–4.00) | modal | 43.8% / [42.4%] / [32.2%] | | +50 bp (4.00–4.25) | | 33.4% / [34.6%] / [38.3%] | | +75 bp (4.25–4.50) | | 7.5% / [8.2%] / [18.3%] | | +100 bp (4.50–4.75) | | 0.0% / [0.0%] / [3.1%] | | +125 bp or more | | 0.0% / [0.0%] / [0.0%] |
Cumulative: ≥+25 bp 84.7% / [85.2%] / [91.9%]; ≥+50 bp 40.9% / [42.8%] / [59.7%]; ≥+75 bp 7.5% / [8.2%] / [21.4%]; any cut 0.0% in all three columns. Column sums 100.1 / 100.0 / 100.0.
(3b) Year-end 2027 ladder — December 8, 2027 meeting (current / [prev-day] / [prev-week])
| Outcome | Range | Probability | | −75 bp (2.75–3.00) | cut | 0.0% / [0.0%] / [0.0%] | | −50 bp (3.00–3.25) | cut | 0.4% / [0.6%] / [0.2%] | | −25 bp (3.25–3.50) | cut | 3.8% / [5.0%] / [1.8%] | | Hold (3.50–3.75) | unchanged | 15.7% / [16.7%] / [9.0%] | | +25 bp (3.75–4.00) | modal | 30.6% / [28.8%] / [22.8%] | | +50 bp (4.00–4.25) | | 29.6% / [27.7%] / [30.4%] | | +75 bp (4.25–4.50) | | 15.2% / [15.3%] / [22.8%] | | +100 bp (4.50–4.75) | | 4.1% / [4.9%] / [10.1%] | | +125 bp (4.75–5.00) | | 0.5% / [0.9%] / [2.6%] | | +150 bp (5.00–5.25) | | 0.0% / [0.1%] / [0.4%] | | +175 bp or more | | 0.0% / [0.0%] / [0.0%] |
Cumulative: ≥+25 bp 80.0% / [77.7%] / [89.1%]; ≥+50 bp 49.4% / [48.9%] / [66.3%]; ≥+75 bp 19.8% / [21.2%] / [35.9%]; any cut 4.2% / [5.6%] / [2.0%]. Column sums 99.9 / 100.0 / 100.1.
(4) Rate-path interpretation. One-day changes. The whole 2026 strip firmed modestly for the first hike and softened for the later ones. September +25 bp went 62.1% → 64.1% (Investing.com) and 58.3% → 63.4% on CME’s cumulative basis (+5.1 pt); October cumulative 73.5% → 74.8%; December cumulative 85.2% → 84.7% with ≥+50 bp falling from 42.8% to 40.9% and ≥+75 bp from 8.2% to 7.5%. The 2027 ladder pulled the same way: December-2027 ≥+25 bp rose 77.7% → 80.0% while any-cut fell 5.6% → 4.2%. Translation: Thursday’s data made a near-term hike marginally more likely (soft growth did not soften wages, and 3.3% core PCE is 130 bp above target) while trimming the depth of the cycle.
One-week changes — the arc that matters. A week ago the market priced 82.3% cumulative September hike odds including a 25.0% chance of an immediate +50 bp; today it prices 63.4% with 0.0% at +50 bp. Over the same week December-2026 ≥+50 bp collapsed from 59.7% to 40.9%, ≥+75 bp from 21.4% to 7.5% and December-2027 ≥+75 bp from 35.9% to 19.8%. Yet across that same week the 30-year yield rose 4 bp to a 19-year closing high of 5.21%, the 20-year rose 2 bp, and 2s30s steepened 18 bp (§6). The front end priced out roughly two hikes and the long end got cheaper anyway. The market is not buying disinflation; it is transferring the tightening from the Fed’s balance sheet to the taxpayer’s discount rate.
The named macro hooks. (i) Core PCE at 3.3% y/y — 130 bp above target, unchanged on the year and undershooting only on the month; the three dissenters’ case survives Thursday intact. (ii) Q2 GDP 1.5% versus 1.8% expected, decelerating from 2.1% — the growth leg that argues against acting. (iii) Claims 197k, +9k — off the multi-decade low but still historically tight, which is why the front end could not rally on the GDP miss. (iv) Brent $89.36 (−1.52%) after topping $90 Wednesday, with Hormuz transits picking up — the crude round trip removed the immediate energy passthrough but left heating oil +97.82% and gasoline +89.99% year-on-year, so the level of fuel costs, not the daily change, feeds core services. (v) The BoE held at 3.75% (6–3) and euro-zone Q2 GDP beat at +0.4%, with markets treating a further ECB hike as near-certain — the G3 policy dispersion keeping the dollar sub-100 (§9).
Base case and tails. Base case (≈64%): +25 bp on September 16 to 3.75–4.00%, then a pause — the modal December-2026 outcome is exactly one hike (43.8%) and the modal 2027 path holds 4.00–4.25% for most of the year. Hawkish tail (≈41%): two or more hikes by December 2026 (cumulative ≥+50 bp 40.9%), which requires a hot ECI and confirming ISM prices-paid. Dovish tail (≈15%): no hike at all in 2026 (December hold 15.4%) — which now requires a growth shock, because the inflation shock has already been priced out. There is still no cut priced anywhere in 2026: 0.0% at every meeting, in every column.
Practical trading implication, tied to the next catalyst. The market has spent a week flattening the front and steepening the back, and Thursday’s benign core PCE did not stop the 30-year making a new 19-year high. The clean expression remains a 2s30s or 3M30Y steepener carried into Friday’s 8:30 a.m. ECI — with an important asymmetry: a soft ECI rewards it (the front rallies further, the back does not follow) and a hot ECI also rewards it (the front sells off less than the back, because the market has just demonstrated it will not re-add +50 bp probability). The position loses only if the long end genuinely re-anchors — restored explicit forward guidance, or a refunding announcement that cuts long-end coupon issuance. Size to 30Y 5.35% as the stop-out for the equity side of the book (§13), and note that equity vol is now the cheap leg: VIX at 17.09 after a 17.28% one-day collapse implies roughly a 1.1% daily move into an ECI print, a payroll week and a 19-year-high long bond. |
| Pair | Level | 1-Day | Context | | DXY | 99.986 | −0.89% | First sub-100 print of the episode; Bloomberg Dollar Spot Index −0.9%; −1.43% on the week, −1.38% on the month | | EUR/USD | 1.15268 | +0.52% | Bloomberg 4 p.m. 1.1532, +0.6%; euro-zone Q2 GDP +0.4% vs. +0.2% expected; markets treat a further ECB hike as near-certain | | USD/JPY | 159.634 | −2.30% | Bloomberg 159.38, yen +2.5% — the yen’s biggest gain since December 2022. Nikkei: Japan carried out yen-buying intervention and the U.S. conducted a rate check. BoJ expected to hold at 1% Friday | | GBP/USD | 1.34653 | +0.71% | Bloomberg 1.3472, +0.8%; BoE held Bank Rate at 3.75% on a 6–3 vote; 10-yr gilt −5 bp to 4.98% | | USD/CHF | 0.80503 | −1.06% | The haven cross — the franc was the second-strongest major on a +1.66% S&P day | | AUD/USD | 0.70276 | +1.04% | Best-performing dollar pair; +5.32% YTD, the strongest major of 2026 | | USD/CAD | 1.40128 | −0.25% | Muted against a −0.9% crude tape — the loonie is trading the dollar, not the barrel | | USD/CNY (onshore) | 6.74615 | −0.21% | −3.30% YTD; offshore CNH was not retrieved on a verified basis this session | | USD/KRW | 1,422.07 | −1.44% | −3.63% on the week and −8.30% on the month — the won has strengthened through a −10.84%, a −6.0% and a −1.23% Kospi |
Take — read the crosses that did not behave. The headline is a broad 0.89% dollar decline that pushed the DXY below 100 for the first time this episode, and the naive story is “risk-on, sell the dollar.” The crosses say something more specific. First, USD/CHF −1.06% contradicts the risk-on framing: the Swiss franc, the purest funding-and-haven currency in the majors, appreciated more than the euro and the pound on a day the Nasdaq 100 rose 3.36% and the VIX fell 17.28%. A haven that bids into an equity melt-up is not tracking equity risk — it is tracking dollar credibility, the same variable the 30-year is trading (§6). Second, the yen’s 2.30% move is policy, not flow. With the Nikkei reporting confirmed MOF intervention plus a U.S. rate check, the yen is the one major whose move carries no information about risk appetite — and the fact that it moved with a 3.36% Nasdaq rally is a useful negative result: the classic carry pairing (yen up / risk down) did not hold, which argues the AI drawdown was a leverage-and-positioning event inside the equity market rather than a global funding unwind. JPMorgan’s deleveraging note and the Situational Awareness liquidation both corroborate. Third, USD/KRW −1.44% completes a three-session pattern: the won has strengthened 3.63% on the week through two circuit breakers. Foreign capital is not fleeing Korea; domestic leverage is unwinding into a currency that is being bought, and the SK Hynix ADR’s +17.52% with a UBS Buy initiation at $204 is the offshore market voting on the other side of that domestic deleveraging. |
Quote basis: TradingEconomics live board, Jul/30 stamps for EUR, GBP, AUD, CHF, CAD, CNY and KRW; DXY and USD/JPY carried a 06:13 clock stamp (Asia Friday morning) rather than a Jul/30 date stamp and are therefore marginally later than the U.S. close — cross-checked against Bloomberg’s 4 p.m. New York marks, which agree to within 0.25 yen, 0.0005 on the euro and 0.0007 on sterling.
| Contract | Last | 1-Day | 1-Week | 1-Month | YTD | Driver | | WTI Crude (front) | $84.230 | −0.27% | −8.63% | +22.82% | +46.69% | Hormuz transits picked up despite hostilities (Bloomberg); Bloomberg’s 4 p.m. mark was −0.9% at $83.67 | | Brent Crude (front) | $89.364 | −1.52% | −11.25% | +24.86% | +46.86% | Back below $90 after Wednesday’s +7.45% spike; the week’s round trip is the largest of the year | | Natural Gas | $2.7438 | +0.80% | −5.91% | −14.79% | −25.56% | The only energy contract with a negative YTD — the AI-power bid has not shown up in the gas curve | | Gasoline (RBOB) | $3.2506 | −4.33% | −7.03% | +10.37% | +89.99% | Largest single-day decline on the board; the level is still the inflation problem (§8) | | Heating Oil | $4.1968 | −3.97% | −3.34% | +30.42% | +97.82% | Distillate remains the tightest barrel; Bloomberg: “US Refiners See Billions in Profits From Global Fuel Crunch” | | Gold (spot) | $4,105.47 | +0.97% | +1.38% | +1.84% | −4.96% (see caveat) | Bloomberg spot 4 p.m. $4,116.73 (+1.2%); Bloomberg board Dec futures $4,160.60 (+1.55%) | | Silver (spot) | $59.036 | +2.47% | +2.43% | −0.08% | −17.15% (see caveat) | Outperformed gold by 150 bp — the industrial leg, not the haven leg | | Copper (spot) | $6.4734/lb | +3.10% | +2.67% | +5.71% | +13.93% | Best base-metal print on the board; +47.01% y/y |
Take — a growth-and-electrification bid inside an energy unwind. The two-day sequence is the story: Wednesday, Brent +7.45% and WTI +6.9% on Iran’s missile attack, Trump’s retaliation pledge and a 7.2 mb EIA crude draw; Thursday, Brent −1.52% and WTI −0.27% to −0.9% on nothing more than the observation that shipping across the Strait of Hormuz has picked up despite hostilities (Bloomberg). A war premium that evaporates on a shipping-traffic observation was a positioning premium, and the weekly numbers confirm how much of it there was: Brent −11.25% and WTI −8.63% on the week even after Wednesday’s spike. Barclays’ Emmanuel Cau flagged rising oil short positions and muted TIPS inflows on Wednesday morning as evidence investors were sanguine on inflation risk; Thursday those shorts were right. Meanwhile copper +3.10% and silver +2.47% against gold +0.97% is a textbook industrial-over-haven rotation lining up precisely with the SOX +8.19% and Technology +5.55% — the metals complex voted for the AI capex cycle on the same day the equity market did, and gold’s 150-bp underperformance of silver mirrors its failure to bid on Wednesday’s −1.5% S&P day. |
Basis caveats, stated explicitly. (i) Energy rows are front-month futures; precious metals and copper are spot, so they are not directly comparable to a futures settlement print. (ii) The WTI figure carries a genuine vendor divergence — TradingEconomics $84.230 (−0.27%) versus Bloomberg’s 4 p.m. $83.67 (−0.9%) and the Bloomberg /markets board’s $83.59, a spread of roughly 64 cents consistent with different snapshot instants around the 2:30 p.m. ET NYMEX settlement; all are shown rather than one being suppressed. (iii) The TradingEconomics YTD column for gold (−4.96%) and silver (−17.15%) cannot be reconciled with the same board’s 1-week, 1-month and 1-year columns (gold +1.38% / +1.84% / +24.76%; silver +2.43% / −0.08% / +60.77%) and appears anchored to an early-January high on a different basis — those two YTD figures should not be traded on, and they are printed only to keep the table on one vendor with the inconsistency disclosed rather than silently dropped. The column mapping (Price | Chg | %Chg | Weekly | Monthly | YTD | YoY | Date) was verified programmatically before any figure was quoted, and every row’s Chg/%Chg pair was checked for internal consistency.
| 11 · Trading Views (desk-style; not personalized investment advice) |
Claude is not a licensed financial advisor. These are illustrative desk-style expressions of the analysis above, not recommendations. Verify independently and size to your own mandate.
1. Long NVDA / short the squeeze basket (SNDK, IREN, NBIS, CRWV), dollar-neutral. Rationale: Nvidia rose 2.65% while the SOX rose 8.19% and the four names above rose 21–31% — a 19-to-28-point single-session dispersion driven by a named, dated, completed liquidation (Situational Awareness → Citadel). Forced-cover dispersion mean-reverts once the basket is placed; fundamental dispersion does not. Catalyst: confirmation the Citadel purchase is fully absorbed; AMD Tue 8/4 4:15 PM and SanDisk/Western Digital Wed 8/5 as the first fundamental checkpoints. Invalidation: the squeeze names hold their gains through both prints, or JPMorgan’s “deleveraging is done” call is followed by a second fund unwind that re-runs the same dispersion. Sizing: dollar-neutral rather than beta-neutral — the short leg’s beta is unstable in a squeeze. |
2. 2s30s (or 3M30Y) steepener, carried into Friday 8:30 a.m. ECI. Rationale: §6 and §8. In one week the front end removed roughly two hikes (CME cumulative September 82.3% → 63.4%; December ≥+50 bp 59.7% → 40.9%) and the 30-year still rose 4 bp to a 19-year closing high of 5.21%. On Thursday a GDP miss, a negative headline PCE and a 17% VIX collapse moved the whole curve exactly 1 bp. Catalyst: ECI Friday 8:30, then ISM prices-paid Mon/Wed and payrolls Fri 8/7. Invalidation: restored explicit FOMC forward guidance, or a Treasury refunding announcement that cuts long-end coupon issuance. Note the asymmetry: both a hot and a soft ECI are constructive for the position; only a re-anchoring of the long end is not. |
3. Long MSFT / short META, beta-neutral. Rationale: the market applied an explicit, twice-tested rule — convert the capex and you get paid ($450bn in one day, the largest ever); commit the capex and you get charged for it (Meta −7.98% after disclosing ~$700bn of committed future AI spending). The pair moved 23.5 points in one session and the rule has not been contradicted by any print this quarter. Catalyst: Azure/AWS disclosure cadence; Meta’s Q3 revenue delivery against the $61–64bn guide. Invalidation: Meta demonstrates revenue conversion on the committed spend, or Microsoft’s FY2027 capex step-up (~$175bn implied) starts being priced as a liability rather than an asset — the same rule cuts both ways, and Microsoft is now the crowded side. |
4. Fade the defensives-funding trade: long Consumer Defensive / Healthcare vs. short S&P 500, small. Rationale: Consumer Defensive −1.89%, Healthcare −1.15% and Real Estate −1.16% on a +1.66% index day is a funding rotation, not a fundamental de-rating — those groups were sold to buy the squeeze, and Consumer Defensive is still +3.19% on the week. Catalyst: the squeeze exhausting; ECI and payrolls re-introducing macro as the driver. Invalidation: a genuine growth re-acceleration (ISM Manufacturing well above 50 with new orders) that justifies cyclical leadership on fundamentals rather than flow. |
5. Long ICE / long CBOE and TW as the fixed-income market-structure read-across. Rationale: ICE is paying $167 cash per MarketAxess share, a 33% premium, ~$5.7–6.0bn EV, closing H1 2027, explicitly to own the electronic institutional bond network (~2,100 institutions across 90+ countries). This is a strategic re-rating of every electronic fixed-income venue in a year when the bond market is delivering 19-year-high yields and record volumes. Cboe reports Fri 7/31 7:30 a.m.; Tradeweb reported 7/30 7:00 a.m. Catalyst: regulatory commentary on the deal; venue volume disclosures. Invalidation: an antitrust challenge that reprices consolidation optionality across the group. |
Vol note. VIX 17.09, −3.57, −17.28%, intraday range 17.00–20.08 — the largest single-day decline of this episode and a full 3.5-point round trip from Wednesday’s 20-handle close. At 17 handles the index implies roughly a 1.1% daily move heading into an ECI print, a full jobs week, an AMD report and a long bond at a 19-year high. Rates vol has been the correctly-priced asset for three weeks; equity vol is now demonstrably the cheap leg, and the cheapest expression of everything above is simply owning August S&P optionality rather than selling it into the squeeze. |
| 12 · S&P 500 Earnings Calendar — Current & Next Week (S&P 500 components only) |
Times are ET. Every day from Friday 7/31 through Friday 8/7 was re-pulled from the Earnings Whispers day pages this session (/1 = before open, /2 = after close), so nothing in the forward calendar is carried stale. Non-S&P-500 names are excluded, as are names whose index membership could not be verified conservatively (listed in the companion US_CrossAsset_Daily_2026-07-30_DataNotes.txt). Re-verify times and membership against company IR before trading any date.
| Day | Reporters | Mon 7/27
— completed | BMO: Baker Hughes (BKR).
AMC: Cadence Design Systems (CDNS) 4:00 — beat: adj. $2.11 vs. $2.05; +4% after hours, Welltower (WELL) 4:05, Cincinnati Financial (CINF) 4:05, F5 (FFIV) 4:05, Universal Health Services (UHS) 4:15 — FY guidance cut; −4%+, Principal Financial (PFG) 4:15, Sun Communities (SUI) 4:15, UDR (UDR) 4:15, Nucor (NUE) 4:30, Brown & Brown (BRO) 5:00. | Tue 7/28
— completed | BMO: Carrier (CARR) 6:00, Centene (CNC) 6:00, UPS (UPS) 6:00, Hilton (HLT) 6:00, TransUnion (TRU) 6:25, HF Sinclair (DINO) 6:30, Royal Caribbean (RCL) 6:30, CMS Energy (CMS) 6:30, Textron (TXT) 6:30, Pentair (PNR) 6:50, Coca-Cola (KO) 6:55 — closed −0.63% Thursday, Invesco (IVZ) 6:55, Xylem (XYL) 6:55, American Tower (AMT) 7:00, Corning (GLW) 7:00, Incyte (INCY) 7:00, IQVIA (IQV) 7:00, PayPal (PYPL) 7:00, Sherwin-Williams (SHW) 7:00 — closed +0.28% Thursday, DTE Energy (DTE) 7:15, S&P Global (SPGI) 7:15, Boeing (BA) 7:30 — +3.22% Thursday, Hubbell (HUBB) 7:30, Ecolab (ECL) 8:00, Illinois Tool Works (ITW) 8:00, PACCAR (PCAR) 8:00.
AMC: Caesars (CZR) 4:00, Arch Capital (ACGL) 4:00, Expand Energy (EXE) 4:00, Skyworks (SWKS) 4:00, Qorvo (QRVO) 4:00, Omnicom (OMC) 4:00, Visa (V) 4:05 — closed −0.67% Thursday, KLA (KLAC) 4:05, Ford (F) 4:05, Enphase (ENPH) 4:05, Seagate (STX) 4:05, Mondelez (MDLZ) 4:05, Boston Properties (BXP) 4:05, CoStar (CSGP) 4:05, Manhattan Associates (MANH) 4:05, Landstar (LSTR) 4:05, PPG (PPG) 4:05, W.P. Carey (WPC) 4:05, Extra Space (EXR) 4:10, NXP Semiconductors (NXPI) 4:10, Unum (UNM) 4:15, FirstEnergy (FE) 4:20, Teradyne (TER) 4:30, Waste Management (WM) 4:30, Veralto (VLTO) 4:30. | Wed 7/29
— FOMC day; completed | BMO: Vertiv (VRT) 5:55, Generac (GNRC) 6:00, Humana (HUM) 6:00, Stanley Black & Decker (SWK) 6:00, Bunge (BG) 6:00, Avantor (AVTR) 6:05, GE HealthCare (GEHC) 6:20, Boston Scientific (BSX) 6:30, Cognizant (CTSH) 6:30, Aon (AON) 6:30, Entergy (ETR) 6:30, Smurfit Westrock (SW) 6:30, Lennox (LII) 6:45, CBRE (CBRE) 6:55, Johnson Controls (JCI) 6:55, ADP (ADP) 7:00, General Dynamics (GD) 7:00, Procter & Gamble (PG) 7:00 — closed −1.47% Thursday, Vulcan Materials (VMC) 7:00, Garmin (GRMN) 7:00, IDEX (IEX) 7:00, Masco (MAS) 7:00, Old Dominion (ODFL) 7:00, WEC Energy (WEC) 7:00, Verisk (VRSK) 7:15, Fortive (FTV) 7:30, Clean Harbors (CLH) 7:30, Watsco (WSO) 7:30, Amphenol (APH) 8:00.
AMC: Meta Platforms (META) 4:00 — −7.98% to $538.88 Thursday; Thursday disclosure: ~$700bn of committed future AI/data-centre spending, Qualcomm (QCOM) 4:00 — −2.62% to $151.60, PTC (PTC) 4:00, Sprouts Farmers Market (SFM) 4:00, Align Technology (ALGN) 4:05, Equinix (EQIX) 4:05, Fortinet (FTNT) 4:05 — +0.62% to $154.17 after trading to $169.26 — the day’s largest fade, Starbucks (SBUX) 4:05 — +1.64%, Lam Research (LRCX) 4:05 — +17.98% to $297.72, L3Harris (LHX) 4:05, Robinhood (HOOD) 4:05 — −3.60%, Electronic Arts (EA) 4:05, Carvana (CVNA) 4:05, C.H. Robinson (CHRW) 4:05, Microsoft (MSFT) 4:10 — +15.51% to $451.10, ~$450bn added — the largest one-day market-cap gain by any U.S. company on record, MGM Resorts (MGM) 4:15, Tyler Technologies (TYL) 4:15, Service Corp (SCI) 4:15, VICI Properties (VICI) 4:15, Invitation Homes (INVH) 4:15, Fair Isaac (FICO) 4:15, O’Reilly (ORLY) 4:30, FMC (FMC) 4:30, American Water Works (AWK) 4:30, Chipotle (CMG) 4:35. | Thu 7/30
— completed | BMO: Cigna (CI) 6:00, Air Products (APD) 6:00, Willis Towers Watson (WTW) 6:00, Xcel Energy (XEL) 6:05, Valero (VLO) 6:30, Trane (TT) 6:30, Norwegian Cruise (NCLH) 6:30 — −7%+; beat but cut guidance; CEO John Chidsey: problems are “self-inflicted”, Regeneron (REGN) 6:30, Hershey (HSY) 6:45, Avery Dennison (AVY) 6:45, KKR (KKR) 6:50, Labcorp (LH) 6:50, Exelon (EXC) 6:50, Bristol Myers Squibb (BMY) 6:55, Quanta Services (PWR) 6:55, A.O. Smith (AOS) 6:55, Martin Marietta (MLM) 6:55, Sirius XM (SIRI) 6:55, American Electric Power (AEP) 6:55, Altria (MO) 7:00, International Paper (IP) 7:00, LKQ (LKQ) 7:00, Builders FirstSource (BLDR) 7:00, Blue Owl (OWL) 7:00, Tradeweb (TW) 7:00, Yum! Brands (YUM) 7:00 — +1%+; adj. $1.62 vs. $1.58; no update on the Taco Bell cyclospora outbreak, Baxter (BAX) 7:15, Huntington Ingalls (HII) 7:15, ICE (ICE) 7:30 — announced the $167/share cash acquisition of MarketAxess, a 33% premium, ~$5.7–6.0bn EV, closing H1 2027, Jones Lang LaSalle (JLL) 7:30, EMCOR (EME) 7:30, Southern Company (SO) 7:30, Mastercard (MA) 8:00 — +2.49%.
AMC: Amazon (AMZN) 4:00 — Q2 net sales $200.6bn (+20%) vs. $196.47bn; AWS $42.2bn (+37%, fastest in 18 quarters) vs. $40.54bn; AWS operating income $16.6bn at a 39.4% margin; total operating income $27.5bn (+43%); Jassy guides 2026 capex to $220bn — shares higher after hours, DexCom (DXCM) 4:00, First Solar (FSLR) 4:00, Monolithic Power (MPWR) 4:00, Neurocrine (NBIX) 4:00, GoDaddy (GDDY) 4:05, Illumina (ILMN) 4:05, Stryker (SYK) 4:05, Western Union (WU) 4:05, Edison International (EIX) 4:05, LPL Financial (LPLA) 4:05, Live Nation (LYV) 4:10, Ingersoll Rand (IR) 4:10, Gallagher (AJG) 4:15, Camden Property (CPT) 4:15, Eversource (ES) 4:15, Erie Indemnity (ERIE) 4:15, Gaming & Leisure (GLPI) 4:15, Mohawk (MHK) 4:15, CubeSmart (CUBE) 4:15, Eastman Chemical (EMN) 4:15, Weyerhaeuser (WY) 4:15, Coinbase (COIN) 4:15 — posted a loss as revenue fell for another quarter, Corteva (CTVA) 4:30, Mettler-Toledo (MTD) 4:30, Ameren (AEE) 4:30, Apple (AAPL) 4:30 — FQ3 EPS $2.02 vs. $1.89 on $109.4bn vs. $108.8bn; iPhone $54.2bn vs. $53.5bn; but Services $30.7bn vs. $31.3bn and Greater China $18.8bn vs. $19.6bn — shares −3% to −4% after hours after closing −1.41%, Alliant Energy (LNT) 6:00. | Fri 7/31
— re-verified | BMO: Linde (LIN) 5:30, Ares Management (ARES) 6:00, Chevron (CVX) 6:15, Exxon Mobil (XOM) 6:30, Eaton (ETN) 6:30, LyondellBasell (LYB) 6:30, Colgate-Palmolive (CL) 6:55, Church & Dwight (CHD) 6:55, T. Rowe Price (TROW) 7:00, Dominion Energy (D) 7:30, Federal Realty (FRT) 7:30, Cboe (CBOE) 7:30, AbbVie (ABBV) 7:45, Franklin Resources (BEN) 8:20. All fourteen names and times confirmed unchanged for a second consecutive session.
AMC: the reviewed Earnings Whispers after-close page for Friday 7/31 again returned “NONE” — no after-close reporters are published for that date. Confirm directly with company IR if you expect one. |
| Day | Reporters | Mon 8/3
— re-verified | BMO: Loews (L) 6:00, Marriott (MAR) 7:00, Tyson Foods (TSN) 7:30. Confirmed unchanged.
AMC: SBA Communications (SBAC) 4:00, Vertex Pharmaceuticals (VRTX) 4:00, Diamondback Energy (FANG) 4:00, Palantir (PLTR) 4:05, Alexandria Real Estate (ARE) 4:10, Clorox (CLX) 4:10, ONEOK (OKE) 4:15, Williams (WMB) 4:15. Confirmed unchanged; ON Semiconductor (ON) remains off this page for a second session. | Tue 8/4
— re-verified | BMO: Archer-Daniels-Midland (ADM) 6:00, Leidos (LDOS) 6:00, DuPont (DD) 6:00, Ball (BALL) 6:00, Gartner (IT) 6:00, Revvity (RVTY) 6:00, Waters (WAT) 6:00, Henry Schein (HSIC) 6:00, Caterpillar (CAT) 6:30, Merck (MRK) 6:30, Kimberly-Clark (KMB) 6:30, Zebra (ZBRA) 6:30, Apollo Global (APO) 6:30, IDEXX (IDXX) 6:30, Pfizer (PFE) 6:45, Aptiv (APTV) 6:45, Marathon Petroleum (MPC) 6:45, Kimco (KIM) 6:50, AMETEK (AME) 6:55, McDonald’s (MCD) 7:00, Entegris (ENTG) 7:00, Broadridge (BR) 7:00, Duke Energy (DUK) 7:00, NRG Energy (NRG) 7:00, Rockwell Automation (ROK) 7:00, TransDigm (TDG) 7:15, FIS (FIS) 7:30, Energy Transfer (ET) 7:30, Cummins (CMI) 7:30, PSEG (PEG) 7:30, Sysco (SYY) 8:00, W.W. Grainger (GWW) 8:00, Progressive (PGR) 8:15, Expeditors (EXPD) 8:30, Pinnacle West (PNW) 8:35. All thirty-five confirmed unchanged. CAT at 6:30 a.m. remains the single most important print of next week after Baird’s data-centre-permitting downgrade.
AMC: Booking Holdings (BKNG) 4:00, Amgen (AMGN) 4:00, Wynn Resorts (WYNN) 4:00, Gilead (GILD) 4:00, Arista Networks (ANET) 4:05, DaVita (DVA) 4:05, Devon Energy (DVN) 4:05, Pinterest (PINS) 4:05, Emerson Electric (EMR) 4:05, Fortune Brands (FBIN) 4:05, Match Group (MTCH) 4:10, Jacobs (J) 4:10, AMD (AMD) 4:15, Mosaic (MOS) 4:15, IFF (IFF) 4:15, Celanese (CE) 4:15, Equitable Holdings (EQH) 4:15, Healthpeak (DOC) 4:15, Prudential Financial (PRU) 4:20, Assurant (AIZ) — time listed only as “After Close.” | Wed 8/5
— re-verified | BMO: Owens Corning (OC) 6:00, CVS Health (CVS) 6:30, BorgWarner (BWA) 6:30, Cencora (COR) 6:30, Zimmer Biomet (ZBH) 6:30, NiSource (NI) 6:30, United Therapeutics (UTHR) 6:30, Iron Mountain (IRM) 6:45, Eli Lilly (LLY) 6:45, Uber (UBER) 6:55, Global Payments (GPN) 6:55, Insulet (PODD) 7:00, CDW (CDW) 7:00, Phillips 66 (PSX) 7:00, Kraft Heinz (KHC) 7:00, Charles River Labs (CRL) 7:00, EOG Resources (EOG) 1:25 AM, Walt Disney (DIS) — time listed only as “Before Open.”
AMC: Western Digital (WDC) 4:00, Axon (AXON) 4:00, Expedia (EXPE) 4:00, SanDisk (SNDK) 4:05, AppLovin (APP) 4:05, Block (XYZ) 4:05, Corpay (CPAY) 4:05, DoorDash (DASH) 4:05, eBay (EBAY) 4:05, Paycom (PAYC) 4:05, Albemarle (ALB) 4:15, Fidelity National Financial (FNF) 4:15, Occidental (OXY) 4:15, Texas Pacific Land (TPL) 4:15, CF Industries (CF) 4:30, Host Hotels (HST) 4:30, Atmos Energy (ATO) 4:35, Allstate (ALL) 5:10. | Thu 8/6
— re-verified | BMO: EPAM (EPAM) 6:00, Targa Resources (TRGP) 6:00, Becton Dickinson (BDX) 6:30, Molson Coors (TAP) 6:30, Kenvue (KVUE) 6:30, Viatris (VTRS) 6:55, ConocoPhillips (COP) 7:00, Datadog (DDOG) 7:00, Fiserv (FISV) 7:00, Keurig Dr Pepper (KDP) 7:00, Howmet (HWM) 7:00, Evergy (EVRG) 7:00, Constellation Energy (CEG) 7:05, Cheniere (LNG) 7:30, APA (APA) 8:00, Fox Corporation (FOXA) “Before Open”; Fox Class B (FOX) 8:00.
AMC: Airbnb (ABNB) 4:00, The Trade Desk (TTD) 4:00, Akamai (AKAM) 4:00, Texas Roadhouse (TXRH) 4:00, Aflac (AFL) 4:05, ResMed (RMD) 4:05, Gen Digital (GEN) 4:05, Monster Beverage (MNST) 4:10, Republic Services (RSG) 4:10, DraftKings (DKNG) 4:15, AIG (AIG) 4:15, Reinsurance Group (RGA) 4:15, Consolidated Edison (ED) 4:30. | Fri 8/7
— re-verified | BMO: MarketAxess (MKTX) 6:30 — now reporting as an announced acquisition target, Take-Two Interactive (TTWO) 7:00, PPL (PPL) 7:30. Confirmed unchanged.
AMC: the reviewed after-close page for Friday 8/7 lists only non-S&P-500 names (Hawaiian Electric, Galaxy Gaming) — there are no S&P 500 after-close reporters published for that date. This corrects the prior report, which stated the calendar had not published an after-close bucket at all; it has, and it contains no index members. |
Changes vs. the prior calendar (7/29 report). Addition to Wed 8/5 BMO: Walt Disney (DIS), listed as “Before Open” with no specific time — a Dow and S&P 500 constituent absent from the prior pull, and at −2.38% Thursday one of the day’s weaker large caps. • Addition to Tue 8/4 AMC: Assurant (AIZ), listed as “After Close” with no specific time. • Addition to Wed 8/5 AMC: Paycom (PAYC) 4:05. • Addition to Thu 8/6 AMC: Monster Beverage (MNST) 4:10. • Timing change on Thu 8/6 BMO: the prior report carried “Fox (FOX/FOXA) 8:00”; the page now splits them — FOXA “Before Open” and FOX 8:00. • Correction to Fri 8/7 AMC: the after-close page does publish, but lists no S&P 500 members. • No removals detected on any of the six days re-pulled. Fri 7/31 AMC remains genuinely empty (“NONE”), not missing, and ON Semiconductor (ON) remains absent from Mon 8/3 AMC for a second session. • Aug 4–7, previously carried from a one-day-stale pull, were fully re-pulled this session and are no longer flagged as carried. • Dominion Energy (D) again appeared on both the Fri 7/31 BMO page and the Mon 8/3 AMC page; the 7/31 BMO slot is shown and the conflict is flagged in the companion US_CrossAsset_Daily_2026-07-30_DataNotes.txt, along with the full list of names excluded for unverified S&P 500 membership. |
| 13 · Risk Map (next 5 sessions) |
1. The equity market repriced positioning and did not reprice the discount rate — that gap is the dominant risk. On a day with a +3.36% Nasdaq 100, a −17.28% VIX, a GDP miss and a negative headline PCE, the 30-year rose to a fresh 19-year closing high of 5.21% and the 20-year to 5.22%, with 2s30s unchanged at +98 bp after steepening 18 bp on the week. A long end that will not rally on soft data is a long end pricing credibility and supply, not the cycle. Watch level: 30Y above 5.35%, at which point the multiple reprices regardless of the flow story. |
2. The squeeze is identifiable, dated and therefore exhaustible. Thursday’s leadership was SNDK +25.99%, IREN +30.54%, NBIS +27.13%, BE +26.42%, CRWV +21.51%, MU +18.36%, LRCX +17.98% against NVDA +2.65% — a 19-to-28-point dispersion against the most liquid name in the complex. The named driver is the Situational Awareness forced liquidation, with Citadel buying the portfolio and BofA, Goldman and JPMorgan marketing the baskets. Once the basket is placed the flow stops. JPMorgan says the deleveraging is essentially complete; Bloomberg’s own strategist says the rebound “looks more like a positioning squeeze than the start of a durable risk-on move.” Both can be true, and if they are, the next 5% is decided by AMD (Tue) and WDC/SNDK (Wed), not by flow. |
3. Friday 8:30 a.m. ECI is the last inflation-adjacent print before August, into a front end that has just removed two hikes. CME cumulative September hike odds have fallen 82.3% → 63.4% in a week and the +50 bp bucket is at 0.0% from 25.0%. That is a lot of removed optionality to defend with one wage print. A hot ECI does not restore the +50 bp bucket — it steepens the curve further, which is the worst outcome for long-duration equities and, on Wednesday’s evidence, for banks. |
4. Apple’s after-hours miss on Services and China is a Friday risk the tape has not yet traded. Services $30.7bn vs. $31.3bn and Greater China $18.8bn vs. $19.6bn, with the stock −3% to −4% after hours on top of a −1.41% session. Apple is the largest weight that rents rather than builds AI infrastructure — the natural beneficiary of the capex rule the market has applied all week — and it still missed on the two lines that define its terminal multiple. Against that, Amazon delivered the cleanest capex-conversion print of the season (AWS +37%, 39.4% margin, $220bn 2026 capex guide) and traded higher. Friday opens with those two pulling in opposite directions on roughly 12% of index weight. |
5. The war has not ended; the war premium has. Brent fell 1.52% to $89.36 and is −11.25% on the week on an observation about Hormuz shipping traffic picking up despite hostilities, while Egypt confirmed a drone strike on two ships — a regasification vessel and a storage vessel — at Damietta, attacks on energy infrastructure widened, and the Caspian Pipeline Consortium suspended Black Sea loadings after two associated tankers were attacked overnight. A price that falls on unchanged physical risk is a price with no risk premium left in it. Watch Brent $96 on the upside and $84 on the downside. |
6. Crowded consensuses to stress-test, with the numbers. (a) “The chip bottom is in” — possibly, but the evidence is a flow call plus a squeeze in the most-deleveraged names, and the SOX is still −21% month-to-date, the worst month since 2008, and 22.9% below its 52-week high. (b) “Microsoft proves AI capex is fine” — Microsoft was rewarded for converting spend; it simultaneously guided FY2027 capex higher and extended building useful lives from 15 to 25 years, and Meta was punished for disclosing ~$700bn of committed future spend. The rule that made MSFT the largest one-day gain ever is the rule that will be applied to MSFT next quarter. (c) “Soft data means the Fed is done” — 0.0% of any cut is priced at any 2026 meeting, in every column, on both vendors. (d) “Risk-on, sell the dollar” — the Swiss franc gained 1.06%, more than the euro or the pound, on a +1.66% S&P day; that is not a risk-appetite trade. (e) Structural watch item: Fitch reported its private-credit default rate hit a record in the second quarter (Bloomberg), and S&P stripped Swarthmore of its AAA over debt load — two small, unglamorous credit datapoints in a week when the 30-year made a 19-year high. |
7. What VIX is and is not pricing. 17.09, −3.57, −17.28%, range 17.00–20.08 — the largest one-day decline of this episode and a full 3.5-point round trip from Wednesday’s 20-handle close. At 17 the index implies roughly a 1.1% daily move. It is pricing the end of the deleveraging. It is not pricing an ECI print in 14 hours, a full jobs week, an AMD report, an Apple miss on Services and China, and a long bond at a 19-year high with 2s30s at +98 bp. Rates vol remains the correctly-priced asset; equity vol has gone from cheap to cheaper. |
Source Links and Data Notes & Conflicts are in the companion text file US_CrossAsset_Daily_2026-07-30_DataNotes.txt.
|