Oil Reprices the Consumer, Rates Reject Relief

2. Complete macro-event table

Rank	Event	Exact time	Actual vs consensus / prior	Surprise	Immediate or session reaction	Why it mattered
1	Iran economic-pressure threat / Hormuz risk	Late Aug. 19 ET; priced Aug. 20	Trump threatened the “most crushing” economic action but gave no implementation detail; commercial normalization remained unverified	Renewed escalation rhetoric	Brent $93.78, +2.4%; 10Y roughly +5 bp to 4.70%; airlines/cruises fell 2.5%–3.5%	Energy inflation, freight/insurance, rates and consumer real income
2	Treasury long-end intervention follow-through	CNBC morning interview; cash close	Bessent said buybacks could exceed $4bn per operation; 10Y/30Y returned near pre-announcement levels	Signal strengthened; market effect faded	10Y ~4.70% from 4.65%; 30Y ~5.23%–5.25%; S&P −0.9%	Tested Treasury's capacity to lean against fiscal/term-premium pressure
3	Philadelphia Fed August manufacturing survey	8:30 a.m. ET	Headline 47.4 vs 25.0, 41.4 prior; employment 27.9 vs 10.0; future activity 73.6 vs 34.4; prices paid 40.9 vs 53.9	Large upside growth/employment surprise; mixed inflation	Combined with claims, pushed long yields higher and equities lower; isolated tick contaminated	Timely factory, hiring, capex and price-setting signal
4	U.S. initial jobless claims	8:30 a.m. ET	206k vs 210k, prior revised 212k from 209k; continuing 1.799m vs 1.790m, prior 1.781m	Fewer layoffs; continuing claims slightly worse	Growth-positive/rates-negative response, inseparable from Philly Fed	Challenged labor-slowdown and near-term easing narratives
5	$8bn 30-year TIPS reopening	Bids 1:00 p.m. ET; result ~1:03 p.m.	2.973% real yield; 2.82 cover; indirect/direct/dealer shares 84.5%/13.5%/2.1% of competitive awards	Exceptionally strong end-user mix at a very high real yield	Long-end nominal weakness persisted; auction itself absorbed cleanly	Test of inflation-protected duration demand amid fiscal pressure
6	Conference Board July U.S. LEI	10:00 a.m. ET	+0.2% vs +0.1%, prior revised −0.1% from −0.2%; six-month growth +0.2%	Modest upside; first positive six-month rate in >4 years	No clean standalone reaction amid oil/rates/Walmart	Soft-landing/AI-investment signal versus weak consumer expectations

Primary cross-checks: DOL claims release, Philadelphia Fed August survey PDF, Conference Board LEI, TreasuryDirect auction data, and AP market wrap.

3. Detailed macro events with opinion clusters

1. Iran economic-pressure threat and the Hormuz oil shock

Impact: very high. Fact pattern. President Trump posted at about 6:59 p.m. ET on August 19 that the United States would mount the “most crushing economic operation” against Iran and punish countries supplying a lifeline. The post named oil smuggling, swap lines, cash transfers, exchange houses, registries, and front companies, but contained neither operative instruments nor an effective date; Treasury Secretary Scott Bessent said detail would come Monday. (Trump post; Reuters, Aug. 20; AP, Aug. 20)

Physical and market evidence. Kpler counted only nine commodity-vessel transits Wednesday, unchanged day over day and far below roughly 45 daily crossings during the June–July truce; AIS-dark movements are excluded. U.S. officials separately claimed a protected Omani-side corridor was moving 15–20 tankers nightly and nearly 10 million barrels a day. Those measures cannot be reconciled cleanly—verified crossings versus estimated barrels including quiet traffic—so the conflict is a material data gap, not a fact to average away. (Reuters/Kpler, Aug. 20; Axios, Aug. 19) Brent settled 2.4% higher at $93.78; at 1:49 p.m. ET, expiring September WTI was up 2.5% at $88.01 and October WTI 2.7% at $86.68. The 10-year yield rose roughly 5 basis points and equities fell, although strong U.S. data, debt supply, and Walmart were material co-drivers. (Reuters oil report, Aug. 20; AP close)

Cluster A — “Prolonged physical deficit; the premium remains too low” (high confidence). Giovanni Staunovo of UBS, Ewa Manthey and Warren Patterson of ING, Mohit Kumar of Jefferies, Angeline Ong of IG, and Deutsche Bank shared the view that diplomatic deadlock, tighter enforcement, and depressed exports prolong the constraint. The support is falling observable traffic, renewed sanctions pressure, and neither side reaching a political pain threshold. The days-to-weeks implication is an upward crude bias and higher inflation/term-premium pressure on duration. The cluster fails if protected traffic normalizes, talks resume, SPR releases grow, or demand destruction accelerates. (Reuters; ING, Aug. 20; Guardian, Aug. 18)

Cluster B — “China limits sanctions efficacy; coercion will be slow” (medium-high confidence). Gregory Brew of Eurasia Group, former Israeli military-intelligence Iran chief Danny Citrinowicz, and Alex Hodes of StoneX argued that meaningful secondary sanctions must reach Iran's partners—especially China—raising retaliation risk without ensuring quick Iranian capitulation. The implication is a weeks-to-months geopolitical premium and more U.S.–China friction, not a clean supply forecast. Broad allied compliance, falling Iranian exports/rial, or an Iranian return to talks would disconfirm it. (Guardian, Aug. 20; Reuters)

Cluster C — “Crude may cap; downstream stress can worsen” (medium confidence). Ole Hansen of Saxo Bank and Arslan Ali of FXEmpire emphasized a 4.4-million-barrel U.S. inventory build and softer demand estimates as crude constraints, while diesel cracks, freight, refiners, airlines, and importers absorb the sharper pain. A new tanker or energy-facility attack is the principal disconfirming risk. (Hansen/source summary; FXEmpire, Aug. 20)

Consensus, disagreement, resolver. Attributed consensus is that the physical constraint is real and détente odds fell. The sharp disagreement is whether $94 Brent materially underprices a prolonged chokepoint shock or already faces an inventory/demand ceiling while stress migrates downstream. Report inference: Thursday repriced policy duration risk; it did not prove incremental Iranian barrels had yet been removed. Monday's sanctions mechanics, daily verified and dark-traffic estimates, China compliance, diesel cracks, and Iranian negotiating behavior resolve the debate.

2. Treasury's larger long-end buyback signal met its market limit

Impact: high. Fact pattern. In a live CNBC interview shortly before an 11:08 a.m. ET report, Bessent said Treasury could buy back more than $4 billion per issue, called 30-year liquidity “very poor,” and described the action partly as a signal that yields did not reflect fundamentals. The underlying August 19 change at least doubled maximum liquidity-support buybacks in 10–20-year and 20–30-year nominals from $2 billion, effective September 9 through November 4. This is debt management financed by Treasury, not Fed reserve creation or QE. (CNBC, Aug. 20; Treasury, Aug. 19) Official gross debt stood at $40.047 trillion on August 18, including $32.266 trillion held by the public. (Treasury FiscalData)

Reaction. Wednesday's announcement initially lowered the 30-year yield roughly 9–10 basis points to about 5.19% and the 10-year 5–7 basis points to about 4.65%. Thursday's verbal escalation produced only a momentary dip: CNBC recorded roughly 5.235% on the 30-year and 4.704% on the 10-year, while AP later put them near 5.23% and 4.69%. DXY was about 98.80 and little changed around the remarks. Oil, data, Walmart, and corporate issuance complicate attribution, but the retracement is the cleanest realized evidence that signaling alone did not reset the long-end regime. (Reuters, Aug. 19; AP analysis, Aug. 20)

Cluster A — “Useful liquidity backstop, but not QE” (medium confidence). Gennadiy Goldberg of TD Securities described the move as Treasury verbal intervention while stressing it was not QE; David Russell of TradeStation said the policy center of gravity could shift toward Treasury. An elastic “Bessent put” could improve off-the-run liquidity and interrupt thin-market momentum over days or months. Thursday's reversal and an absence of market dysfunction are the chief counterevidence. (Axios, Aug. 19; Reuters, Aug. 20)

Cluster B — “Too small for fiscal, inflation, and supply fundamentals” (high confidence). Goldberg, Daleep Singh of PGIM, Macquarie strategists, and UBS Wealth Management converged on the view that buybacks cannot sustainably defeat deficits, oil/inflation risk, and roughly $550 billion of estimated quarterly corporate bond issuance. The horizon is quarters: elevated/volatile long yields, mortgages, and all-in funding costs. A credible fiscal plan, falling inflation, smaller coupon supply, or consistently larger operations would disconfirm it. (AP analysis; Reuters)

Cluster C — “Treasury activism adds policy and dollar risk” (medium-high confidence). Krishna Guha of Evercore ISI said activist Treasury policy is becoming as material as central-bank policy; Mark Cabana of BofA located the deeper pressure in uncertainty over the Fed's inflation response; MUFG warned intervention without consolidation could weaken Treasury or dollar demand. Michael Feroli of JPMorgan provided the counterweight: short-rate control remains intact. The medium-term implication is more curve and FX volatility, not necessarily immediate Fed impairment. (Axios, Aug. 20; MUFG, Aug. 20)

Consensus, disagreement, resolver. Consensus grants genuine liquidity value but denies a durable solution to supply and fiscal credibility. The dispute is whether repeated operations create a helpful backstop or advertise a politically sensitive yield ceiling. Report inference: the rapid reversal presently favors the structural-fundamentals cluster. September buyback take-up, the promised Bessent/Vought fiscal discussion, November refunding guidance, auction tails, DXY, and mortgage rates are the tests.

3. U.S. initial jobless claims: low layoffs, limited payroll signal

Impact: medium. Fact pattern. At 8:30 a.m. ET, DOL reported 206,000 initial claims for the week ended August 15, 4,000 below Reuters' 210,000 consensus and down 6,000 from a prior revised to 212,000 from 209,000. The four-week average rose to 204,000. Continuing claims rose 18,000 to 1.799 million, above a 1.790 million forecast, with the prior revised to 1.781 million from 1.777 million; the insured unemployment rate held at 1.2%. (DOL, Aug. 20; Reuters, Aug. 20; Investing.com) The upside employment surprise was modest: layoffs stayed low, but continuing claims preserved the soft-hiring signal.

Reaction. The simultaneous Philadelphia Fed release was much stronger than expected, making causal decomposition impossible. From 8:29 to 8:35 a.m., E-mini S&P futures slipped about 2.5 points and EUR/USD fell roughly 0.065%; the 10-year yield moved choppily by less than one basis point in the first minutes. (Google Finance ES; EUR/USD; 10-year) The larger closing risk-off move should not be assigned to claims.

Cluster A — “Low-fire resilience and rough balance” (high confidence on layoffs; medium on policy). Matthew Martin of Oxford Economics, Carl Weinberg of High Frequency Economics, and Jim Baird of Plante Moran said demand is soft but labor supply has weakened more, leaving stable rather than rapidly expanding conditions. Claims remained well below year-ago levels and did not show wildfire-related deterioration. The one-to-three-month implication is low layoff-loop risk and room for the Fed to prioritize inflation. Rising continuing claims, the July payroll loss, or evidence that supply rather than demand explains stability are risks. (Reuters; AP; Plante Moran, Aug. 20)

Cluster B — “Claims are a poor payroll forecaster” (medium-high confidence). Thomas Simons of Jefferies noted that claims failed to predict July's payroll decline and have not forecast payrolls well in this regime. The implication is to avoid mechanically upgrading August payrolls or Fed pricing from survey-week claims alone. A persistent multiweek rise aligned with JOLTS and survey hiring would restore signal. (Reuters)

Consensus, disagreement, resolver. Consensus is a low-hire, low-fire labor market, not reacceleration. The disagreement is how much low claims say about payroll growth and policy. Report inference: the print modestly weakened the labor-downside case but was confirmatory, not policy-changing. August payrolls on September 4, JOLTS, subsequent claims, and PMI employment indexes decide it.

4. Philadelphia Fed: exceptional expectations, less uniform current acceleration

Impact: medium-high. Fact pattern. At 8:30 a.m. ET, the August general-activity index rose to 47.4 from 41.4, versus a 24.1 median on Investing.com and 25.0 on several calendars—the strongest since April 2021. Employment jumped to 27.9 from 10.0 and the workweek to 26.5. Yet current new orders eased to 30.1, shipments to 27.7, and inventories turned negative. Prices paid/received cooled to 40.9/17.7 but remained elevated, while future price indexes rose. Six-month activity surged to 73.6, the highest since August 1983, and future capex to 48.2, a 53-year high. Responses were collected August 10–17. (Philadelphia Fed primary PDF; calendar consensus)

Reaction. AP attributed part of the 10-year's close at 4.70%, up from 4.65%, to this report and claims, but the simultaneous releases cannot be decomposed. DXY recovered only about 0.12% from its 98.50 area despite the data. (AP close; FXStreet, Aug. 20)

Cluster A — “Growth acceleration is rate-bearish” (medium confidence). Drew Fisher of Pure Rate Mortgage, a public mortgage-market commentator rather than an institutional economist, treated the employment, capex, and expectations strength as confirmation of higher Treasury and mortgage yields. The days-to-weeks implication is reduced easing conviction and pressure on duration. A weak national flash PMI or ISM would disconfirm the regional read-through. (Fisher, Aug. 20)

Cluster B — “Powerful details, limited policy and dollar follow-through” (medium-high confidence). Joshua Gibson of FXStreet stressed that the internals were stronger than the headline but observed only a small dollar response and no forced short-end regime shift. The implication is resilience that still needs national confirmation. Another strong PMI/ISM would reject the regional-noise discount. (Gibson, Aug. 20) Public same-day, freely verifiable sell-side commentary specific to this release was sparse; secondary commentary is labeled.

Consensus, disagreement, resolver. Consensus is an unequivocal upside surprise, directionally hawkish for rates. The disagreement is whether the headline marks national acceleration. Report calculation/inference: an equal-weight ISM-style composite of orders, shipments, employment, delivery times, and inventories is about 57.5, slightly below July's 58.1; the headline accelerated, the component composite did not. Friday's flash PMI, September ISM, manufacturing hours, durable goods, and next month's regional survey resolve it.

5. Thirty-year TIPS: buyers arrived at nearly 3% real yield

Impact: medium. Fact pattern. Competitive bids for the $8 billion, 29-year 6-month reopening closed at 1:00 p.m. ET. The 2⅜% TIPS stopped at a 2.973% high real yield with 2.82 bid-to-cover. Indirects won $6.708 billion, directs $1.068 billion, and dealers only $167 million—84.4%/13.4%/2.1% of competitive awards. A contemporaneous tape estimate put when-issued at 2.991%, implying a strong 1.8-basis-point stop-through; unlike the official statistics, that comparison is secondary. Versus February, the real yield rose 50 basis points while cover improved from 2.75 and indirect participation from 78.3%. (Treasury announcement; official result; February result)

Reaction. Long-bond futures rose 3/32 between 12:59 and 1:05 p.m.; E-mini S&P futures were 2.5 points lower and EUR/USD 0.01% higher by 1:05. The clean but tiny window indicates a positive rates auction with negligible standalone FX/equity impact. (bond futures; ES; EUR/USD)

Cluster A — “Strong end-user demand; no buyer's strike” (high confidence on quality, medium on persistence). No named public post-auction strategist note was found. The stop-through, 2.82 cover, 97.9% nondealer take, and 2.1% dealer residue objectively support the report's inference that the sale cleared strongly. A same-day market participant made that reading; Julius Baer's August 18 institutional view similarly argued that supply was clearing without persistent tails or dealer accumulation. The near-term implication is support for long TIPS; repeated tails or higher dealer shares would refute it. (auction tape; Julius Baer view)

Cluster B — “Restrictive warning versus rare value” (medium confidence). James Okafor framed near-3% real yields as evidence of restrictive conditions and underpriced hike risk; Elias Haddad of BBH emphasized supply/liquidity and fiscal-credibility strain. Michael Lebowitz countered that 3% real TIPS are historically cheap and diversifying, with price upside if inflation and sentiment normalize. These are contextual same-day/nearby views, not claimed as post-auction notes. The quarterly implication spans persistent restrictive real rates versus unusually attractive forward real returns. (Okafor, Aug. 20; BBH, Aug. 20; Lebowitz, Aug. 19)

Consensus, disagreement, resolver. The auction was strong at a punitive real yield. The dispute is whether 3% compensates for durable structural risk or is a rare entry point. September auctions, dealer shares, real-yield/breakeven decomposition, CPI/PCE, Fed guidance, and indirect demand decide it.

6. Conference Board LEI: broader expansion, still a two-speed economy

Impact: low-to-moderate. Fact pattern. At 10:00 a.m. ET, July LEI rose 0.2% to 99.5, versus +0.1% consensus; June was revised to −0.1% from −0.2% and May to +0.2% from +0.1%. Seven of ten July components contributed positively, led by lower claims, permits, and the yield spread. Consumer expectations and core capital-goods orders detracted; manufacturing hours were flat. The six-month change turned positive, +0.2%, for the first time in more than four years, with nine of ten components higher. The coincident index rose 0.2% to 114.8, although its payroll component was flat. Several orders/income inputs were imputed and remain revisable. (Conference Board technical notes; official release)

Reaction. From 9:59 to 10:05, the 10-year yield moved only about 0.2 basis point, DXY was nearly flat, SPY slipped roughly 0.06%, QQQ was flat, and IWM about 0.25% lower. The tiny, mixed response is consistent with a composite built mainly from known inputs; Deere's call and the day's larger oil/rates news contaminate it. (10-year minute series; DXY; SPY)

Cluster A — “Moderate expansion / soft landing” (medium-high confidence). Justyna Zabinska-La Monica of the Conference Board said the fourth increase in six months and newly positive six-month rate point to moderate growth. The Board maintained 1.9% real GDP growth for 2026 and 2027, driven partly by AI business investment. The implication is lower near-term recession risk and support for capital spending; renewed declines in orders, permits, or claims would disconfirm it. (Conference Board, Aug. 20)

Cluster B — “AI/capital strength versus a squeezed consumer” (medium confidence). Zabinska-La Monica emphasized consumer expectations as a drag and high living costs as a constraint on lower- and middle-income spending. Same-day RISMedia analysis, with a secondary quote from Keller Williams co-founder Gary Keller, reinforced the household-budget risk. The implication is industrial/business-investment outperformance against vulnerable consumer cyclicals. Real-wage gains and a sustained confidence/retail rebound are the counterevidence. (RISMedia, Aug. 20)

Cluster C — “Positive but low-information market signal” (high confidence on immediate impact). Public same-day sell-side commentary was sparse. The muted reaction supports the report's inference that a one-tenth beat in a largely pre-released composite does not alter Fed pricing by itself. Repeated broad upside would change that conclusion.

Consensus, disagreement, resolver. Moderate expansion is the base case; consumer confidence is the weak link. The dispute is whether broad diffusion marks durable reacceleration or financial/housing support masking household fragility. Report inference: diffusion matters more than the one-tenth beat, but flat payrolls and negative expectations prevent an all-clear. August LEI, ISM orders, permits, consumption, confidence, and revisions resolve it.

4. Complete earnings / call table

The impact rank below is authoritative; detailed-section numbers are navigation labels, not a second ranking.

Rank	Company / ticker	Release / call (ET)	Result versus consensus	Guidance / decisive call point	Price reaction	Sector / macro read-through
1	Walmart (WMT)	7:00 / 8:00 a.m.	Revenue $187.94bn vs $186.62bn; EPS $0.81 vs $0.74; U.S. comp +2.6% vs +3.8%	Q3 EPS $0.62–$0.64 vs $0.68; consumer softer, refund reinvested	−9.15%	Broad consumer, grocery/value, index-heavy
2	Advance Auto Parts (AAP)	6:30 / 8:00 a.m.	Sales $2.00bn vs $2.03bn; comp −0.5% vs +1.4%; EPS $1.03 vs $0.81, including ~$0.31 refund	Operating guide unchanged; DIY low-double-digit decline	−24.5%; AZO ~−4.4%	Lower/middle-income DIY and auto aftermarket
3	Alibaba (BABA)	5:30 / 7:30 a.m.	Revenue RMB268.95bn vs RMB268.88bn; EPS RMB8.52 vs RMB10.53	Cloud +45%, capex +75%, FCF −RMB44.7bn; cloud acceleration guided	Premarket ~−6%; close +1.3%	China AI/cloud positive; commerce/cash quality mixed
4	Deere (DE)	6:20 / 10:00 a.m.	EPS $5.10 vs $4.69; equipment sales $10.999bn vs ~$10.73–10.81bn	FY net income $4.75–$5.00bn; measured ag recovery, strong construction backlog	+7.3%	Farm trough, construction/data-center demand
5	Ross Stores (ROST)	4:11 / 4:15 p.m.	Sales $6.265bn vs $6.15bn; EPS $2.66 vs $1.94; ~$2.06 ex-refund; comp +10%	FY EPS $8.61–$8.77 incl. refund; traffic and brand access strong	+8.46% after hours	Off-price/value share gain; TJX/BURL positive
6	Nordson (NDSN)	Aug. 19 4:32 p.m. / Aug. 20 8:30 a.m.	Sales $817.7m vs $779.5m; EPS $3.25 vs $3.09	Backlog +35%; FY sales/EPS raised; electronics near peak but U.S. fabs not in orders	+7.9%	Advanced packaging/inspection and medical precision
7	NetEase (NTES)	4:58 / 8:00 a.m.	Revenue RMB30.11bn vs RMB29.10bn; EPS RMB12.02 vs RMB15.56	Games margin 76.1%; Sea of Remnants/pipeline need proof	−5.8%	China gaming margins versus hit concentration
8	JD Sports (LSE: JD.)	2:00 a.m.; no call	Organic −1.3% vs +0.5%; LFL −3.1%; North America LFL −6.8%	Profit guide cut to £700–£800m	−14.3% London; NKE ~−2.1%	U.S. discretionary footwear and Nike product cycle
9	Coty (COTY)	Aug. 19 4:30 p.m. / Aug. 20 8:00 a.m.	Revenue $1.269bn vs $1.19bn; loss $0.02 vs $0.01	No FY guide; Q1 EPS below consensus, margins weaker	−9.4%	Company execution/portfolio warning, not broad beauty collapse
10	OSI Systems (OSIS)	4:02 / 4:30 p.m.	Revenue $484.1m vs $529.7m; EPS $3.78 vs $3.77	$50m Middle East pushout; record backlog; FY27 2H-weighted	−13.4% after hours	Security-screening/RF demand versus conversion timing
11	ScanSource (SCSC)	~8:30 / 10:30 a.m.	Sales $953.1m vs $822.0m; EPS $1.46 vs $1.14	Organic FY27 sales +6%–10%; $220.5m MicroAge acquisition	+9.8%	IT-distribution demand; services capital allocation
12	LSI Industries (LYTS)	Premarket / 11:00 a.m.	Revenue $234.6m vs $222.8m; EPS $0.38 vs $0.35	Legacy pricing creates 50–100bp margin headwind for 1–2 quarters	−14.2%	Grocery/C-store renovation strong; integration risk
13	Futu (FUTU)	6:00 / 7:30 a.m.	Revenue HK$7.20bn vs HK$5.91bn; ADS EPS HK$26.08 vs HK$25.85	Mid-single-digit asset outflows; Q3 activity modestly softer	+3.0%, off intraday high	China/overseas brokerage scale and regulation
14	Aegon (AEG)	1:00 / 8:00 a.m.	Operating result €804m vs €752m; OCG €416m vs €376m; FCF miss	Buyback raised; U.S. strain, transition, CFO departure	−2.9% ADR	U.S. life growth versus distributable capital quality
15	Atour (ATAT)	~6:00 / 7:00 a.m.	Revenue RMB3.49bn, +41.4%; public EPS comparison low-confidence	FY revenue growth raised to 30%; margins fell	−7.5%	China lodging/retail share gains versus mix
16	ATRenew (RERE)	4:09 / 8:00 a.m.	Revenue RMB6.609bn vs RMB6.326bn; GAAP NI RMB129.1m vs RMB124.5m	Q3 +23.1%–25.1%; inventory/borrowings rose	−10.9%	Recommerce share gain versus working capital
17	Daqo (DQ)	6:30 / 8:00 a.m.	Revenue $62.7m vs $114.7m; loss $1.20 vs $0.53	ASP below cash cost; output guide raised; policy floor optionality	+5.7% after deep early loss	Polysilicon trough and China supply discipline
18	Hovnanian (HOV)	9:15 / 11:00 a.m.	EPS −$0.70 vs −$0.56; revenue $705.8m vs $723.5m	Q4 margin recovery guided; first pretax guide miss in 23 quarters	−7.4%	Homebuilder execution plus higher rates
19	Flowers Foods (FLO)	4:10 p.m.; call Aug. 21	Revenue $1.193bn vs $1.23bn; EPS $0.21 vs $0.22	Volume −5.8%; every FY range cut	Regular −4.8%; AH another −4.2%	Price-sensitive packaged bread; premium snacks stronger
20	Huize (HUIZ)	5:00 / 8:00 a.m.	H1: premiums +29.8%, revenue +5.8%; no comparable consensus	No numerical H2 guide; Asia/AI optionality	Premarket +177%; close +26% on 94x volume	Real insurtech improvement, large microstructure overlay

5. Detailed company sections with opinion clusters

1. Walmart (WMT): a beat that failed the comp, cadence, and valuation tests

Release/call: 7:00/8:00 a.m. ET. Impact: very high. Revenue was $187.94 billion versus $186.62 billion FactSet and adjusted EPS $0.81 versus $0.74. The decisive miss was U.S. comparable sales ex-fuel: +2.6% versus +3.8% LSEG, with traffic +1.5% and ticket +1.1%. A 125-basis-point Medicare/pharmacy-deflation drag meant ex-pharmacy comps were 3.4%. U.S. e-commerce grew 24%, Walmart Connect ex-VIZIO 43%, and global membership fees 17%. Adjusted operating income rose 17.4% constant-currency but included about 750 basis points of net benefit from nearly $2.9 billion of tariff refunds. (Walmart release; 8-K; AP; Reuters)

Guidance/call. FY sales growth rose to 4%–5%, operating-income growth to 7%–8.5%, and EPS to $2.80–$2.87, but FactSet expected $2.90. Q3 EPS of $0.62–$0.64 missed $0.68 consensus. CFO John David Rainey said gasoline above $4 forced June trade-offs and the consumer was “arguably” softer than in February. Management had expanded rollbacks to 11,000 items and said unit response precedes durable traffic/share gains; it asked investors to judge refund-assisted Q2 together with price-investment-heavy Q3. Digital incremental margins recently ran high-single-digit to low-double-digit. (call transcript) WMT fell 9.15%, erasing more than $80 billion and recording its worst day since May 2022. (AP close)

Cluster A — “The trade-down tailwind is fading” (medium-high confidence). Neil Saunders of GlobalData and Brian Jacobsen of Annex Wealth Management saw the six-year-low comp, June trade-offs, and gasoline pressure as a warning that consumer stamina and higher-income migration may be fading. The one-to-two-quarter implication is downside for broadline discretionary demand and lower-income credit. Ex-pharmacy comps, share gains across cohorts, and improving July demand are the counterevidence. (Axios, Aug. 20; Reuters)

Cluster B — “The structural flywheel is intact” (high operating evidence; medium near-term-stock confidence). Oliver Chen of TD Cowen, Michael Lasser of UBS, and Sarah Henry of Logan Capital emphasized share, low-cost positioning, e-commerce, advertising, and improving digital economics. The multi-quarter implication is profit growing faster than sales and widening competitive advantage. The refund is one-time, rollback elasticity unproven, and the entering multiple high. (Axios)

Cluster C — “Guidance quality missed the valuation bar” (high confidence). Kate McShane of Goldman Sachs said investors expected a larger annual raise. Q3 EPS was five cents below consensus and full-year EPS remained below FactSet despite the windfall; deferred Flipkart timing and successful price investment could repair the bridge. (Axios)

Consensus/disagreement/resolver. Walmart remains the best-positioned value retailer, but its cadence was less clean than its multiple demanded. The dispute is true consumer deterioration versus pharmacy/timing noise. Report inference: the selloff combined macro anxiety with earnings-quality disappointment; it does not yet prove structural share loss. Q3 ex-health traffic/comps, rollback elasticity, fuel, ad growth, digital margin, and Q4 Flipkart reversal resolve it.

2. Alibaba (BABA): cloud acceleration won a volatile vote over cash burn

Release/call: 5:30/7:30 a.m. ET. Impact: high. Revenue rose 9% to RMB268.953 billion, essentially in line with LSEG's RMB268.88 billion. Non-GAAP EPS fell 42% to RMB8.52 versus RMB10.53, a 19% miss; adjusted EBITA fell 30%. AI Cloud and Compute revenue grew 45% to RMB48.437 billion, segment EBITA 133%, and margin reached roughly 12%. E-commerce revenue rose 4%, but China commerce fell 8% and reported customer-management revenue 7% (about +1% like-for-like). Capex rose 75% to RMB67.678 billion and free cash flow was negative RMB44.670 billion. (official release PDF; Reuters)

Call/reaction. CEO Eddie Wu forecast further cloud acceleration, sequential margin expansion, AI-product annualized revenue near $10 billion next quarter, and Model/Application Services ARR above RMB30 billion by year-end. Management said server/chip investment generally breaks even inside three years and moved quick-commerce's profitability target to FY29. Analyst questions focused on supply limits, model/API mix, and which layer captures economics; management said infrastructure/chips lead near-term value. (same-day transcript) The ADR fell 6.3% premarket, opened down 5.1%, then closed up 1.3% at $130.53; China internet ETFs stayed weak. (AP)

Cluster A — “AI/cloud inflection is real” (medium-high confidence). Public pre-print views from Morgan Stanley, UBS analyst Kenneth Fong, and Barclays analyst Jiong Shao expected roughly 45% cloud growth and improving margins; the print delivered both. The multi-quarter implication is estimate and valuation migration toward infrastructure. Segment redefinition, chip constraints, pricing, and undisclosed concentration are risks. (Morgan Stanley public summary, Aug. 16; UBS summary; Barclays, Aug. 14)

Cluster B — “Growth quality waits for capex returns” (medium confidence). Zacks analyst Vasundhara Sawalka and same-day Investing.com analysis emphasized the 19% EPS miss, capex at roughly 25% of revenue, and worse FCF. Cloud's 133% EBITA growth and the claimed three-year server payback are the counterevidence. The quarters-ahead test is utilization, depreciation, and FCF recovery. (Zacks, Aug. 18; Investing.com, Aug. 20)

Cluster C — “Commerce is stabilizing, not reaccelerating” (medium confidence). Shao's stabilization thesis found support in improving instant-commerce economics and AliExpress operating profit, but +1% underlying customer revenue and intense competition keep it defensive. CMR/GMV, subsidy intensity, and FY29 profitability decide it.

Consensus/disagreement/resolver. AI demand and cloud leverage were proved; near-term EPS/FCF were deliberately sacrificed. The disagreement is whether 45% growth at 12% margin validates a self-funding high-return platform or whether quarterly capex exceeding cloud revenue defers cash too far. Report inference: the reversal voted for cloud, not for the China consumer complex. No verifiable public post-print rating/target changes were available by cutoff; the named research is dated and publicly accessible.

3. Deere (DE): diversification validated a trough, not a sharp farm recovery

Release/call: 6:20/10:00 a.m. ET. Impact: high. Deere reported $5.10 EPS versus $4.69 Refinitiv, $12.608 billion of total revenue, and $10.999 billion equipment sales versus a publicly cited $10.73–$10.81 billion range. Production & Precision Ag sales/profit fell 6%/9%; Small Ag & Turf rose 12%/28%; Construction & Forestry rose 18%/84%. FY net-income guidance narrowed upward to $4.75–$5.00 billion and equipment cash flow to $5.0–$5.5 billion. (release; 8-K; Kiplinger/Refinitiv)

Call/reaction. Management called 2026 the ag-cycle bottom but described 2027 as a measured, uneven recovery. Planter/sprayer early orders rose mid-single digits; See & Spray factory orders nearly doubled, but 8R tractor orders were only as expected. Construction had a four-to-five-month backlog versus two to three normally, supported by infrastructure, data centers, pipelines, and rental. Deere expects roughly $1 billion of FY27 tariff run rate versus about $750 million net in FY26 absent further refunds. (call transcript) DE closed 7.3% higher near $622.78.

Cluster A — “Execution and diversification validate the trough” (medium confidence). Investing.com editor Vlad Schepkov viewed the clean beat, higher guide floor, improved profitability, and construction resilience as sufficient for repricing. Underproduction, better early orders, and technology take-rates support FY27 earnings improvement without a fast large-ag snapback. Crop economics, Brazil/Europe, construction normalization, and tariffs are risks. (Schepkov, Aug. 20)

Cluster B — “The stock priced recovery before the core delivered” (medium-high confidence). Itai Smidt noted that PPA still contracted, the comparable equipment-sales beat was only about 2%, and tariff recovery helped EPS. JPMorgan's public August 14 Neutral context warned that flat-to-slightly higher orders would not support consensus's FY27 PPA growth. The six-to-twelve-month downside is valuation compression if orders/margins disappoint; high-single-digit early-order growth or faster inventory clearing would refute it. (Smidt, Aug. 20; JPMorgan public summary, Aug. 14)

Consensus/disagreement/resolver. Operations were strong and construction/small ag cushioned large ag; any recovery is measured. The disagreement is whether order/inventory evidence justifies the FY27 inflection already in price. Report inference: AGCO/CNH receive a constructive inventory signal, not proof of a sharp equipment cycle; construction/rental peers get the cleaner positive read-through. November guidance, final early-order programs, PPA margins, Brazil, backlog conversion, and tariffs decide it. No verifiable same-day public post-call sell-side note was found; named commentators and the pre-print note are labeled.

4. NetEase (NTES): operating margins passed; pipeline and investment marks failed

Release/call: 4:58/8:00 a.m. ET. Impact: high for China gaming. Revenue rose 7.9% to RMB30.11 billion versus RMB29.10 billion FactSet, but non-GAAP diluted EPS was RMB12.02 versus RMB15.56. The primary bridge was equity-investment losses, impairments, and a 25.5% tax rate rather than gross-profit deterioration. Games revenue rose 9.7%; games gross margin expanded 590 basis points to 76.1%, consolidated margin reached 70.5%, and expenses grew only 1.5%. Cloud Music revenue was flat at RMB1.98 billion, with margin up to 37.4%; member revenue grew in the half, but mix diluted ARPPU. Net cash was RMB167.5 billion and operating cash flow RMB10.0 billion. (primary release; 6-K; Dow Jones)

Management acknowledged Sea of Remnants' steep learning curve and criticism, promised updates, and withheld bookings guidance; Ananta awaits a Gamescom update, while other titles will launch only after polishing. The ADR fell about 5.8%. (call transcript)

Cluster A — “Operating compounder intact” (medium-high confidence). Goldman Sachs' Lincoln Kong called the quarter solid; Goldman's public June Buy view emphasized evergreen/global titles, lower channel costs, operating-profit growth, and FCF. The 590-basis-point games-margin expansion supports it; slate concentration and marketing/R&D are risks. (Goldman public summary)

Cluster B — “Pipeline execution is the binding risk” (high confidence on debate; low on outcome). Citi's Alicia Yap, CICC's Xue Zhang, and Kong used Q&A to test Sea of Remnants, Ananta differentiation, and the lean slate. These are attention signals, not recommendations. Bookings remain catalyst-heavy; player retention and Gamescom reception decide it.

Cluster C — “The profit miss is real but non-operating” (high confidence). Nomura analysts cited publicly by Dow Jones had warned that marks on Alibaba/PDD holdings could cause a miss. That supports near-term earnings-noise/quality risk, while cash, buybacks, and operating leverage argue against treating it as a franchise-margin break.

Consensus/disagreement/resolver. Game economics were strong; the slate needs proof. The dispute is whether to look through the 23% EPS miss. Report inference: the selloff overweights the headline mark but reasonably discounts launch concentration. Sea of Remnants bookings, Ananta, Q3 contract liabilities, consolidated margin, and Cloud Music ARPPU settle it. Public post-print analyst commentary was sparse and is not inferred.

5. Futu (FUTU): record activity met a regulatory and Q3 run-rate warning

Release/call: 6:00/7:30 a.m. ET. Impact: medium-high. Revenue rose 35.6% to HK$7.20 billion versus HK$5.91 billion in an Investing.com compilation; ADS EPS was HK$26.08 versus HK$25.85. Funded accounts grew 33.6% to 3.843 million, client assets 43.6% to HK$1.40 trillion, trading volume 78.8% to HK$6.42 trillion, and margin balances 85.1%; wealth assets rose 10.4%. (6-K exhibit; consensus)

No revenue/EPS guidance was given. Management said Malaysia reached break-even and Prediction Markets traded more than 200 million contracts in their first month. Q&A also disclosed mid-single-digit cumulative client-asset outflows after May 22 rules and modestly softer Q3-to-date accounts and trading. (transcript) FUTU rose 8.6% premarket but closed only +3.0%; TIGR gained 4.3%, while HOOD/IBKR fell.

Cluster A — “International scale overwhelms the regulatory shock” (medium confidence). Investing.com editor Rachael Rajan emphasized record volume, the large revenue beat, account growth, and overseas leverage. The FY26–27 implication is a broader earnings base if U.S./ASEAN cohorts hold ARPU and profitability. Activity normalization, CAC, margin losses, and new rules are risks. (Rajan, Aug. 20)

Cluster B — “A peak quarter with regulatory cyclicality” (medium-high confidence). CMB International's Nika Ma, explicitly older context, centered mainland restriction risk; the call's asset outflow and softer Q3 evidence strengthen that caution. Q2's 79% volume and 85% margin-balance growth may not be a run rate. (CMBI public report)

Consensus/disagreement/resolver. Diversification is real; regulation and trading cyclicality bind. The disagreement is higher earnings base versus market-driven peak. Report inference: TIGR gets the cleanest positive read-through; HOOD/IBKR do not. Q3 net inflows, retention, activity, take rate/CAC, overseas profit, and Prediction Markets monetization decide it. No public same-day sell-side note was found.

6. Advance Auto Parts (AAP): refund-aided EPS could not rescue demand credibility

Release/call: 6:30/8:00 a.m. ET. Impact: high. Sales were about $2.00 billion versus $2.03 billion Zacks, comps −0.5% versus +1.4% consensus, and adjusted EPS $1.03 versus $0.81. A $26 million tariff refund contributed about 130 basis points of gross margin and $0.31 EPS, making normalized EPS roughly $0.72; refund-excluded gross margin still improved about 110 basis points. Pro grew low single digits, but DIY fell low double digits as price-sensitive households absorbed fuel and living costs. YTD FCF swung to +$120 million. (official IR; AP/Zacks; transcript)

FY sales, comp, margin, and FCF guidance was retained; EPS rose $0.20 largely because of interest income, not operations, and no meaningful second-half refund is expected. AAP fell 24.5%; AutoZone fell roughly 4.4%. (reaction; peer move)

Cluster A — “Turnaround margins and cash are real” (medium confidence). RBC's Steven Shemesh expected product-margin and supply-chain gains; the ex-refund improvement validated the mechanism. Guggenheim's Steven Forbes and Goldman's Mark Jordan used Q&A to test hub productivity and Pro wins. Inventory, distribution, and assortment can support 2027 margin; volume deleverage and share loss threaten it. (RBC, Aug. 13)

Cluster B — “Low-quality beat; fragile guide” (high confidence). Citi's Steven Zaccone, UBS's Michael Lasser, and Morgan Stanley's Simeon Gutman publicly challenged the back-half comp and core-margin bridge on the call. The sales miss, DIY contraction, refund-adjusted EPS shortfall, and unchanged operations justify estimate pressure. Early-Q3 transactions, positive Pro, and pricing are the counterevidence. (same-day analysis)

Cluster C — “Mixed consumer, not broken aftermarket” (medium confidence). The evidence separates stressed DIY households from resilient professional repair. Aging vehicles remain supportive, but customers trade down and defer optional work.

Consensus/disagreement/resolver. Self-help is working; sales recovery is late and the beat overstated core performance. The dispute is whether ≥100 basis points of 2027 margin improvement survives weak DIY. Report inference: AZO's sympathy move is directionally valid, but AAP-specific execution/refund quality makes the full decline non-transferable. Ex-refund margin, FCF, DIY transactions, fuel, and hub comps decide it.

7. Aegon (AEG): capital beat, lower-quality bridge

Release/call: 1:00/8:00 a.m. ET. Impact: medium-high. Operating result was €804 million versus €752 million company-collected consensus, operating capital generation €416 million versus €376 million, but free cash flow €392 million versus €415 million. Holding cash reached €1.66 billion; the second-half buyback rose €150 million to €350 million and dividend to €0.21. Transamerica operating result grew 14% and individual-life sales 54%; U.S. RBC was 420%. (release; company consensus)

The call emphasized instant underwriting, >12% expected new-business returns, and higher new-business strain. Offsets were a €231 million assumption/model-review hit, €350 million of transition cost, and CFO Duncan Russell's April 2027 departure ahead of a planned U.S. domicile. (transcript) The ADR fell 2.9%, more than U.S. life peers.

Cluster A — “Real beat, partly non-recurring and priced” (high confidence). Morgan Stanley analyst Hadley Cohen maintained Equal-weight/€7.70 and said about €34 million of OCG came from one-offs plus favorable experience. The implication is limited upside without normalized OCG growth. (same-day summary)

Cluster B — “U.S. growth and capital enable rerating” (medium confidence). Pre-print Buy views from JPMorgan's Farooq Hanif and Berenberg's Michael Huttner contrasted with UBS analyst Nasib Ahmed's Hold. Bulls see higher-return U.S. life sales, asset-management improvement, and buybacks offsetting transition cost; strain, assumptions, succession, and transaction timing are disconfirmers. (public analyst page)

Consensus/disagreement/resolver. Commercial and capital-return momentum are strong; headline quality is less exceptional. The dispute is whether U.S. growth compounds future OCG or consumes too much present capital. Report inference: fast underwriting is positive for peers, but sales growth is not immediate distributable cash. H2 normalized OCG/FCF, RBC, strain, experience, the EGM, CFO hire, and U.S.-GAAP dry runs decide it.

8. ScanSource (SCSC): demand beat, then the capital-allocation debate

Release/call: announced about 8:30/10:30 a.m. ET. Impact: medium-high. Sales rose 17.3% to $953.1 million versus $822.0 million, and adjusted EPS was $1.46 versus $1.14. Hardware mix compressed gross margin 35 basis points to 12.6%, but EBITDA margin edged higher. FY27 organic guidance calls for 6%–10% sales growth, $158–$165 million EBITDA, and at least $85 million FCF. (primary release; transcript/consensus)

ScanSource also agreed to pay $220.5 million cash for MicroAge, financed by its credit facility. It claims first-year margin, EPS, and FCF accretion but withheld acquired metrics and combined guidance until Q1. (deal release) The stock jumped 17% premarket, touched $66, and closed +9.8% at $56.39.

Cluster A — “Demand and leverage inflected” (high confidence). Call participants Greg Burns of Sidoti, Keith Housum of Northcoast, and Guy Hardwick of Barclays highlighted leverage, breadth, and whether supply/macro could interrupt it. Management said most growth was demand—not price—across physical security, networking, cloud, CX, and cybersecurity. The 6–12-month implication is positive for CNXN/NSIT and enterprise refresh; large-deal lumpiness, mix, Brazil, and harder comparisons are risks.

Cluster B — “Strategic services bridge or oversized bet?” (medium confidence). Raymond James analyst Adam Tindle noted the acquisition exceeds 20% of market cap and questioned it versus repurchasing shares at 6–7x EBITDA. Management argued services returns should be higher. Integration can structurally improve mix—or credit-funded leverage and channel conflict can erase the benefit.

Consensus/disagreement/resolver. The quarter and distribution-demand signal were clearly strong. The dispute is whether MicroAge converts a hardware rebound into durable services growth. Report inference: the intraday fade rewarded demand more confidently than the deal. Q1 combined guidance, acquired revenue/margins, retention, leverage, and FCF decide it. Public post-print notes were sparse; named evidence is Q&A, not claimed recommendations.

9. Atour Lifestyle (ATAT): share gains without enough profit conversion

Release/call: about 6:00/7:00 a.m. ET. Impact: medium-high. Revenue grew 41.4% to RMB3.490 billion; adjusted net income rose 30.8%. A displayed $0.58 per-ADS consensus versus $0.60 reported is low-confidence because vendors mix measures. Hotels grew 19.2% to 2,175, system RevPAR 0.7%, but mature-hotel RevPAR was only 97% of prior-year. Retail revenue surged 63.2%. Atour raised FY retail growth to 40% and total revenue growth to 30%, yet adjusted net and EBITDA margins fell 130/120 basis points on supply-chain, product mix, and marketing. (release PDF; transcript) ATAT fell 7.5%.

Cluster A — “Brand share and a real hotel-to-retail flywheel” (medium confidence). Morgan Stanley's Dan Qi and Citi analysts Jiwei Liu and Lydia Ling used Q&A to test the pipeline, demand mix, and retail durability. These are attention signals. Stable system RevPAR, unit growth, an 811-hotel pipeline, and product extension support multi-quarter branded share gains; mature-hotel contraction and franchise/product execution are risks.

Cluster B — “Growth quality is deteriorating” (medium-high confidence). BofA's Ronald Leung focused on volatile RevPAR and UBS analyst Xin Chen on margin guidance. Revenue outgrew profit, both segment margins compressed, and management still expects annual margin decline. Summer stabilization, retail scale, differentiated pricing, and buybacks could disconfirm the bear case.

Consensus/disagreement/resolver. Atour is gaining lodging and sleep-product share; investors demand conversion. The dispute is temporary mix/investment versus structural dilution from supply chain and online retail. Report inference: this is a margin/customer-acquisition warning, not proof of collapsing China travel. Q3 mature RevPAR, openings, retail retention, gross margins, marketing, and taxes decide it. No same-day rating change was publicly verifiable.

10. Hovnanian (HOV): margin recovery could not prevent an execution penalty

Release/call: 9:15/11:00 a.m. ET. Impact: medium-high. GAAP EPS was −$0.70 versus −$0.56 in thin public consensus and revenue $705.8 million versus $723.5 million. Adjusted homebuilding gross margin improved sequentially to 14.6%, but contracts fell 4.6%; backlog dollars rose 5.1%. A small adjusted pretax loss marked the first guide miss in 23 quarters because JV income and delivery timing lagged. Q4 guidance is $800–$900 million revenue, 15.0%–16.5% adjusted margin, and $15–$30 million adjusted pretax income. (release; transcript) HOV recovered from −12.7% premarket to −7.4%, still about five points worse than ITB as the 10-year rose.

Cluster A — “The margin inflection is real, but conditional” (medium confidence). Citizens analyst Jay McCanless called the margin news strong and tested durability. Two sequential gains, lower incentives, newer land, and a 15.75% Q4 midpoint support improvement; rates, incentives, lumber, and volume threaten it.

Cluster B — “Execution and demand deserve a discount” (medium-low confidence). Same-day Investing.com synthesis emphasized the first pretax miss in 23 quarters; Q&A from Housing Research Center's Alex Barron and Zelman's Natalie Kulasekere focused on incentives, rates, construction cost, and labor. These questions are not recommendations. Backlog, August contracts, optioned lots, and liquidity are counterevidence.

Consensus/disagreement/resolver. Near-term miss, credible conditional margin path. The dispute is whether Q4 is an inflection or a stable-rate assumption. Report inference: the outsized decline was a company execution/credibility repricing layered on a rates selloff, not a new sector-wide demand shock. Q4 deliveries, JV income, margin, communities, incentives, and mortgages decide it. No same-day authored broker note was public.

11. Daqo New Energy (DQ): disastrous economics, valuable policy optionality

Release/call: 6:30/8:00 a.m. ET. Impact: medium-high for upstream solar. Revenue was $62.7 million versus $114.7 million public consensus and adjusted loss $1.20/ADS versus $0.53. Daqo produced 43,675 metric tons but sold only 15,190; its $4.04/kg ASP was below $4.57 cash cost and $5.95 total cost. Inventory rose $105 million sequentially to $363.5 million and first-half operating cash burn reached $276 million. The defense is $1.92 billion liquidity and no debt; FY production guidance rose to 160,000–180,000 tons. (primary release; transcript/consensus) The ADR reversed from $12.19 to close about +5.7%.

Cluster A — “Enforced rationalization creates a floor” (medium confidence). Roth Capital's Oscar Chen and Jefferies analyst Alan Lau used Q&A to test government enforcement and higher-priced transactions. Management argued price-law and energy-efficiency rules could remove high-cost capacity over 6–18 months. The balance sheet makes Daqo a survivor; noncompliance, weak demand, and inventory liquidation are risks.

Cluster B — “Still an inventory/cash-burn trough” (medium-high confidence). Lau and Goldman analyst Mengwen Wang focused on inventory and shipment strategy. Production nearly triple sales, below-cash-cost ASP, inventory growth, and thin transactions near RMB40/kg support caution. Upstream pain benefits wafer/cell/module input cost until real discipline takes hold.

Cluster C — “AI data-center optionality” (low-medium confidence). Management targets prototypes by year-end and 2027 sales, but the project is pre-revenue and capital intensive. Investing.com editor Vahid Karaahmetovic attributed part of the reversal to bargain hunting/short covering, a market view rather than research. (recap)

Consensus/disagreement/resolver. Q2 economics were poor; Daqo can survive and benefits if policy lifts prices. The dispute is real above-cost transactions before inventory forces discounting. Weekly volumes/prices, inventory, cash flow, Q3 sales/production, enforcement, and closures decide it. No public rating change was found.

12. ATRenew (RERE): 1P share gain versus working-capital quality

Release/call: 4:09/8:00 a.m. ET. Impact: medium-high. Revenue rose 32.4% to RMB6.609 billion versus RMB6.326 billion in a two-analyst S&P feed; GAAP net income beat by 3.7%. Product revenue grew 35.9%, while service revenue fell 4.2% on merchant discounts. First-party gross margin improved to 15.7% and adjusted operating margin 69 basis points to 3.1%, but inventory rose 36.6% and short borrowings 41.2%. Q3 revenue guidance implies 23.1%–25.1% growth. (primary release; transcript) RERE fell 10.9% despite the beat.

Cluster A — “Countercyclical recommerce share gainer” (medium-high confidence). UBS analyst Mandy Lu and Zacks SCR's Brian Lantier used Q&A to frame cautious new-device demand, trade-ins, and overseas scale. Sourcing density, refurbishment, and value can gain share as consumers trade down; subsidies, depreciation, and handset cycles are risks.

Cluster B — “Real 1P margins, weakening capital-light marketplace” (medium confidence). CICC's Wang Zhao focused on service decline and DBS analyst Raphael Tse on store count. Management called fee waivers temporary and closed weak stores. Continued concessions, working capital, debt, and decelerating guidance threaten quality; service recovery and stable inventory days would refute it.

Cluster C — “Overseas optionality remains unproved” (low-medium confidence). The three-year FoneSquare plan could export rising used-device supply, but revenue, take rate, and margin are undisclosed.

Consensus/disagreement/resolver. First-party scaling was strong; public post-print rating notes were unavailable. The dispute is durable quality upgrade versus working-capital-heavy growth masking 3P weakness. Report inference: the 11% selloff prioritized the unresolved cash/mix question over the headline beat. Service growth, operating cash flow, inventory turns, store productivity, and overseas disclosure decide it.

13. LSI Industries (LYTS): demand beat, inherited pricing failed the quality test

Release/call: premarket/11:00 a.m. ET. Impact: medium-high. Revenue was $234.6 million versus $222.8 million and adjusted EPS $0.38 versus $0.35. Display grew 18% organically with 12.4% EBITDA margin; Lighting sales fell 3%, but orders grew 5% and book-to-bill exceeded one. A new $30 million, 2,500-site oil-retailer award was excluded from orders. The problem was legacy SignResource/Royston pricing: management expects a 50–100-basis-point headwind for one or two quarters, with Q1 margin about 30 basis points worse sequentially. (SEC-filed release; transcript) LYTS fell 14.2%.

Cluster A — “Demand is real, not pull-forward” (medium confidence). Oppenheimer's Brent Thillman, Craig-Hallum's Aaron Spychalla, and Texas Capital's Alex Rygiel used Q&A to test pull-forward, QSR, and award timing. Installation schedules, book-to-bill, backlog, and the new award support a refurbishment runway; permitting, weak QSR/auto, and concentration are risks.

Cluster B — “Integration and margin proof required” (medium-high confidence on concern). Those analysts plus H.C. Wainwright's Amit Dayal focused on pricing and backlog burn. The selloff says acquisition-driven growth gets discounted until low-margin legacy work clears. Rapid repricing and otherwise strong organic margins are counterevidence.

Consensus/disagreement/resolver. Healthy grocery/C-store project demand, two-quarter margin test. The dispute is an isolated inherited-pricing issue versus broader integration risk. Report inference: positive narrow demand read-through for renovation/signage, not a broad industrial boom. Q1/Q2 margins, backlog burn, new-order pricing, leverage, and award releases decide it. Q&A is an attention signal; no rating change was public.

14. Nordson (NDSN): a high-quality backlog and electronics call

Release/call: August 19 after close/8:30 a.m. ET Aug. 20. Impact: medium-high. Sales were $817.7 million versus $779.5 million and adjusted EPS $3.25 versus $3.09. Organic sales grew 12%, backlog 35%, and Advanced Technology 31% organically with 600 basis points of EBITDA-margin expansion. FCF conversion was 144%; FY guidance rose to $3.035–$3.075 billion revenue and $11.80–$12.00 EPS. (primary release; consensus)

Management said backlog normally converts within six months while lead times improved, arguing against bottleneck inflation. Electronics inspection/dispensing and Medical & Fluid accelerated; Industrial remained steady. ATS is near a cycle peak and FY27 growth should normalize, but U.S. chip-fab investment has not yet entered orders. (transcript) NDSN rose 7.9%, far ahead of SOXX/XLI.

Cluster A — “Beat-and-raise supports upside” (high confidence). D.A. Davidson reiterated Buy/$345, citing sales/operating profit above model, 12% organic growth versus 6%, guidance, cash, and balance sheet. The target offered little room after the rally, and normalization/backlog slippage are risks. (D.A. Davidson, Aug. 20)

Cluster B — “Durability real, but concentration needs proof” (medium confidence). Baird's Michael Halloran, KeyBanc's Jeff Hammond, D.A. Davidson's Matt Summerville, and Oppenheimer's Christopher Glynn used Q&A to test lead times, ATS dependence, FY27, and incrementals. The positive semicap/medical read-through is strongest through FY26; broader industrial acceleration is not yet shown.

Consensus/disagreement/resolver. High-quality beat-and-raise with cash/order support. The dispute is durable semicap/medical inflection versus peak electronics pull-forward. Report inference: relative performance credited order quality, not just EPS. FY27 guidance, ATS book-to-bill, conversion, U.S.-fab orders, and Industrial margin decide it.

15. Huize (HUIZ): genuine improvement, unreliable price discovery

Release/call: 5:00/8:00 a.m. ET. Impact: medium due exceptional tape. Huize reported H1, not Q2; no reliable consensus exists. Gross written premiums rose 29.8% to RMB4.196 billion and first-year premiums 48.7%; revenue rose only 5.8% to RMB719.8 million. GAAP profit was RMB25.3 million versus RMB2.3 million, but non-GAAP profit only RMB3.1 million because a negative RMB22.3 million share-compensation adjustment inflated GAAP relative to adjusted. (primary release; 6-K)

Management cited AI, product innovation, and Asia expansion; Hong Kong is profitable, Singapore should turn profitable in FY26, and Vietnam is near break-even. No H2 numerical guide was given. (transcript) HUIZ spiked 177% premarket to $4.04, then closed +26% at $1.84 on 81.7 million ADS—about 94 times prior volume and 54% below the premarket peak.

Cluster A — “New-business and AI-efficiency inflection” (medium confidence). Investing.com synthesis and Sidoti analyst Aashi Shah's Q&A focused on first-year premiums, expense leverage, and productivity. Digital insurers could benefit if AI lowers service/acquisition cost while persistency holds. Revenue grew only 6%, adjusted profit is tiny, and AI attribution is management-supplied.

Cluster B — “International optionality with regulatory risk” (medium-low confidence). Citi's Amy Chen called performance resilient in Q&A but tested Hong Kong tax-policy risk. Overseas profitability can diversify China; enforcement or visitor-demand weakness could derail it.

Cluster C — “Fundamentals improved; the tape was microstructure” (medium confidence). Independent commentator Atul Ghandhi warned before the open that thin liquidity made quotes unreliable; the reversal supports that. (Ghandhi, Aug. 20)

Consensus/disagreement/resolver. Operating improvement is genuine; a triple-digit revaluation was not supported. The dispute is whether AI/overseas scale can turn slow revenue into margins. Report inference: useful insurtech operations, useless peer-valuation signal. H2 adjusted profit/cash flow, premiums, persistency, overseas profit, and normalized volume decide it.

16. Coty (COTY): stabilization without visibility

Release/call: August 19 about 4:30 p.m./8:00 a.m. ET Aug. 20. Impact: medium-high. Revenue was $1.269 billion versus $1.19 billion, but adjusted loss $0.02 versus $0.01 expected. Like-for-like sales fell 1%; Prestige fell 0.5% and Consumer Beauty 3%. Gross margin contracted 140 basis points to 60.9% and EBITDA 26%. FY FCF improved to $348 million, leverage was 3.4x, and no full-FY27 guidance was given. Q1 EPS $0.11–$0.13 was below roughly $0.14 consensus, with sales and EBITDA declines guided. (SEC-filed release; transcript) COTY fell 9.4%.

Cluster A — “No visible path to durable growth” (high confidence). RBC's Nik Modi downgraded to Sector Perform and cut his target to $3 from $8, citing early-stage execution, Gucci's exit, restructuring, and the Consumer Beauty review. Absent annual guidance, margin erosion, and sell-out underperformance support him; FCF, debt reduction, or faster U.S. recovery are counterevidence. (RBC, Aug. 20)

Cluster B — “Early stabilization, still show-me” (medium confidence). Same-day recap showed TD Cowen Hold/$2.90 and Morgan Stanley Equal Weight/$2.50; Q&A from Citi's Filippo Falorni, Evercore's Javier Escalante, Canaccord's Susan Anderson, and Deutsche Bank's Stephen Powers tested FY27, shelf space, mass beauty, and the ex-Gucci bridge. Revenue and cash improved; category underperformance blocks conviction. (recap)

Consensus/disagreement/resolver. Sequential stabilization, no proof of profitable growth. The dispute is whether launches and SKU reduction restore sell-out before Gucci dilution and restructuring. Report inference: execution/portfolio warning, not a broad beauty-demand collapse. Q1 sell-out, gross margin, shelf resets, review terms, annual guidance, and leverage decide it.

17. OSI Systems (OSIS): backlog could not excuse a back-end-loaded guide

Release/call: 4:02/4:30 p.m. ET. Impact: medium-high after hours. Revenue was $484.1 million versus $529.7 million, while adjusted EPS $3.78 versus $3.77 and operating margin +200 basis points. Middle East/site constraints deferred roughly $50 million of Security deliveries; Mexico declined $20 million. Security margin still reached 20.8%, Opto/Healthcare grew, backlog hit $1.9 billion, and Q4 operating cash was $182 million. FY27 guidance is $1.875–$1.930 billion revenue and $11.13–$11.49 EPS, weighted to the second half. (primary release; consensus/transcript) OSIS traded near $188.97, −13.4% after hours at 7:41 p.m.

Cluster A — “Timing deferral, not demand loss” (medium confidence). B. Riley's Josh Nichols, Citi's John Godyn, and Roth's Jeff Martin used Q&A to test whether delayed deliveries remain live and when they convert. Backlog, book-to-bill, and cash support a 6–18-month recovery; prolonged access constraints or another pushout would refute it.

Cluster B — “Security/RF offers multi-year optionality” (medium-high on pipeline, lower on timing). Oppenheimer's Christopher Glynn and Nichols focused on $285 million of CBP IDIQ ceilings and RF awards. Only about $21 million of firm CBP task orders entered backlog and most benefit is FY28+, so the opportunity cannot rescue near-term estimates.

Cluster C — “Cash/margins real, growth algorithm questioned” (medium confidence). CJS analyst Larry Solow challenged why EPS growth barely outruns sales despite buybacks; Glynn and Nichols tested margins/cash conversion. Investment and mix are management's answer, but the back-end cadence leaves risk.

Consensus/disagreement/resolver. Strong franchises, cash, and backlog; poor revenue/guide setup. The dispute is recoverable timing versus recurring conversion risk. Report inference: the after-hours decline priced conversion, not demand destruction. Middle East deliveries, CBP task orders, RF milestones, first-half cadence, margins, and Mexico cash decide it. No public post-print rating note was available; Q&A is labeled.

18. JD Sports Fashion (LSE: JD.): U.S. footwear warning, with timing caveats

Update: 2:00 a.m. ET; no call. Impact: medium-high global/sector. Organic sales fell 1.3% versus +0.5% Visible Alpha and like-for-like sales 3.1%. North America fell 4.5% organic/6.8% LFL, though excluding standalone Finish Line conversions organic sales fell only 1.0%. Footwear and promotions were weak; apparel/accessories, running, and newer styles did better. FY pretax-profit-before-adjusting-items guidance fell to £700–£800 million from £750–£850 million; FCF guidance stayed £460–£520 million. (JD Sports RNS; Visible Alpha) London shares fell 14.3%; Nike, more than 40% of JD revenue, fell about 2.1%, a channel co-move rather than proven causality. (Reuters)

Cluster A — “U.S. discretionary warning” (medium-high confidence). Wealth Club's Susannah Streeter saw darkening U.S. clouds as younger/lower-income buyers treat sneakers as discretionary. The one-to-two-quarter implication is negative for mall footwear, Nike wholesale, and DICK'S Foot Locker business. August back-to-school timing and Finish Line conversions are meaningful counterevidence. (Guardian, Aug. 20)

Cluster B — “Structural sneaker/product-cycle problem” (high confidence). Investec's Kate Calvert said shares need estimate stabilization, clean inventory, and better Nike momentum; GlobalData's Chloe Tedford-Jones and Guardian columnist Nils Pratley emphasized market maturation, innovation gaps, Nike dependence, and price. Promotions and downgrades can persist 6–18 months; better Nike launches, inventory, and running/new-style growth would refute it. (GlobalData; Pratley)

Cluster C — “Cash discipline cushions, not cures” (medium confidence). Pratley noted unchanged cash-flow guidance but viewed buybacks/cash preservation as no substitute for growth. Net cash limits balance-sheet risk; another guide cut still drives rerating.

Consensus/disagreement/resolver. The downgrade cycle is not demonstrably over. The dispute is temporary timing/conversion distortions versus saturation and brand innovation. Report inference: a real U.S. discretionary warning, less severe than the −6.8% headline. August sales, September 23 results, U.S. retail/labor data, margin, inventory, and Nike sell-through decide it. No call occurred; Investec was the only attributable sell-side view found.

19. Ross Stores (ROST): the tariff-adjusted beat was still exceptional

Release/call: 4:11/4:15 p.m. ET. Impact: high after hours. Sales rose 13.3% to $6.265 billion versus $6.15 billion FactSet and EPS was $2.66 versus $1.94. A $253 million tariff refund contributed $0.60, leaving roughly $2.06 normalized EPS, still 6% above consensus. Comps rose 10% on top of 2%, driven mostly by transactions. Refunds supplied 405 of 610 basis points of reported margin expansion; underlying margin still improved 205 basis points, above plan. (8-K; primary release)

Q3/Q4 comps are guided +6%–7%/+4%–5%; FY EPS is $8.61–$8.77 including the refund, or $8.01–$8.17 normalized. Management said new, recaptured, and more-frequent existing shoppers all contributed; access to better brands and closeouts improved. ROST ended extended trading at $248.75, +8.46% from the $229.34 regular close at 7:59:55 p.m.; TJX and BURL firmed. (company webcast; options context)

Cluster A — “Traffic/share gains have runway” (medium-high evidence). JPMorgan's Matthew Boss, BofA's Lorraine Hutchinson, and Jefferies' Corey Tarlowe used call questions to test exit momentum, initiatives, and customer composition. Broad transaction growth and still-early execution support multi-quarter gains; back-half deceleration and comparison/timing effects are risks.

Cluster B — “A vendor/brand flywheel” (medium evidence). Gordon Haskett's Chuck Grom, Citi's Paul Lejuez, and Evercore's Michael Binetti focused on vendor access and competition for buys. Growth, cleaner stores, and brand positioning can open more vendors; ample supply also benefits rivals and higher inventory can create markdown risk.

Cluster C — “Better-than-headline quality, not a clean $2.66” (high factual confidence). The refund explains most of the EPS surprise, but normalized EPS and margin still beat. The gain accelerating during the call indicates the market rewarded operations/guidance, not merely the refund.

Consensus/disagreement/resolver. Execution and traffic are real; public post-call notes were unavailable by 8 p.m. The dispute is a durable share-gain flywheel versus normalization in the second-half guide. Report inference: positive off-price and value-channel read-through, negative for full-price apparel. Q3 traffic, repeat rates, brand penetration, inventory turns, and underlying margin decide it.

20. Flowers Foods (FLO): price could no longer cover bread-volume erosion

Release/call: 4:10 p.m. ET; call August 21 at 8:30 a.m. ET. Impact: medium. Revenue fell 4% to $1.193 billion versus $1.23 billion and adjusted EPS to $0.21 versus $0.22. Price/mix added 1.8%, but volume fell 5.8%; branded volume fell 7.6%. EBITDA declined 19.2% and margin 180 basis points to 9.3%. Flowers cut every FY range: sales to $5.070–$5.142 billion, EBITDA to $453–$481 million, and EPS to $0.75–$0.85. (SEC-filed release; prepared remarks) Simple Mills was the countertrend, with retail sales +13%, cookies +39%, and crackers +8%. FLO closed regular trading −4.8% at $7.10, then fell another 4.2% to $6.80 by 8 p.m. (StockAnalysis/Cboe)

Cluster A — “Core category erosion is broadening” (medium confidence on opinion; high on facts). An early-August Jefferies Hold/target cut to $7 warned deterioration had broadened; today's volume, deleverage, and guide cuts validated it. Smaller packs, Nature's Own recovery, and $20 million annualized savings are the countercase. (Jefferies public summary)

Cluster B — “Premium diversification cushions, does not reverse” (medium confidence). Zacks' August 14 preview identified Simple Mills, keto/protein, and snacks as offsets to loaf and margin softness. Simple Mills confirmed the pocket; consolidated results show it is not large enough. Co-manufacturing and premium-consumer risk remain. (Zacks)

Consensus/disagreement/resolver. Legacy bread is deteriorating; the question is whether innovation and savings stabilize it before volume/mix worsen. Report inference: negative signal for price-sensitive packaged staples, healthier premium better-for-you demand. The August 21 Q&A, weekly share/volume, Nature's Own, Simple Mills velocity, and H2 EBITDA conversion decide it. No named same-day post-print opinion was public by cutoff; the call was unavailable.

6. Cross-event themes and notable contradictions

1. Physical scarcity beat financial signaling

Facts: observable Hormuz traffic remained severely restricted, Brent rose 2.4%, and the 10-year returned to about 4.70% even after Treasury enlarged its long-end buyback signal. (Reuters/Kpler; CNBC/Bessent; AP close) Attributed consensus: buybacks can improve liquidity; they cannot neutralize oil inflation, coupon/corporate supply, and fiscal credibility. Report inference: the market is demanding a lower realized barrel constraint or a fiscal/disinflation bridge before treating policy language as durable duration relief.

2. The economy looked stronger in surveys and weaker at the cash register

Facts: claims stayed low, the Philadelphia Fed reached 47.4, and LEI breadth turned positive, while Walmart comps slowed, AAP DIY fell low double digits, and JD Sports' North American LFL sales fell 6.8%. (DOL; Philadelphia Fed; Conference Board; Walmart; JD Sports) Report inference: business investment, manufacturing expectations, and low layoffs can coexist with squeezed discretionary budgets. That is a two-speed expansion, not a contradiction to be averaged into “mixed.”

3. Normalization, not the reported headline, set the equity multiple

Walmart's tariff refund, AAP's $0.31 refund contribution, NetEase's investment marks, Aegon's one-offs, Daqo's impairment, and Huize's share-compensation reconciliation all required a repeatable bridge. The tape punished the cases in which the normalized comp, margin, or guide missed, even where reported EPS beat. Report inference: in this regime, the market capitalizes recurring unit economics and cash conversion; it expenses windfalls and reversible marks immediately.

4. Backlog won only when conversion looked credible

Deere's construction book, Nordson's six-month backlog with improving lead times, and ScanSource's broad demand supported strong gains. Hovnanian's delivery/JV miss, OSI's Middle East pushouts, and LSI's low-margin legacy backlog showed the opposite: bookings without timing, price, or margin can become a liability. Report inference: “record backlog” is no longer sufficient disclosure; investors are underwriting age, cancellation, lead time, margin, and cash cadence.

5. China growth split into three investable categories

• Self-funding platform growth: Alibaba cloud and NetEase game gross margin showed genuine operating leverage, though Alibaba's capex and NetEase's marks complicated equity capture.
• Working-capital or activity growth: ATRenew, Futu, Atour, and Daqo grew through inventory, trading activity, unit expansion, or production that may not persist at the same cash return.
• Policy/microstructure optionality: Daqo's anti-involution reversal and Huize's extreme volume were real market events, but neither constituted clean proof of normalized earnings power.

Report inference: “China growth” was not one factor. The decisive split was cash-funded repeatability versus policy, inventory, or trading-beta dependence.

6. The day's sharpest contradictions—and their likely resolution

Contradiction	Why both sides can be true	Deciding evidence
Strong TIPS auction / rising nominal yields	Buyers accepted nearly 3% real yield; supply and inflation still lifted the nominal curve	September auctions, breakevens, dealer shares
Strong Philly Fed / softer current orders	Expectations, labor and capex surged; current orders/shipments eased	National PMI/ISM and durable goods
Walmart beat / 9% decline	Digital and profit engines worked; comps and forward EPS missed the valuation bar	Q3 ex-health traffic, rollback elasticity, Q4 timing
Daqo miss / stock gain	Current economics were below cash cost; policy made the survival option more valuable	Actual above-cost trades, inventory draw, enforcement
Alibaba profit miss / positive close	Capex consumed cash; cloud growth and margin validated a larger future profit pool	Cloud acceleration, utilization, FCF recovery
Treasury backstop / yield retracement	Liquidity tools are real; they do not retire debt or remove inflation	Buyback take-up, fiscal plan, refunding supply

7. Coverage audit

Inventory and calendar cross-check

The master inventory was built before event-level research, then each included item received a dedicated independent research pass. Source sets checked were:

• Macro calendars and primary schedules: DOL release calendar/UI claims, Philadelphia Fed survey calendar, Conference Board indicator calendar/technical notes, Treasury auction announcements/results, Treasury news releases/FiscalData, plus Reuters and AP daybooks/market wraps for unscheduled geopolitical and policy developments.
• Earnings calendars: Nasdaq's August 20 earnings API, Kiplinger/Briefing.com with Refinitiv estimates, TipRanks, StreetInsider, company investor-relations calendars, and SEC EDGAR 8-K/6-K submissions. Foreign local-date items were mapped to their U.S. ET release time.
• Primary company record: earnings releases, filings, presentations, webcast/event pages, and public transcripts where available. Vendor consensus was used only when the measure and unit could be reconciled; thin or inconsistent ADS/ordinary-share feeds are flagged.
• Opinion search: dated public sell-side summaries, company-call analyst Q&A, buy-side/strategist comments, and reputable media quoting named experts. No inaccessible note is represented as reviewed. A question is treated as an attention signal, never as a rating.

Borderline or erroneous calendar items excluded

Item	Decision and one-line reason
EIA weekly natural-gas storage (+15 Bcf)	Excluded: routine release with no identified outsized gas, utility, or cross-asset impact today.
Canaan (CAN) “Q2 earnings”	Excluded after primary-source correction: no August 20 release, call, or filing; IR results stop at Q1 and latest item was the August 17 July-mining update. CAN's +12.8% move tracked crypto beta, not earnings. (IR news; quarterly results; SEC July update)
Intuit, Workday, Medtronic	Excluded as stale/misdated calendar hits: official dates are August 25, August 27, and September 1, respectively. (Intuit IR; Workday IR; Medtronic)
Autohome (ATHM)	Reported, but its roughly −1.7% move and results produced no identified material sector or macro read-through beyond the included China consumer set.
PSEC, GRRR, YSG, DDL, KRKR, FLUX and other calendar microcaps	Excluded after results/move review: no durable index, sector, commodity, credit, or crowded-security read-through sufficient for “major.”
China LPR and Japan trade data dated August 20 locally	Excluded from the U.S. trading-date inventory because they occurred August 19 ET and were covered in the prior U.S.-date issue.

Availability and material gaps at cutoff

• Flowers Foods: the release is included; its Q&A call is scheduled for August 21 at 8:30 a.m. ET and therefore was unavailable. Any conclusion is pre-call and should be refreshed after management Q&A.
• JD Sports: the company explicitly held no call for this update; the next scheduled results are September 23.
• Macro attribution: claims and Philadelphia Fed arrived together at 8:30 a.m.; oil, fiscal supply, corporate issuance, and Walmart co-drove the session. Tick and closing moves are not falsely assigned to one catalyst.
• Hormuz measurement: Kpler's verified AIS crossings and U.S. officials' estimated protected barrels use different scopes; the report preserves the conflict.
• TIPS commentary: no named public post-auction strategist note was found; objective auction statistics, a labeled market-participant tape, and nearby contextual institutional views are used.
• Small-cap analyst coverage: public same-day notes were frequently sparse. In those sections, named Q&A, dated pre-print research, or explicitly secondary market commentary replaces a manufactured “Street view.”
• Consensus quality: Atour, Hovnanian, Huize, ATRenew, Daqo and some China ADR feeds had thin contributors or mixed ADS/ordinary-share units. Comparisons are qualified or marked unavailable rather than forced.
• Price timestamps: regular closes are labeled separately from after-hours snapshots. Foreign ADR/local-share reactions and WTI expiry effects are not conflated.

Coverage count

Included: six macro events/developments and 20 company releases, calls, or trading updates. The company set contains every verified mega-/large-cap item identified, both August 20 calls on material August 19 releases (Nordson and Coty), and smaller names admitted only for a large realized move or important sector signal. Explicit category zeros: no separate major central-bank rate decision, sovereign-rating action, material credit event, or U.S.-date global statistical release beyond the table qualified today.