Market Pulse · Daily Scan
Daily Market Pulse
Jul 30, 2026 · sentiment/flow window ≈ 48h
Pre-marketnot publishedPost-market
TLDR;
A near-total reversal of the prior session's hawkish-repricing selloff, driven by one earnings print. Microsoft's FQ4 - revenue $90.0B against an $87.6B estimate, Azure +43% in constant currency crossing $100B of annual revenue - supplied the demand proof point the market had been demanding, and the entire AI complex re-rated off it. The S&P recovered most of Wednesday's decline, the Nasdaq posted its best session in weeks and ended a six-day losing streak, and the SOX rallied roughly 8%, its biggest advance since April 2025. The structure of the day matters more than the magnitude: Microsoft and Meta reported the same night, both raised capex, and the market rewarded one and punished the other - Microsoft +15.51%, Meta -7.95%. The AI-capex revolt that defined the prior run did not end; it became DISCRIMINATING. Spending with a demonstrated revenue offset was bought, spending without one was sold. Three supporting forces: Samsung warned the memory shortage could persist into 2028 while posting a record Q2 operating profit up 1,813.8% y/y, rebutting the CXMT-oversupply panic; JPMorgan told clients hedge-fund deleveraging out of tech, chip and memory positions was essentially complete, having run faster than expected - corroborated by the disorderly unwind of Leopold Aschenbrenner's roughly $24B fund, which sold most of its listed book to Citadel after margin calls from Goldman, JPMorgan and Bank of America; and macro cooperated, with Q2 GDP at +1.5% against a +1.8% forecast and headline PCE easing to +3.7% y/y from +4.1%. September hike odds fell from roughly 80% to roughly 54%. Breadth: NARROW, and this is the single most under-appreciated fact of the session. The headline was a 1.66% S&P gain, but 23 of the Dow's 30 constituents FELL and the Dow was the laggard at +1.19%. Information Technology added roughly 5% - its largest one-day advance since April 2025 - while Salesforce fell 5.4% and Nike and Johnson & Johnson also declined. Shortly after the open only 243 S&P names were higher and only five of eleven sectors were positive. The Russell 2000 participated at +1.37%, so this was not purely a mega-cap event, but it was overwhelmingly a technology-and-semiconductor event: strip out Microsoft and the chip complex and the tape was close to flat. That is the mirror image of Jul 29, which was broad-based selling with the Russell falling LESS than the Dow..A · Daily Pulse
What moved — and why
▲ Rising — dailycomposite
| # | Ticker | Move | Driver | Score |
|---|---|---|---|---|
| 1 | MSFT | — | The single name that reframed the entire AI-capex debate. FQ4 revenue $90.0B beat the $87.6B estimate, Azure grew +43% in constant currency against a 40% estimate, and annual Azure revenue crossed $100B for the first time in fiscal 2026 with Copilot above 30M paid seats (up from 20M in April). Added roughly $450B of market capitalisation in one session - the largest single-day gain ever recorded for an individual stock - and its best day since 2008. The analytically important point is not the size of the move but its direction of causation: a day earlier Meta's capex guide was read as proof the AI buildout does not pay, and Microsoft's Azure number was read as proof it does. The entire semiconductor complex re-rated off this one print. Market cap not independently pulled post-move; unambiguously above threshold. ↔ deep-dive | 85 |
| 2 | SNDK | — | Largest single-session gain of any universe-passing name and the sharpest reversal in the artifact: -7.32% on Jul 29, then +26% on Jul 30. The catalyst is a supply statement, not a company print - Samsung warned the memory shortage could persist into 2028 while posting a record Q2 operating profit up 1,813.8% y/y on AI memory demand. That reframes the CXMT oversupply panic that drove the July liquidation. PUMP_RISK stays flagged from the prior run on flow characteristics - +/-15-26% daily swings and persistent Tier-3 crowding - but the flag is a position-sizing warning, not a fraud signal: the underlying business has real cash generation. Note the magnitude conflict below. pump risk | 83 |
| 3 | MU | — | Round-tripped the memory panic: -9.94% on Jul 29, +18% on Jul 30. Two forces converged - Samsung's memory-shortage-to-2028 warning restored the pricing-power narrative, and JPMorgan told clients that hedge-fund deleveraging out of tech, chip and memory positions was likely complete, having happened faster than expected. That second point matters more than the first: the July drawdown was substantially a forced-selling event, and the marginal seller being finished is a different fact from demand improving. Market cap carries an unresolved conflict inherited from the prior run; universe qualification is not in doubt at any of the candidate figures. ↔ deep-dive | 79 |
| 4 | LRCX | — | Cleanest catalyst-to-move ratio in the window and its best day since 1999. Record fiscal Q4: revenue $6.72B with adjusted EPS $1.82, revenue up 15% sequentially and 30% y/y, non-GAAP operating margin 38.4%, and a Q1 guide of $8.1B implying over 20% sequential growth. Jefferies raised its target to $335 from $315 on what it called a clean beat and raise. The read-through is the analytically valuable part: wafer-fab equipment is the capex layer furthest upstream, so a >20% sequential guide from Lam is a direct statement that fab spending is accelerating, not digesting - which is why AMAT, KLAC and the wider SOX complex traded off this print rather than their own. | 73 |
| 5 | AMD | — | Pure high-beta expression of the semiconductor snapback, up 13.00% with no company-specific news - the move is entirely Microsoft's Azure print transmitted through sector beta. Worth reading alongside its Track B position: AMD is Track B downside rank 8 on a relative-value case against Nvidia, and a 13% single-session rally on someone else's earnings makes that short more expensive, not less valid. Reports Aug 4 AMC, the first real test of whether its own numbers can support the re-rating. | 68 |
| 6 | EME | — | The most overlooked name on this list and the only non-semiconductor in the top six. Blowout Q2 beat with raised full-year guidance and strong project-backlog growth, up 18% on the session. EMCOR is the electrical and mechanical construction contractor that physically builds data centres, so it sits on the same AI-capex thesis as Vertiv and GE Vernova but is priced as an industrial rather than a technology name. Its rallying 18% on the same day the industrials sector was recovering from being the prior session's worst performer argues the move is company-specific. Market cap $41.07B is a May 2026 figure and therefore STALE and understated after the move; the name clears the threshold by a wide margin either way. | 67 |
| 7 | AMAT | — | Cleanest sympathy trade in the window: up 15% on Lam Research's guide with no print of its own. AMAT had fallen 8% or more alongside VRT, KLAC, MU and SMCI on Jul 29, so this is a same-week round trip in the semicap complex rather than a re-rating. The signal here is about the wafer-fab-equipment cycle, not about Applied Materials. Market cap not independently pulled; unambiguously above threshold as a mega-cap semicap name - flagged for form. | 66 |
| 8 | INTC | — | Up 11% on sector beta with no company catalyst, immediately after finishing July down 41.36% - its worst month since September 2000. The reflexivity is the point: a name that can fall 41% in a month and rise 11% in a day on someone else's earnings is being traded on positioning, not on 18A yields or foundry economics. This is the same observation that places INTC at Track B downside rank 2, and the Jul 30 rally makes that short more expensive without addressing the thesis. | 66 |
| 9 | MRVL | — | Custom-silicon and optical-interconnect beta, up 13% with no own catalyst. Relevant because Marvell sits in the same custom-ASIC lane as Broadcom (Track B upside rank 4): if hyperscaler capex holds, the merchant-versus-custom split matters, and Jul 30 repriced the whole lane at once without discriminating between them. Market cap not independently pulled; comfortably above threshold - flagged for form. | 62 |
| 10 | TSM | — | Up 7.6% - roughly the SOX move, and materially less than MU, AMD, INTC or AMAT. That relative lag is informative rather than disappointing: TSMC is the lowest-beta way to own the capex cycle, so it neither fell as far in the July liquidation nor bounced as hard. It is Track B upside rank 1 for exactly that reason, and the Jul 30 session is consistent with the thesis rather than a re-rating of it. | 60 |
▼ Falling — dailycomposite
| # | Ticker | Move | Driver | Score |
|---|---|---|---|---|
| 1 | ALNY | — | Largest single-session decline of any universe-passing name in the window by a wide margin, and the only large decliner not tied to the AI complex. Missed Q2 revenue expectations and cut full-year guidance. The context makes it worse than a single bad quarter: Alnylam was already down roughly 30% in 2026 against a broadly recovering biotech tape, and the Wainua Phase 3 failure had already raised questions about the wider RNAi-silencer franchise including nucresiran. A guidance cut on top of that reads as thesis erosion rather than timing. MARKET CAP CONFLICT: sources span $41.2-41.8B (MacroTrends Jul 6, Trading Economics July) against $59.3-60.1B (SimplyWall.st) - roughly a 45% spread that could not be reconciled, so no point estimate is published. Universe qualification is not in doubt at either figure. | 78 |
| 2 | META | — | The one large name that fell on a day everything else rallied, and the cleanest statement of the session's central divergence: Microsoft and Meta reported the same night, both raised capex, and the market rewarded one and punished the other. Q2 EPS of $6.18 missed the ~$7.22 estimate on higher legal and severance charges even as revenue grew 28% to $60.80B; free cash flow fell about 91%; 2026 capex was guided to $130-145B; and Q3 revenue was guided BELOW consensus - that last item is new versus the prior run. The most important disclosure is Tier 1 and was easy to miss: a Thursday securities filing put future lease obligations, largely tied to AI data centres, at $279B, a 53% jump from three months earlier. That is off-balance-sheet capex commitment growing faster than the guided number. Now on an 11-session losing streak, the longest on record. ↔ deep-dive | 78 |
| 3 | CVNA | — | Full-year earnings guidance of $2.7-3.0B against Deutsche Bank at $3.0-3.2B and Morgan Stanley at $4.45B. The spread between those two sell-side estimates is the story: a roughly 48% dispersion in analyst expectations for the same fiscal year means the Street has no shared model of this business, and the guide landed below even the conservative end. On a session when the S&P rose 1.66%, a 12% decline is a ~14-point relative move. MAGNITUDE CONFLICT: -10% premarket, -14% in Jul 29 after-hours, -12% at the close across sources; the close-basis figure is used. | 70 |
| 4 | QCOM | — | The counter-example that keeps the semiconductor rally honest. Qualcomm reported into the same tape that lifted every other chip name and fell anyway: revenue of $9.95B beat the $9.67B estimate but adjusted EPS of $2.21 slightly missed $2.23, margins thinned, and handset revenue fell 20% y/y. On a day the SOX rose about 8%, that is a 13-point relative underperformance. The distinction the tape is drawing is precise - chip names levered to AI infrastructure were bought, chip names levered to handsets were not, and being in the same index bought Qualcomm nothing. MAGNITUDE CONFLICT: -4% premarket, roughly -5% per the close-basis recap, and 'more than 7%' in a CNBC live-blog entry of uncertain timestamp; the close-basis figure is used. | 61 |
| 5 | LLY | — | Included because the absence of a catalyst is itself the datapoint. Lilly fell 4.55% on a session the S&P rose 1.66% - a ~6.2-point relative move - with no company news identified in any source consulted. The most likely mechanical explanation is rotation: healthcare and defensives funded the technology rally, having been the refuge during the prior week's semiconductor liquidation. That reading is an INFERENCE, not a documented cause, and the catalyst_unverified flag is set accordingly. LLY is Track B upside rank 8 and reports Aug 5 BMO, the largest single-name catalyst of that day. unverified | 56 |
B · One-Year Conviction
Asymmetric 12-month skew
VALUATION FIELDS THROUGHOUT TRACK B ARE AS OF THE JUL 29 CLOSE and do NOT reflect Jul 30 price action. This is deliberate: the underlying aggregator snapshots were pulled on 2026-07-30 reflecting the prior close, and re-deriving forward multiples from unverified post-move prices would be fabrication. The consequence must be read explicitly - MSFT's 20.03x forward and 24% PT upside are stale by a +15.51% session, MU's 5.15x by +18%, INTC's 54.78x by +11%, AMD's 48.20x by +13%, and META's 17.63x by -7.95% in the other direction. Each affected entry carries a note field saying so. Tier-1 primary sources verified directly across this and the prior run: Micron FQ3-2026 via SEC EDGAR, NVIDIA Q1 FY2027 IR, GE Vernova Q2-2026 IR, Alphabet Q2-2026 slides, FOMC statement, FactSet Earnings Insight, and the Meta lease-obligation disclosure. Sell-side notes were accessed only as quoted excerpts inside Tier-2 media, never as primary notes.
▲ Upside skew — 1yrconviction
| Ticker | Thesis | Key risk | Score |
|---|---|---|---|
| TSM | Unchanged and reinforced. TSMC is the unavoidable chokepoint of the AI buildout and, unlike its customers, converts the capex boom into cash rather than consuming it. Full-year 2026 revenue growth guidance above 40% with capex lifted to $64B, and brokerages flagging EUV-driven capex and 2027 price hikes. Jul 30 added corroboration from an unexpected direction: Lam Research guided to over 20% sequential revenue growth, which is a direct read on the fab spending TSMC commands. Sentiment nudged 78 to 80 on that read-through - a documented second-order datapoint, not score drift. | Taiwan geopolitical tail risk is the entire bear case and is not diversifiable; Customer concentration in Nvidia; Rising capex intensity compresses FCF conversion; A 2027 hyperscaler capex cut hits with a 2-3 quarter lag | 87 |
| NVDA | Unchanged thesis, lowered sentiment. Q1 FY2027 (Tier 1, company IR): revenue $81.6B +85% y/y, data center $75.2B +92%, non-GAAP gross margin 75.0%, Q2 guide $91.0B +/-2% while ASSUMING ZERO China datacenter compute revenue, plus an $80B buyback authorisation. At 19.6x forward against 85% growth the market has already discounted a substantial capex deceleration. TWO NEW ADVERSE SENTIMENT DATAPOINTS on Jul 30, both Tier 2 and both worth taking seriously without over-weighting: Michael Burry disclosed shorting Nvidia and buying puts, arguing its future revenues are majority financed in a circular arrangement, and Mark Cuban compared the AI economy to the dot-com bubble. Separately, NVDA rose only 2.65% on a day the SOX gained about 8% - the weakest participation of any major semiconductor name. Sentiment cut 70 to 66 on that combination. | The circular-financing critique is now being expressed by a named investor with capital behind it, not just commentators; Memory input-cost inflation is hitting the bill of materials - memory is now 29% of total cost against a stated ~20% target; $410.6M of insider selling over three months with zero buys; Custom-ASIC share loss to Broadcom, Google TPU and Meta MTIA; Beta of 2.21 means it is the wrong vehicle if the capex thesis is wrong | 84 |
| MSFT | The thesis was that Microsoft is the only hyperscaler demonstrably MONETIZING AI at scale while still generating $67B of free cash flow. On Jul 30 the market agreed, violently. FQ4: revenue $90.01B beat $87.6B, Intelligent Cloud $39.31B +32%, Azure +43% in constant currency against a 40% estimate and crossing $100B of annual revenue for the first time, commercial backlog $678B +84%, Copilot above 30M paid seats versus 20M in April, operating margin 45.6%. The stock added roughly $450B of market value, the largest single-day gain ever recorded for any stock. Sub-scores moved in BOTH directions and that is deliberate: estimate_revisions 84 to 88 and sentiment 82 to 86 on confirmed acceleration, structural 88 to 90 on the Azure milestone, but fundamental_valuation CUT 80 to 76 because a 15.5% move consumed a meaningful share of the 24% consensus upside in a single session. A thesis being validated makes a position less cheap, not more attractive. ↔ deep-dive | A 15.51% single-session move front-loads a large share of the 12-month case; Capex up 84.4% y/y in a recent quarter and guided higher again; Azure growth is materially OpenAI-linked, concentrating exposure to one counterparty's funding; Options underpriced this print by roughly 2x (7.0-7.3% implied vs 15.51% realized), so the market's read on Microsoft is unstable in both directions | 84 |
| AVGO | Carried unchanged - no new company-specific information in the window. The custom-ASIC leg remains the structural AI trade: every hyperscaler dollar spent reducing Nvidia dependence is by definition spent with Broadcom. Q1 FY2026 AI semiconductor revenue $8.4B +106% y/y on total revenue $19.31B +29%, with a $73B AI backlog deliverable over 18 months. VMware software mix cushions semiconductor cyclicality. Does not report until Sep 3, outside the horizon, so there is no near-term event to reprice it. | Customer concentration (Google, Meta, OpenAI); The '$100B AI revenue in 2027' figure is a CEO statement on a call, NOT formal guidance; Trailing 61.6x reflects acquisition amortisation so GAAP optics stay poor and screen badly; No catalyst inside the horizon means it re-rates only on sector beta | 81 |
| MU | STILL SPLIT CONVICTION, but the bear leg weakened materially on Jul 30 and the sub-scores say so. The single most important new datapoint in this entire run: Samsung warned the memory shortage could persist into 2028, while posting a record Q2 operating profit up 1,813.8% y/y on AI memory demand. That is the largest memory supplier in the world stating that supply stays tight for two more years - a direct rebuttal of the CXMT-oversupply thesis that drove the July liquidation. Structural_thesis raised 62 to 70 and sentiment 50 to 58 on that plus JPMorgan's assessment that hedge-fund deleveraging in the complex is essentially complete. fundamental_valuation CUT 92 to 88 after an 18% single-session move. The bear case is not dead: an 85% gross margin remains roughly 25 points above any historically sustainable level, and customer inventories at 7-9 weeks still sit against an 8-week warning threshold and a 10-week exit trigger. Bull: FQ3 (Tier 1, SEC-filed) revenue $41.46B against $9.30B a year earlier, GAAP diluted EPS $24.67, operating cash flow $25.39B, FQ4 guidance $50.0B +/-$1.0B at ~86% gross margin, on 5.15x forward earnings. ↔ deep-dive | Peak-cycle earnings at a trough multiple remains the textbook value trap in semis - Samsung's warning changes the timing, not the existence, of the cycle; Customer inventories 7-9 weeks vs an 8-week warning threshold; A supplier forecasting a two-year shortage has an obvious incentive to do so; Highest-variance name on either list - the 12-month distribution is genuinely bimodal | 79 |
| GEV | Carried unchanged. Electricity, not silicon, is the binding constraint on AI, and GEV sells the scarcest physical asset in the buildout - gas turbine slots - into a multi-year sold-out book. Q2 2026 (Tier 1, company release): orders $24.2B +88% organic, backlog $176B, gas power equipment backlog 100GW to 116GW, revenue $11.1B +22%, adjusted EBITDA margin +340bps organic to 11.3%, free cash flow $5.1B in the quarter - more than all of 2025. FY26 FCF guidance raised from $6.5-7.5B to $11.5-12.5B. Microsoft and Amazon both reaffirming aggressive capex on Jul 30 is supportive at the margin but does not change a book that is already sold out. | Highest headline multiple on this list at ~52x forward; Wind segment remains a drag; Turbine slot reservations are not firm revenue; A one-year New York State moratorium on 50MW+ data centres shows regulatory bottlenecks are now real | 76 |
| GOOGL | Best full-stack AI position in the world - TPUs, models, distribution and cloud - and the only Mag-7 name whose REVENUE growth accelerated on the back of AI. Q2 2026: revenue $119.8B +24%, operating income $40.8B +30% at a 34.0% margin, Google Cloud $24.8B +82% with cloud operating margin expanding to 35.6% from 20.7%, and a $514B cloud backlog. Sentiment raised 55 to 60 on a specific, documented Jul 30 development: Alphabet stabilised in the $335-342 range, extending its recovery from the post-earnings selloff, and both Microsoft's Azure beat and Amazon's AWS beat validated the cloud-capex-pays-off case that Alphabet's own negative free cash flow print had called into question two weeks earlier. The capex risk is unchanged; the market's willingness to underwrite it is not. | 2026 capex raised to $195-205B; Q2 free cash flow -$5.9B, the first negative quarter since the 2004 IPO; buybacks suspended; Street models FCF going from +$13B in 2026 to -$13B in 2027; EARNINGS QUALITY: Q2 diluted EPS of $9.11 was materially inflated by unrealised equity-securities gains and is excluded from the valuation sub-score; Antitrust remedies remain live | 74 |
| LLY | Clear structural winner of the GLP-1/obesity duopoly, and the contrast is quantitative rather than narrative: Lilly is compounding roughly 29% EPS while Novo Nordisk has guided 2026 to a 5-13% constant-currency SALES DECLINE. Oral orforglipron extends the franchise past injectable supply constraints. Included as much for portfolio construction as for absolute upside - it is the only top-10 upside name with no exposure to the hyperscaler capex question, which matters more than usual in a week when a single Azure number moved nine of the other ten. fundamental_valuation raised 60 to 62 on the -4.55% Jul 30 move, sentiment cut 74 to 70 on that same unexplained underperformance. Net effect on rank: swaps with GOOGL, 7 to 8. Reports Aug 5 BMO. | Consensus PT implies only ~11% upside - the easy multiple expansion is behind it; Fell 4.55% on Jul 30 with no identified catalyst, which is itself an unresolved question ahead of the Aug 5 print; US drug pricing / most-favored-nation policy risk; 34x forward leaves no room for a manufacturing stumble | 74 |
| META | CONTRARIAN AND DETERIORATING - held, but cut hard, and the honest read is that this position is now the weakest on the upside list. The valuation case is stronger than ever: below 17.6x forward before a further 7.95% decline, for a business growing revenue 28%. The capital-allocation case got materially worse on new evidence. Two items are new versus the prior run and both are adverse. First, Q3 revenue was guided BELOW consensus - the de-rating is no longer purely a capex-ROI question, the top line is now implicated. Second, and more important, a Thursday securities filing (Tier 1) disclosed future lease obligations, largely AI-data-centre-related, of $279B, up 53% in three months. That is a commitment growing far faster than the $130-145B guided capex figure and it sits largely outside the headline number. estimate_revisions cut 45 to 38, structural 70 to 66, sentiment 40 to 35; fundamental_valuation raised 90 to 92. Composite falls 66 to 63. THE POSITION IS RETAINED under no-churn discipline because the valuation leg is intact and no promotion or invalidation condition has been met - but this is now flagged watch_for_invalidation, and one more quarter of below-consensus revenue guidance should retire it rather than re-score it. ↔ deep-dive | The structural flaw: unlike MSFT, GOOGL and AMZN, Meta has NO external cloud revenue line to monetise the buildout; $279B of future lease obligations, +53% q/q, disclosed in a Thursday SEC filing - off the headline capex number; Q3 revenue guided below consensus - the first sign the de-rating is not purely about spending; 11-session losing streak, the longest on record; free cash flow down ~91% y/y; Widest gap on the list between the valuation case (92) and the revisions case (38) | 63 |
| VRT | Held at the bottom of the list, sentiment cut again. Pure play on data-centre thermal management and power distribution - the physical bottleneck layer - where rack densities from Rubin-class systems make liquid cooling non-optional. The problem is that the options market has not forgiven the Q2 revenue miss: Market Rebellion's Jul 30 pre-market report put VRT 30-day IV at 73 against a 52-week range of 45-85 with call:put skewed 1:1.4 toward PUTS on 104,000 contracts. That is persistent bearish positioning two sessions after the -17.26% print, on a day the entire AI-infrastructure complex rallied hard. Sentiment cut 45 to 40. Ranked last because it is the only name on this list where in-window evidence actively contradicts the thesis. | Options skew stayed 1:1.4 toward puts on 104K contracts even during a session when its whole sector rallied; Highest beta on this list to any hyperscaler capex cut; Competitive entry from Schneider and Eaton; The Q2 2026 revenue miss ($3.27B vs $3.38B) is exactly the kind of execution slip a 29x forward multiple cannot absorb repeatedly | 62 |
▼ Downside skew — 1yrconviction
| Ticker | Thesis | Key risk | Score |
|---|---|---|---|
| ORCL | Carried unchanged and still the highest-conviction downside. The only name on either list where the risk is solvency-adjacent rather than valuation. Oracle levered its balance sheet to fund a roughly $300B OpenAI contract and the credit market has repriced it: CDS spreads at a record 198.23bp and an S&P downgrade to BBB- - one notch above junk, which would force selling by investment-grade-mandated funds. Largest non-financial corporate borrower in Bloomberg's US high-grade index. Total liabilities $218.7B against $43.06B of equity, with ~$40B of further financing planned for FY2027. IMPORTANT AND UNCOMFORTABLE: the Jul 30 tape cuts against this position. Microsoft and Amazon both validated cloud-capex monetisation and ORCL appeared on Market Rebellion's popular-stock and active-options lists in both Jul 30 reports. Sub-scores are held rather than cut because the thesis is a CREDIT thesis that resolves on refinancing dates, and a good week for cloud demand does not change a funding structure - but this is the argument to watch. | THE HONEST COUNTER-ARGUMENT: the stock is already -49%, 14.6x forward is not demanding, and the 3-year EPS growth forecast is 27.37%; Jul 30 cloud-demand validation from MSFT and AMZN is a direct headwind to the demand leg of the bear case; Any OpenAI equity or financing event that de-risks the counterparty; A successful $40B raise at reasonable spreads would collapse the CDS and the thesis with it | 78 |
| INTC | Thesis intact, entry improved, timing worse. A 4.5x move in twelve months on a company still losing $11B a year, driven by POLICY and PARTNERSHIP headlines rather than profits - $8.9B of CHIPS funding converted to a government equity stake, a $5B Nvidia investment, and an Apple chip-making agreement. At 54.8x forward it is the most expensive large-cap semi on either list by a wide margin with the weakest competitive position and no trailing earnings at all. Jul 30 sharpened rather than weakened the reflexivity argument: INTC finished July -41.36%, its worst month since September 2000, then rose 11% the next session on Microsoft's earnings. A name that moves like that on someone else's news is trading on positioning. fundamental_valuation raised 90 to 92 and sentiment 65 to 68 accordingly. | A sovereign/strategic backstop makes a fundamental short genuinely dangerous - the US government is now an equity holder; The Jul 30 +11% shows how violently this name reprices on sector beta alone; 18A/14A yield news is binary and headline-driven; Consensus PT of $115.27 sits 26% ABOVE the Jul 29 spot - the Street is not on this side of the trade | 76 |
| AAPL | THE THESIS WAS CONFIRMED BY THE PRINT - this is the largest justified sub-score move in this run, and it moves AAPL from downside rank 7 to rank 3. The prior run argued Apple is the clearest AI laggard among mega-caps, trading at 36.6x forward for ~11% growth with a consensus target BELOW spot. FQ3 reported Jul 30 AMC delivered exactly the split that thesis predicted: revenue $109.4B (+16% y/y) and EPS $2.02 (+29%) both BEAT, but Services revenue of $30.7B missed the ~$31.2B estimate and Greater China again fell short. Services is the high-multiple, AI-adjacent segment that justifies the premium; hardware is the low-multiple segment that beat. A print that beats on the cheap line and misses on the expensive line is a de-rating event even when the headline is green. Compounding it: this was Tim Cook's final earnings call before John Ternus succeeds him in September - a CEO transition executing into an unresolved AI strategy. estimate_revisions 62 to 72, structural 72 to 76, sentiment 68 to 72. ↔ deep-dive | $123.3B of TTM free cash flow and the largest buyback in history put a hard floor under it; The headline beat on revenue and EPS gives management a defensible quarter to point at; Any credible AI/Siri launch closes the narrative gap instantly; The pre-print options market priced only a ~5% move (straddle 340) and 1.2:1 calls - the crowd was not positioned for this and may not chase it; A new CEO has an incentive to reset expectations early, which can front-load bad news and clear the path | 75 |
| NVO | Carried unchanged. Still the cleanest VERIFIED downward-revision case on either list. The company guided 2026 to a full-year sales decline of 5-13% at constant currency against 10% growth in 2025, driven by US market-access and demand pressure, with the CEO warning it will get worse before it gets better. Novo has also cut US GLP-1 prices by up to 70%, defending volume at the cost of mix. The sell-side has not finished cutting: consensus PT of $47.23 sits 8.4% BELOW spot - one of only two names in this artifact where that is true, the other being Apple. Reports Aug 5, inside the horizon. | A trailing P/E of 12.1 is genuinely cheap and limits downside velocity; A price-cut-driven volume inflection, or a strong oral launch, could stabilise the story; Already -25% over twelve months, so a meaningful portion is priced; Reports Aug 5 - a single print can resolve this in either direction inside the horizon | 75 |
| MRK | Carried unchanged. Merck has rallied 57% INTO the Keytruda loss of exclusivity, and the deterioration is already visible in aggregate estimates: FactSet attributes the Health Care sector's Q2 downward revision (-8.9% to -17.8%) specifically to Merck and Gilead. The company has announced $3B of annual cost cuts ahead of the cliff - a defensive signal, not an offensive one. Paying 21x forward for 5.6% EPS growth against a patent cliff is the worst growth-adjusted multiple among mega-cap pharma. Fell only 0.44% on Jul 30, essentially unmoved by the tape. Reports Aug 4 BMO with consensus EPS of -$0.26. | Pipeline or M&A can re-rate it - the company is actively acquisitive; Subcutaneous Keytruda conversion could materially blunt the cliff; a KEYTRUDA QLEX sNDA has a PDUFA date of Aug 17; A 2.6% dividend limits downside velocity; Consensus PT is still above spot, if only by 4.2% | 73 |
| CRWV | HELD BUT CUT - the invalidation watch flagged in the prior run is now the live issue. The balance-sheet case is unchanged and severe: $35B of debt against $40B of equity value funding depreciating GPUs, -$10.6B of free cash flow, Altman Z-Score 0.36 and Piotroski F-Score 2, both in elevated-distress territory. The thesis is that this is the first place a funding-market tightening shows up. What changed on Jul 30 is the demand leg, and it changed against the short: Microsoft's Azure beat and Amazon's AWS beat with capex hiked to $220B both argue neocloud capacity gets absorbed rather than stranded, and CRWV appeared on the popular-stock and active-options lists in both Jul 30 reports. structural_thesis cut 82 to 78 and sentiment 60 to 55. Composite falls 75 to 73, moving it from rank 4 to rank 6. If the Aug 4-5 hyperscaler-adjacent cohort confirms capex acceleration again, this position should be retired rather than re-scored - the standing watch_for_invalidation flag is maintained and escalated. | INVALIDATION RISK IS NOW THE PRIMARY RISK: two hyperscalers validated capex monetisation inside the window; Contracted revenue with investment-grade counterparties is real and materially de-risks the near term; Nvidia's strategic support is an explicit backstop; Borrow cost is likely punitive and the -32% move means much is already priced; Consensus PT of $138.03 is 87% above spot | 73 |
| TSLA | Carried unchanged. $1.22 trillion of market capitalisation resting on $3.81B of earnings. In a regime where the Fed's next move is still priced toward a hike - prediction markets moved from roughly 80% to 54% probability of a September hike across the Jul 29-30 window, so the pressure eased but did not reverse - Tesla carries the largest pure discount-rate exposure of any US mega-cap. The underlying auto business is a 14%-growth, 3.7%-net-margin manufacturer; the entire valuation is optionality on robotaxi and Optimus with no verifiable revenue line. Capex is up 142% to ~$25B and free cash flow is negative. TSLA rose on Jul 30 per an intraday-only source; no close-basis figure was confirmed, so no move is recorded and no sub-score is adjusted on it. | ESTIMATE-REVISION DIRECTION REMAINS UNVERIFIED for TSLA - the sub-score is set to neutral 50 rather than guessed; September hike odds fell from ~80% to ~54% across the window, which relieves the discount-rate leg of the thesis; Persistent retail bid; TSLA remained on the active-options list both sessions; Any credible autonomy milestone re-rates it violently; Consensus PT of $398.30 is still 29% ABOVE spot | 72 |
| AMD | Relative-value short, entry improved, conviction unchanged. AMD trades at roughly 2.5x Nvidia's forward multiple while being the structurally weaker accelerator franchise - squeezed from above by Nvidia's ecosystem and from below by Broadcom, Google TPU and Meta MTIA custom silicon. It faces the same memory-cost inflation now hitting Nvidia's bill of materials WITHOUT Nvidia's pricing power, and Samsung's Jul 30 warning that the memory shortage persists into 2028 makes that input-cost problem worse, not better, for the weaker buyer. AMD rose 13.00% on Jul 30 on Microsoft's earnings rather than its own, which improves the entry: fundamental_valuation raised 80 to 84, sentiment cut 55 to 50. Reports Aug 4 AMC - the first test of whether its own numbers support a 174% twelve-month move. | Genuine second-source demand from hyperscalers who want a Nvidia alternative is real and structural; Server CPU share gains against Intel are real and margin-accretive; A 46.6% revenue growth forecast is not fantasy; Reports Aug 4 AMC - a strong print inside the horizon invalidates quickly; This is a relative-value short (vs NVDA) far more than an absolute one | 67 |
| PLTR | Carried unchanged. After a 22% decline the stock still trades at roughly 56x SALES. The balance sheet is pristine and the growth is real, so this is explicitly NOT a business-quality short - it is a pure multiple-compression short. At 56x sales a 53% grower needs roughly five years of flawless execution just to reach a normal software multiple. Notably PLTR fell 0.6% on a session the Nasdaq rose 2.78% - it did not participate in the risk-on reversal at all, which is the first tape evidence in a while that the retail bid may be thinning. Reports Aug 3 AMC, the marquee report of that day. NOTE ON SOURCE DISCIPLINE: several widely used data providers list the Q2 date as August 2, a Sunday; the company's own release says August 3, and the issuer is the authority. retail-heavy | Government and defense budget tailwind ($1.5T FY27 request); Extremely strong retail base cuts both ways; Crowded short; Reports Aug 3 AMC - highest single-name event risk to this position inside the horizon; Consensus PT of $182.20 is 49% above spot | 67 |
| WDAY | Carried unchanged. The cleanest pure-play on the seat-based-SaaS de-rating. HCM headcount licensing is the most literally AI-exposed pricing model in enterprise software: if AI reduces headcount, Workday's billable units shrink mechanically. Morgan Stanley initiated at Underweight with a $145 target on 2026-07-21 - below spot - acknowledging Workday has the strongest moat in enterprise software while concluding its AI initiatives are unlikely to generate meaningful growth acceleration in the near term. Ranked last on this list for the reason stated in the audit note below. | 15x forward is not expensive; Consensus is still Buy across 41 analysts with a PT above spot; Morgan Stanley itself concedes the moat; AUDIT NOTE: the same firm published in March 2026 that AI-disruption fears in software are overblown, then downgraded ADBE and initiated WDAY at Underweight in July. That internal inconsistency is why this cohort ranks in the bottom third rather than the top | 66 |
C · Options Heat
Where the premium is paying up
| Ticker | Signal | Read | Implication | Lag |
|---|---|---|---|---|
| MSFT | vol_oi | Pre-print weekly IV 111 vs August 48 (2.06x); straddle 392.50 pricing ~7%; call:put 2.1-2.2:1. Post-print 322,000 calls in the first 30 minutes = ~8x average intraday volume, most popular Aug 450 call. Realised +15.51%. | The most important options fact of the window: the market UNDERPRICED this event by roughly 2x. A 7% implied move against a 15.51% realised move in the largest company in the S&P is a warning that event pricing across this earnings season is not calibrated to the dispersion actually being realised. | Pre-print from Market Rebellion Jul 29 (intraday). Post-print from Schaeffer's published Jul 30. Broker-desk and aggregator snapshots, NOT exchange data. No bid/ask side tagging, so the 2.1:1 call ratio does not distinguish buying from overwriting. |
| META | vol_oi | Pre-print weekly IV 128 vs August 54 (2.04x); straddle 592.50 pricing ~9%; call:put 2.0-2.4:1 in the 590 calls. Post-print ~70,000 puts = ~4x average, led by the 7/31 500-strike put, with the Sep 460 put showing sell-to-open. Realised -7.95%. META then appears in the Jul 30 mid-session DECREASING-IV list. | Bearish, and the one large-cap event this window that the options market priced roughly correctly. The contrarian wrinkle survives: sell-to-open in the September 460 put means someone is writing downside into the decline. | Same sourcing as MSFT. Post-print figures published Jul 30. The decreasing-IV listing is a mid-session (11:16 ET) snapshot, intraday and non-final. |
| VRT | pc_skew | 30-day IV 73 against a 52-week range of 45-85; call:put 1:1.4 toward PUTS on 104,000 contracts (Jul 30 pre-market). | Bearish and persistent. This is the only name in the dataset carrying a clear directional put skew two sessions AFTER its event, and it did so on a day its entire sector rallied. Drives the Track B upside sentiment cut on VRT. | Market Rebellion pre-market Jul 30, reflecting the Jul 29 close. Broker-desk snapshot, not exchange data. Volume ratios do NOT identify buy versus sell side - a 1:1.4 put ratio is equally consistent with protective buying and with put-selling. |
| AAPL | iv_spike | Jul 31 weekly 340 straddle priced for a ~5% move with call:put 1.2:1 (pre-market); weekly IV 76 vs August 31 against a 52-week range of 18-33, call:put flat 1:1 (mid-session). Realised roughly -6.6% after hours. | The options market priced the SMALLEST relative event premium of the mega-cap slate and was wrong-footed again - realised exceeded implied. Weekly IV at 76 against an annual band topping out at 33 shows the event premium was real in absolute terms; it was simply insufficient. | Market Rebellion pre-market and mid-session Jul 30. Mid-session figures are intraday and non-final. The realised move occurred after the close and is a Jul 31 price event, outside the Track A window. |
| AMZN | iv_spike | Jul 31 weekly 227 straddle priced for a ~7.5% move, call:put 1.6:1 (pre-market); weekly IV ~150 vs August 50 against a 52-week range of 23-50, call:put 1.5:1 (mid-session). Realised +6.97% after hours to $251.92. | Correctly priced, and the cleanest example in the window of a 3x term-structure ratio resolving inside the implied. Q2 revenue $200.6B beat $196.5B, up 20% y/y, with 2026 capex hiked to $220B - the AWS read-through that reinforced the Microsoft signal. | Market Rebellion pre-market and mid-session Jul 30; the mid-session weekly IV figure contains an apparent typographical error in the source ('at 15, 0August is at 50') and is read as ~150 vs 50. Flagged rather than silently corrected. Realised move is a Jul 31 price event. |
| XOM | pc_skew | Jul 31 weekly 155 straddle priced for a ~3.5% move; call:put 1.7:1 pre-market rising to 2.1:1 mid-session; weekly IV 58 vs August 33 against a 52-week range of 18-37. | Bullish tilt that BUILT through the session - the call ratio rose from 1.7:1 to 2.1:1 intraday even as crude fell 0.7%. Positioning into the Jul 31 BMO print is leaning long despite the spot move going the other way. Weekly IV at 58 against an annual top of 37 is genuine event premium. | Market Rebellion pre-market (06:03 ET) and mid-session (11:16 ET) Jul 30. Broker-desk snapshots; the mid-session figure is intraday and non-final. Volume ratios carry no side tagging. |
| CVX | iv_spike | Jul 31 weekly 192.50 straddle priced for a ~3.5% move; call:put 1.1:1 both pre-market and mid-session; weekly IV 65 vs August 31 against a 52-week range of 18-33. | Neutral. Same event premium as XOM by term structure (2.1x) but with a flat skew - the options market has a volatility view on Chevron and a directional view on Exxon. Both report Jul 31 BMO with OPEC+ meeting Aug 2. | Market Rebellion pre-market and mid-session Jul 30. Broker-desk snapshots, not exchange data. |
| MU | iv_spike | 30-day IV 97 against a 52-week range of 40-108 with call:put balanced at 1.1:1 (as of the Jul 28 close). MU was the FIRST name listed under 'popular stocks with increasing option volume' in both the Jul 30 pre-market and mid-session reports. | Two-way hedging and gamma demand at near-record vol rather than directional conviction - consistent with forced de-risking followed by forced re-risking. The most persistently active single name in the window across both sessions. | IV level is from the Jul 29 pre-market report and therefore reflects the JUL 28 close - a two-session lag by the time of RUN_DATE. No updated name-level IV for MU was published on Jul 30 in either report. No exchange vol/OI table was retrievable. |
| SNDK | iv_spike | 30-day IV 143 against a 52-week range of 44-163, call:put flat 1:1 (as of the Jul 28 close). Appeared on both the popular-stock and active-options lists in both Jul 30 reports. | Near-record implied vol with zero skew - a pure volatility event, which the +26% Jul 30 session against -7.32% on Jul 29 confirms. Reports fiscal Q4/FY26 Aug 5 AMC, so premium should persist through the horizon. | Same two-session lag as MU: the IV level reflects the Jul 28 close. No updated name-level figure published Jul 30. |
| ALNY | vol_oi | Appeared simultaneously on the 'increasing unusual option volume' and 'increasing unusual CALL volume' lists in the Jul 30 mid-session report, on a session the stock fell 29.4%. | GENUINELY AMBIGUOUS and included as an interpretive caution rather than a signal. Unusual call volume during a 29% decline is equally consistent with dip-buying, with closing of short-call hedges, and with put-side spread construction. Without side tagging it cannot be resolved, and it should not be read as a bullish reversal signal. | Market Rebellion mid-session Jul 30 (11:16 ET), intraday and non-final. No IV level, term structure or call:put ratio was published for ALNY. No free source tags bid versus ask. |
| TER | pc_skew | 30-day IV 102 against a 52-week range of 35-118; call:put 1:1.2 toward puts with the share price up 8.1% (Jul 30 pre-market). | Bearish tilt into strength. Teradyne rallied 8.1% with the semicap complex while its options skewed toward puts - a positioning divergence worth noting given TER was among the names down 8%+ on Jul 29. One of very few name-level skews published for the window. | Market Rebellion pre-market Jul 30, reflecting the Jul 29 close. Broker-desk snapshot; the quoted share-price move is the Jul 29 print, not Jul 30. |
| FXY_YEN_COMPLEX | iv_spike | FXY (Japanese yen ETF) was one of only FOUR names on the Jul 30 mid-session increasing-IV list (VCX, ANF, ULTA, FXY), alongside a roughly 2.9% USD/JPY move below 160. | Corroborates suspected Bank of Japan intervention through an independent channel. Matters for equities because a yen-carry unwind is a documented cross-asset transmission mechanism, and the BoJ's scheduled rate decision falls Jul 31, inside the horizon. ETF excluded from ranked lists by universe filter. | Market Rebellion mid-session Jul 30, intraday and non-final. IV direction only - no level, term structure or skew was published. |
| POST_EVENT_IV_CRUSH_COHORT | iv_spike | Names appearing on the Jul 30 mid-session DECREASING implied-volatility list that had just reported: META, CVNA, FTNT, CMG, CROX, SOFI, BSX, plus MSFU (2x MSFT ETF). | Included as a methodology observation. This cohort is the mechanical signature of event resolution, not a directional signal, and it spans winners and losers indiscriminately - META fell 7.95% and FTNT rose while both crushed. Any model reading falling IV as bearish will misclassify half of this list. | Market Rebellion mid-session Jul 30 (11:16 ET), intraday and non-final. Direction of IV change only; no levels published for these names. |
| CBOE_EQUITY_PC | pc_skew | NOT OBTAINED for either Jul 29 or Jul 30. | No conclusion drawn. Recorded so the gap is visible rather than silently absent. Given the VIX fell roughly 17% on Jul 30, the total put/call almost certainly declined - but that is an INFERENCE and is not published as a datum. | The official CBOE daily statistics page returned navigation content only in the prior run and was not re-attempted successfully this run; YCharts was stale to Jul 10 (equity) and Jul 2 (total) at last check. A third aggregator running roughly 2x the CBOE historical band was previously discarded as methodologically incompatible and remains excluded. |
D · Catalyst Calendar
Upcoming events
| Date | Event | Tickers | IV move |
|---|---|---|---|
| 2026-07-31 | Employment Cost Index Q2 8:30 ET (consensus +0.9% q/q); Chicago PMI 9:45; University of Michigan final July sentiment 10:00. The ECI is the highest-signal print of the day given three FOMC dissents favouring a hike. | macro | unknown |
| 2026-07-31 | BANK OF JAPAN rate decision plus quarterly Outlook Report. Elevated importance after suspected BoJ intervention on Jul 30 drove USD/JPY down ~2.9% below 160 and lifted FXY implied vol. A yen-carry focal point; affects MUFG, SMFG, MFG, NMR, HMC and broad risk assets. | macro | unknown |
| 2026-07-31 | Q2 earnings, BMO. Options price roughly +/-3.5% for both. XOM call:put built from 1.7:1 to 2.1:1 through Jul 30; CVX flat at 1.1:1. Both sit on a crude price that FELL 0.7% on Jul 30 despite fresh US strikes on Iran, as Hormuz transit improved. | XOMCVX | normal |
| 2026-07-31 | Large-cap earnings. ABBV weekly IV 95 vs August 36 (2.6x), call:put 1:1. ARES weekly IV 118 vs August 60, call:put 1:1.4 puts. KKR weekly IV 105 vs August 45, call:put 1:1.6 PUTS - the clearest put skew of the day's slate. MRNA is a double catalyst: earnings today plus a PDUFA date Aug 5. | ABBVARESKKRLINUBSETNENBCLCCJTROWMRNALYBCBOE | normal |
| 2026-08-02 | OPEC+ ministerial meeting (Sunday) to set September output. Non-trading day; market reaction lands Aug 3. Reporting on the direction of the prior decision remains CONFLICTED across outlets and is UNVERIFIED; the meeting date is verified. | XOMCVXSHELBPOXYFANGDVNEOGVLOMPCPSX | unknown |
| 2026-08-03 | ISM Manufacturing PMI July 10:00 ET - consensus 54.0, up from 53.3, with prices-paid forecast to FALL to 70 from 73.0 and new orders to improve to 57 from 56. The prices-paid component is the one to watch given the hawkish dissent bloc. Also S&P Global Manufacturing PMI 9:45, Construction Spending June 10:00, and US Treasury quarterly refunding borrowing estimates plus TBAC statements. | macro | unknown |
| 2026-08-03 | Q2 earnings, AMC, webcast 5:00 pm ET. Marquee report of the day; PLTR is Track B downside rank 9. SOURCE-DISCIPLINE NOTE: several widely used data providers list this as August 2, which is a Sunday; the company's own release says August 3 and the issuer is the authority. | PLTR | high |
| 2026-08-03 | Large-cap earnings (mixed BMO/AMC). Berkshire Hathaway and Marriott are the size anchors; ON and MUFG carry read-through to the semiconductor and yen-carry themes respectively. | BRKMARVRTXSNAPWMBOKEFANGONMELITSNMUFG | normal |
| 2026-08-04 | June JOLTS job openings 10:00 ET; June international trade balance 8:30; June factory orders 10:00 (consensus +0.1% m/m). JOLTS is the leading tell into the Aug 7 payrolls print that sits just OUTSIDE this horizon. | macro | unknown |
| 2026-08-04 | SpaceX FIRST-EVER earnings report as a public company, AMC. Options previously priced a ~15% implied move with 30-day IV at the top of its 52-week band. Triggers a tiered insider lock-up whose first tranche opens Aug 6 - up to 911.5M shares - against reported short interest near one-third of available float. Shares have stumbled since the post-IPO surge and fell 0.31% on Jul 30. | SPCX | high |
| 2026-08-04 | Q2 earnings, AMC. AMD is Track B downside rank 8 AND Track A rising rank 5 - it gained 13.00% on Jul 30 on Microsoft's earnings rather than its own, so this is the first test of whether its own numbers support the move. | AMD | high |
| 2026-08-04 | Large-cap earnings, BMO. MRK is Track B downside rank 5 (consensus EPS -$0.26); CAT guidance will be read as a tariff and cost-inflation tell for the whole industrial complex. | CATMRKMCDPFEBPDUKSPOTSUAPTVDDHSBC | normal |
| 2026-08-04 | Large-cap earnings, AMC. ANET and ALAB are the AI-networking read-throughs most directly exposed to the Jul 30 capex validation. | AMGNANETGILDBKNGEMREOGDVNALABPINSLSCCTOSTWYNN | normal |
| 2026-08-05 | ADP National Employment Report July 8:15 ET and ISM Services PMI July 10:00; US Treasury quarterly refunding statement plus TBAC report. ADP is the second leading tell into Aug 7 payrolls. | macro | unknown |
| 2026-08-05 | Q2 earnings, BMO. Largest single-name catalyst of the day. LLY is Track B upside rank 8 AND Track A falling rank 5 - it dropped 4.55% on Jul 30 with no identified catalyst, so this print resolves an open question rather than merely updating a model. | LLY | normal |
| 2026-08-05 | Earnings, AMC. SNDK is Track A rising rank 2 with PUMP_RISK flagged and carried 30-day IV of 143 against a 52-week range of 44-163. The single highest-variance scheduled event in the horizon after SpaceX, and the first hard test of Samsung's memory-shortage-to-2028 claim against a Western supplier's own guidance. | SNDKWDC | high |
| 2026-08-05 | PDUFA goal date for the mRNA-1010 seasonal influenza vaccine. FDA VRBPAC voted 9-0 in favour on 2026-06-18 (separate 9-0 votes for adults 50-64 and 65+). Second leg of a double catalyst after Jul 31 earnings. | MRNA | high |
| 2026-08-05 | Large-cap earnings (mixed BMO/AMC). NVO is Track B downside rank 4 and this print can resolve that thesis in either direction inside the horizon. UBER is on the deep-dive watchlist and did NOT surface organically in either track this run. | DISNVOSHOPCVSUBERPSXKHCAPPMCKDASHOXYEBAYAXONEXPENYTIONQ | normal |
| 2026-08-06 | SpaceX first insider lock-up tranche OPENS - up to 911.5M shares, with an extra 10% unlocking if the stock closed at least 30% above the $135 IPO price on at least 5 of the 10 sessions before earnings (Musk excluded). Lands 48 hours after the maiden earnings report into short interest near one-third of float. Two opposing forces - a squeeze setup and a supply cliff - resolve within two sessions of each other. | SPCX | high |
| 2026-08-06 | Q2 productivity and costs; ConocoPhillips earnings BMO. Productivity matters more than usual: with Q2 GDP at 1.5% and ECI landing Jul 31, unit labour costs are the arithmetic bridge between the two and feed directly into the hawkish-dissent debate. | COP | normal |
E · Trade Ideas
What the data implies
MSFTlong · track A/B
Rationale The thesis and the tape now agree, which is exactly when position sizing gets dangerous. FQ4 delivered revenue of $90.0B against $87.6B, Azure +43% in constant currency against a 40% estimate, annual Azure revenue above $100B for the first time, backlog $678B +84% and Copilot above 30M paid seats. This is the only hyperscaler demonstrably monetising AI at scale while still generating $67B of free cash flow. The structural point is that Microsoft has now defined the test the market will apply to every other AI spender: does the capex show up in someone else's revenue line.
Invalidation Azure growth decelerating below the high-30s in constant currency, or a capex guide that rises again without a corresponding backlog print. Note the backlog, not the quarterly revenue, is the leading indicator here.
Risk The options market underpriced this event by roughly 2x (7% implied vs 15.51% realised), which cuts both ways: the market's model of Microsoft is unstable, and the next surprise is as likely to be negative. Azure growth is materially OpenAI-linked, concentrating exposure to one counterparty's funding. Do not hold alongside TSM, NVDA, AVGO, GEV and VRT and call it diversification - all six are the same capex bet.
MUneutral · track A/B
Rationale Held at NEUTRAL for a second consecutive run, but the balance shifted. Samsung warning the memory shortage persists into 2028 while posting a record Q2 operating profit up 1,813.8% y/y is the most direct rebuttal available to the CXMT-oversupply thesis that drove the July liquidation, and JPMorgan's assessment that hedge-fund deleveraging in the complex is essentially complete removes the mechanical seller. Structural_thesis was raised 62 to 70 on that basis. It remains NEUTRAL rather than long because an 85% gross margin is still roughly 25 points above any historically sustainable level, customer inventories at 7-9 weeks still sit against an 8-week warning threshold, and a supplier forecasting a two-year shortage has an obvious incentive to do so.
Invalidation LONG invalidated if customer inventories cross 10 weeks (the stated exit trigger) or if FQ4 gross margin guidance is cut. SHORT invalidated by the Samsung shortage guidance being corroborated in Western supplier guidance - the first hard test is SNDK and WDC reporting Aug 5 AMC.
Risk Highest-variance name in this artifact and the market cap carries an unresolved source conflict inherited from the prior run - size off verified fundamentals, not off a quoted cap. Track A rising rank 3 and Track B upside rank 5 now point in the SAME direction, which removes the useful tension the prior run had; do not read agreement between two independently scored tracks as confirmation, since both are reading the same Samsung headline.
METAshort · track A/B
Rationale The clearest single-name expression of the session's organising idea. Microsoft and Meta reported the same night, both raised capex, and the market rewarded one and punished the other - the difference is that Microsoft has an external revenue line for its buildout and Meta does not. Two items are new and both are adverse: Q3 revenue guided BELOW consensus, which extends the problem from capex to the top line, and a Thursday securities filing (Tier 1) disclosing $279B of future lease obligations largely tied to AI data centres, up 53% in three months. That last number is growing faster than the $130-145B guided capex figure and sits largely outside it.
Invalidation A Q4 revenue guide back at or above consensus, or any credible disclosure that the lease commitments carry third-party revenue against them. Note the position is HELD on the Track B upside list at rank 9 on a 17.6x forward multiple - the valuation leg is intact and one more quarter of below-consensus revenue guidance should retire that position rather than re-score it.
Risk THIS IS THE MOST INTERNALLY CONTESTED NAME IN THE ARTIFACT AND THAT IS STATED DELIBERATELY: META is Track A falling rank 2 AND Track B upside rank 9 simultaneously. The tracks are scored independently and are not netted. The honest summary is that the 12-month valuation case and the near-term revisions case point in opposite directions, and anyone acting on one should know the other exists. Short interest is already at its highest since 2022.
AAPLshort · track B
Rationale Thesis confirmed by the print in its specifics rather than merely its direction, which is why this moved from downside rank 7 to rank 3 - the largest justified re-rank in the run. FQ3 beat on revenue ($109.4B, +16%) and EPS ($2.02, +29%) but missed on Services ($30.7B against ~$31.2B) and Greater China. Services is the high-multiple, AI-adjacent segment that justifies a 36.6x forward multiple; hardware is the low-multiple segment that beat. A print that beats on the cheap line and misses on the expensive line is a de-rating event even when the headline is green. Consensus PT sits BELOW spot - one of only two names in this artifact where that is true.
Invalidation A Services re-acceleration above the low-teens, a credible Siri or on-device AI launch with a shipping date, or a China stabilisation. Any of the three closes the narrative gap quickly.
Risk $123.3B of TTM free cash flow and the largest buyback in history put a hard floor under this. The pre-print options market priced only a ~5% move with call:put 1.2:1, so the crowd was not positioned for the miss and may not chase it. A new CEO takes over in September and has every incentive to reset expectations early, which can front-load bad news and clear the path - a genuine risk to a short held over that transition.
CRWVshort · track B
Rationale Retained but explicitly DOWNGRADED, and this entry exists mainly to document that. The balance-sheet case is unchanged and severe: $35B of debt against $40B of equity value funding depreciating GPUs, -$10.6B of free cash flow, Altman Z 0.36 and Piotroski F 2. The demand leg, however, was actively contradicted inside the window - Microsoft's Azure beat and Amazon's AWS beat with capex hiked to $220B both argue neocloud capacity gets absorbed rather than stranded. structural_thesis cut 82 to 78, sentiment 60 to 55, composite 75 to 73, rank 4 to 6.
Invalidation ESCALATED WATCH. If the Aug 4-5 cohort (AMD, ANET, ALAB, then the memory names) confirms capex acceleration again, this position should be RETIRED rather than re-scored. A successful refinancing at reasonable spreads, or any Nvidia support event, invalidates immediately.
Risk Borrow cost is likely punitive, the stock is already -32% over 52 weeks, and consensus PT sits 87% above spot. This is a credit thesis expressed in equity, which is an imperfect instrument, and it is now running against the demand tape. Contracted revenue with investment-grade counterparties materially de-risks the near term.
XOMCVXneutral · track A
Rationale Framed as NEUTRAL and reversed from the prior run's long, on evidence rather than on price. The geopolitical bid stopped working: the US launched fresh strikes on Iranian targets including Qeshm Island overnight and crude still FELL 0.7% to roughly $83.90, while gold was flat at +0.05%. Hormuz transit improved to 14 vessels from single digits. When a live state-on-state military exchange cannot lift crude, the risk premium is being sold, not bought. The options tape disagrees, which is what makes it interesting: XOM's call:put BUILT from 1.7:1 to 2.1:1 through the session even as spot fell.
Invalidation Resolves fast and in a scheduled way: both report Jul 31 BMO with options pricing roughly +/-3.5%, and OPEC+ meets Aug 2 with the direction of its prior decision still UNVERIFIED across sources. A Hormuz closure or a CPC escalation re-rates crude regardless of positioning.
Risk Headline-driven and two-sided with a documented same-week reversal already on the record. The Caspian Pipeline Consortium suspended Black Sea loadings for a second time after drone strikes, so the supply side is not clean either. Note the divergence between XOM's directional skew and CVX's flat 1.1:1 - the options market has a view on one and not the other.
SPCXneutral · track A
Rationale Unchanged from the prior run and now closer: the most asymmetric scheduled event in the horizon and the one with the least usable prior. First-ever public earnings report Aug 4 AMC, immediately followed on Aug 6 by a tiered lock-up releasing up to 911.5M shares, into reported short interest near one-third of available float. Both legs now sit INSIDE the horizon, which was not true last run. SPCX remained on Market Rebellion's popular-stock and active-options lists through Jul 29-30 while falling 0.31% on a day the Nasdaq rose 2.78% - persistent attention without participation.
Invalidation n/a (volatility view). The directional read only becomes tradeable AFTER the print clarifies whether the extra 10% unlock tranche triggered.
Risk No earnings history whatsoever - there is no base rate for how this reacts. The market-cap figure available is stale and sources conflict by roughly $500B, so do not size off it. A squeeze setup and a supply cliff resolving within 48 hours of each other is a structure that can produce a large move in either direction and then reverse it.
Rationale An OBSERVATION and a model-calibration note, not a position. Jul 30 was a discrimination event inside the semiconductor complex, and the two names bracket it. Lam Research beat and guided to over 20% sequential revenue growth and rose 18.2%, its best day since 1999. Qualcomm beat on revenue ($9.95B vs $9.67B), missed narrowly on EPS ($2.21 vs $2.23), reported handset revenue down 20% y/y, and FELL 5% on a day the SOX rose about 8% - a 13-point relative underperformance. Being in the same index bought Qualcomm nothing. The prior run's calibration rule was 'beats WITHOUT demonstrated AI monetisation are punished'; Jul 30 refines it to 'beats without AI-capex EXPOSURE are punished, regardless of quality'.
Invalidation The pattern breaks if handset, auto or consumer-exposed semis start being rewarded for good numbers. Watch the Aug 4-5 cohort - AMD, ANET, ALAB, LSCC, then SNDK and WDC - for whether the discrimination persists or collapses back into sector beta.
Risk The counter-evidence is that one earnings print (Microsoft's) drove the entire dispersion, so this may be a one-day artifact rather than a regime. QCOM's magnitude is itself conflicted across sources at -4%, -5% and 'more than 7%', which limits how hard the observation can be leaned on.
Methodology & limitations
- INTRADAY-VERSUS-CLOSE HAZARD WAS THE DOMINANT DATA RISK THIS RUN, more so than last. Jul 30 was a session that ACCELERATED into the close - LRCX was quoted at +6% after hours Jul 29, 'nearly 9%' premarket, and closed +18.2% - so any figure sourced before 4pm ET understates the move. Several widely indexed sources (HDFCSky, proactiveinvestors, and an AP wire snapshot) published index levels that do not match the confirmed closes, marking their single-name figures as intraday. Those sources were used only where a close-basis figure corroborated them.
- STALE-SOURCE FAILURE MODES CONFIRMED AGAIN. The Zacks/Yahoo 'Stock Market News for July 30, 2026' page recaps the JUL 29 session (Dow 51,594.14) - the third consecutive run in which this behaviour has been caught, and it is now treated as a known property of that source rather than an error. MacroTrends reported MSFT's 'latest closing price as of July 30, 2026' as $451.10, which is the JUL 29 close, and gave a META price of $585.61 irreconcilable with the $539.03 in the prior run's aggregator snapshot. A Yahoo Finance sidebar widget gave Russell 2000 at 2,936.49 (+30.18) against two independent sources agreeing on 2,946.10 (+39.79) - the documented Yahoo caching failure mode. All four were discarded rather than averaged.
- NO LIVE OPTIONS TAPE. Barchart Unusual Options Activity and Market Chameleon render client-side and returned page chrome only. No ticker-level volume-to-open-interest table for Jul 29 or Jul 30 was retrievable from any free source. The named options activity in this artifact is the COMPLETE verified set - nothing was invented to fill the gap. Track A options sub-scores are neutral-50 for 5 of 15 ranked names.
- NO BID/ASK SIDE TAGGING on any free source. Every call:put figure is a VOLUME ratio. This mattered acutely this run: ALNY showed increasing unusual CALL volume on a session it fell 29.4%, which is equally consistent with dip-buying, hedge closing and put-spread construction, and is published as ambiguous rather than resolved.
- NAME-LEVEL IV IS TWO SESSIONS STALE FOR THE MEMORY COMPLEX. The MU (97) and SNDK (143) IV levels used here come from the Jul 29 pre-market report and therefore reflect the JUL 28 close. Neither the Jul 30 pre-market nor mid-session report published an updated name-level figure for either. Given both names moved 18-26% on Jul 30, those IV levels should be treated as historical context, not as current positioning.
- REDDIT AND SOCIAL MENTION-VELOCITY TRACKERS WERE NOT RETRIEVED THIS RUN. The prior run carried ApeWisdom, Tradestie and AltIndex mention counts; no equivalent data was obtained for Jul 29-30. Rather than carry stale counts forward or estimate them, sentiment_velocity sub-scores are built on documented proxies only - Market Rebellion's 'popular stocks with increasing option volume' and 'active options' lists, plus observed Tier-1/2 coverage volume - and every sentiment block names those proxies explicitly. This is a genuine reduction in evidentiary quality versus the prior run and is stated rather than papered over.
- VIX CLOSE IS A THREE-WAY CONFLICT (17.09 / 17.64 / 17.77), all three reconciling arithmetically against the confirmed Jul 29 close of 20.66. Published as a point estimate with the full range stated; not averaged.
- UNRESOLVED SINGLE-NAME MAGNITUDE CONFLICTS, published with flags rather than point-estimated silently: LRCX (18.0 / 18.2 / 18.26 / 18.3 / 19.77 / 20.20 across six sources - the 18.2-18.3 cluster is used); SNDK (16.5 / 20.95 / 23.16 / 26); MU (13.18 / 15.47 / 18); QCOM (-4 / -5 / 'more than 7'); AMD (7.3 / 12.35 / 13.00 - the confirmed close-basis 13.00 is used); NVDA (1.63 / 2.65 / 2.93); ALNY (-21.1 intraday versus -29.4 close); CVNA (-10 premarket / -14 after-hours / -12 close); SOX/SOXX (7 / 8 / 8.2).
- MARKET CAP FOR ALNY COULD NOT BE RECONCILED: $41.17B (Trading Economics, July), $41.79B (MacroTrends, Jul 6) versus $59.3B and $60.1B (SimplyWall.st) - roughly a 45% spread. No point estimate is published. Universe qualification is not in doubt at either figure.
- TRACK B VALUATION FIELDS ARE AS OF THE JUL 29 CLOSE and are materially stale for the six names that moved more than 7% on Jul 30 (MSFT +15.51, MU +18, AMD +13, INTC +11, TSM +7.6, META -7.95). Each affected entry carries an explicit note. Re-deriving forward multiples from unverified post-move prices would be fabrication and was not done.
- BREADTH is described qualitatively from the Dow constituent count (23 of 30 lower) and an early-session S&P advancer count (243). A full NYSE composite advance/decline line for Jul 30 was not obtainable, and the S&P-internal advance/decline ratio published for Jul 29 has no Jul 30 equivalent in the sources consulted.
- BLOCKED OR RESTRICTED SOURCES: Bloomberg and CNBC return 403 to automated fetch; the Globe and Mail and Seeking Alpha are robots-disallowed. Their content appears here only via search-index headlines or third-party citation and is tiered accordingly.
- SCORES ARE HEURISTIC AND NOT BACKTESTED. Sub-scores are analyst judgment applied to cited evidence, not a fitted model. Every sub-signal that could not be verified from a free source is set to neutral 50 and flagged.
- SINGLE-FACTOR CONCENTRATION: 8 of 10 Track B upside names and 3 of 10 downside names remain expressions of one variable. Jul 30 demonstrated the risk directly - one earnings print moved nine of the ten upside names in the same direction on the same day.
Not financial advice. Educational/analytical only — generated by Daily Market Pulse template v1 - multi-desk research run (market regime, Track A movers, options/IV, catalyst calendar, Track B conviction), RUN_DATE 2026-07-30, synthesized 2026-07-31. Do your own diligence.