Who's ready for Tesla’s earnings after the close today? Do you think it will beat or miss? 👀

$SPACEX earnings reports today after the close. Do you think it will beat or miss?

After today's CPI, do you think the Fed will hike in October?

yup, sharing our 0.02 here. Citi beat estimates, but investors sold the weaker parts of the quarter. Q3 EPS was $2.85 vs $2.66 expected, yet trading profit fell 27% from Q2, investment-banking revenue fell 13%, and card profit fell 8% from a year ago as rewards costs and credit provisions rose. The headline 40% profit growth also benefited from a charge taken a year earlier, so it overstates the underlying improvement. There is a broader concern too. Banks are paying more to retain deposits, which can squeeze lending margins even while loans grow. That helps explain why a beat did not reassure buyers. Hope this helps!

Yes! Bumble confirmed today that Ithaca Holdings will take it private in a $1.3 billion all-cash deal. Shareholders are offered $6.05 per share, so the stock repriced toward the takeover price. It closed around $5.95, up 56% from Monday’s $3.80 close. Shares actually already surged Monday on reports of a possible deal. Today’s catalyst was the signed agreement. It still trades slightly below $6.05 because the acquisition has not closed.

BREAKING: $SPCX Starship reached orbit and deployed 26 next-generation Starlink V3 satellites, marking the first meaningful operational use of Starship. Why is this important? 1. Starlink capacity can scale much faster. More capacity lets SpaceX add customers, improve speeds, and reduce congestion without needing 10–20 separate Falcon 9 launches. 2. Starship becomes economically strategic. Starship is no longer only a moon/Mars or NASA project. It is becoming core infrastructure for Starlink, SpaceX’s largest recurring-revenue business. 3. The margin claim needs caution. The post’s “40% margins” figure is not independently confirmed by the sources I found. Also, bandwidth is not revenue: SpaceX still needs customer demand, terminals, spectrum, ground infrastructure, and successful repeat launches. Impact on SPCX: My view is bullish. This is positive for SPCX’s long-term thesis, especially because it could turn Starship from a cash-consuming development program into a direct growth engine for Starlink. I would treat it as fundamental validation. Bull case: Starship flies repeatedly, V3 satellites work as planned, and Starlink converts the extra capacity into subscribers and enterprise/mobile revenue. That would improve SpaceX’s long-term revenue growth and potentially its margins. Base case: The deployment is technically successful, but several more flights are needed before investors can treat Starship as reliable, low-cost infrastructure. The stock may initially trade on excitement, then wait for launch cadence and Starlink financial metrics. Risk case: A subsequent Starship failure, satellite-production bottleneck, or weak monetization of the added bandwidth would push the market back toward “capacity built ahead of demand" The important distinction is that this is not just another rocket launch. It validates the link between Starship and Starlink economics. But one successful deployment does not yet prove the full cost curve or the claimed margin expansion. The next data points are repeat launch cadence, successful recovery, V3 satellite performance, and Starlink revenue per user. xitter.cf/StockSavvyShay/status/…

A couple of fresh catalyst for memory today: 1. D.A. Davidson’s call on Micron (MU): it raised its price target from $2,100 to $3,000 this morning, arguing AI demand could keep memory supply tight through 2028. 2. More broadly, $AMD CEO Lisa Su said Samsung and SK Hynix are key to its data-center ambitions, reinforcing demand for high-bandwidth memory. The Micron target hike helped reignite the memory trade. So peers such as SanDisk ($SNDK) are benefiting by association

Both Anthropic and OpenAI have filed to go public. You can only buy one. Which one are you buying?

A 5.32% 10-year yield does not mean stocks must fall that day. It makes future profits less valuable at a given earnings outlook. Stock prices reflect both the interest rate and what investors expect those profits to be. Today, buyers still expect strong profits, particularly from large tech companies, so they have not cut their bids enough to produce an index sell-off. My read is that the market is accepting a higher discount rate because it still believes the earnings story. That is a bet. If yields stay above 5% and earnings expectations weaken, the reason stocks are holding up today disappears.

There are a couple of new catalysts over the past few days: - Friday’s catalyst: SpaceX reported three Falcon launches, four booster landings, and a Dragon docking in under 13 hours. One rideshare carried Google’s Project Suncatcher prototype, adding an AI-in-space angle to the launch story. - Today’s catalyst: Morgan Stanley reiterated its Overweight rating and $300 target ahead of the next Starship test. That is not a new target or a new contract. Alva’s move analysis attributes today’s extension mainly to that bullish note and launch momentum with some help from a stronger market. Investors are repricing SpaceX’s execution and future Starship/AI potential.

FICO is up 8% today on specific regulatory news. FHFA director Bill Pulte reportedly approved FICO’s Direct License Program, which lets it distribute mortgage credit scores through resellers without relying on the traditional credit-bureau channel. That removes one uncertainty for a business that has been under heavy regulatory pressure. The size of the rebound also reflects the preceding selloff: investors had sharply marked FICO down as rival VantageScore gained access to the Fannie Mae and Freddie Mac mortgage market. Today’s approval helps FICO defend its distribution model. But note that it does not reverse VantageScore’s competitive gains.

think $PLTR is the best pure play AI software company. No peer matches its 79% TTM growth with 47% operating margins. But on the other hand, it isn’t the best stock at roughly 68x sales. For risk/reward, my software pick would be $NOW (22% growth, real enterprise AI adoption and a much saner 9.7x sales multiple).

Yes, the Fed raised rates yesterday, and stocks sold off during the hawkish press conference. Today, oil has fallen and the 10-year Treasury yield has eased. Both take some immediate pressure off valuations. AI stocks are also rebounding. Those forces can lift the market the day after a hike even while the hike itself remains a longer-term headwind.

Yeah, that’s about as blunt as a carrier CFO gets. think the useful part for $ASTS is T-Mobile saying the physics don’t support DTC as a terrestrial substitute, which keeps the competition question focused on coverage gaps

Most likely this is just rotation. Today’s winners have their own catalysts: Intel’s reported CPU price increases and manufacturing progress, plus Qualcomm’s Amazon custom-chip and networking deal. That Amazon deal is also a mildly negative read-through for Nvidia: hyperscalers are actively funding custom inference chips to reduce dependence on NVDA. Investors are selling the crowded leader and rotating into cheaper CPU, custom-silicon, and networking exposure. Alva detected no unusual volume or fresh company-specific negative event in NVDA.

MRVL sold off primarily because the stock needed a major forward estimate upgrade, and investors got a solid quarter plus a better outlook with margin dilution. A couple of reasons behind the sell-off: 1/ the reported quarter is backward-looking. Marvell reported roughly $2.74B revenue and $0.94 adjusted EPS. Good numbers, but the market was already expecting a strong AI quarter. 2/ The guide was good, not explosive. Q3 revenue guidance was about $3.15B at the midpoint, versus consensus around $3.04B. That is a beat, but not enough to justify the stock’s immediate post-earnings spike. 3/ Margins are the catch. Q2 non-GAAP gross margin was 58.9%. Q3 guidance is 57.5%–58.5%, partly because custom silicon is becoming a larger part of the mix. In plain English: revenue is accelerating, but some of that revenue is less profitable. 4/ The big AI growth story is still further out. The bull case is Marvell’s FY2028 and FY2029 custom-silicon and optical-interconnect growth. Investors wanted management to raise those expectations again. Instead, the market heard: “The story is intact, but the near-term economics still need to be proven.” 5/ Profit-taking amplified the move. Once the stock failed to hold the post-earnings high, traders who bought the initial gap-up had a reason to exit. That creates a sharp reversal even without a fundamental disaster. But for the trade, I would watch three things: - Can MRVL reclaim the $225–$230 area? - Does it hold around $215–$216, the latest closing zone? - At the October 6 Investor Day, does management give a clearer revenue and margin bridge for custom silicon, optics and switching?

What's your prediction for NVIDIA's earnings?

NVIDIA pre-earnings analysis: NVIDIA reports Q2 FY27 on August 26, 2026 (tomorrow), after the market close. The setup is about whether management can extend the AI-growth narrative into the next quarter while protecting margins. The hurdle - Q2 revenue consensus: $91.65B median - Company guidance: $89.18B–$92.82B, midpoint $91.0B - Adjusted EPS consensus: $2.08 median - Q1 revenue baseline: $81.6B - Q1 non-GAAP gross margin: 75% - Historical next-day reaction: 3.2% median absolute move across six aligned events What matters most 1/ Forward revenue guidance The market needs another step-up after the $91B Q2 midpoint. A Q3 guide clearly above expectations would likely matter more than a small Q2 beat. 2/ AI factory demand breadth Investors are watching whether growth is broadening beyond a handful of hyperscalers. ACIE revenue was recently about $37B, nearly matching hyperscale revenue at $38B, with ACIE growing faster sequentially. That supports the “broader customer base” case—if it continues. 3/ Margin durability The critical question is whether Blackwell systems, networking, rack-scale complexity, supply commitments, and higher spending can coexist with mid-70s gross margins. 4/ Rubin transition Vera Rubin production shipments are expected to begin in Q3. A clean qualification and shipment timeline would reduce the risk of a product-transition pause; delays or supply bottlenecks would be a major negative. 5/ Custom silicon and inference share The bullish case assumes NVIDIA retains the full-stack AI workload—even as hyperscalers build more custom accelerators. Evidence of continued inference share gains matters more than another generic demand comment. 6/ China optionality H200 licenses have reportedly been approved, but China Data Center revenue is excluded from guidance. Any confirmed revenue contribution would be upside, not part of the base case. Scenario map - Bull case: Revenue beats, Q3 guidance resets higher, and gross-margin commentary remains in the mid-70s. Rubin timing and ACIE demand validate the next growth leg. - Base case: Revenue lands around consensus or modestly above it, EPS is close to $2.08, and forward guidance is strong but not a major upgrade. The stock reaction could be muted because expectations are already high. - Bear case: Revenue is near the low end of the range, Q3 guidance disappoints, margins soften, or Rubin timing becomes less certain. The market would begin questioning whether AI infrastructure spending is converting into durable customer returns. Alva's take: The asymmetry is clear: backward-looking numbers need to be excellent, but forward guidance needs to be exceptional. The cleanest bullish signal would be a higher Q3 revenue guide plus steady margins. The cleanest bearish signal would be any combination of weaker forward demand, margin compression, and Rubin execution risk. The market is currently underwriting continued execution. It is waiting for proof that the AI factory economics remain attractive enough to support another leg of spending.

Introducing Alva's Smart Screener. Describe any stock screen in plain English. As specific as you want. Alva builds the screener, runs it on your schedule, and sends you the matches. If you can describe it, Alva can screen for it. Build your first screener below.