Warren Buffett Watch: It's Buffett, not Abel, who appears to be calling the shots on stocks

Affirm enters its November 5 earnings report with analysts expecting growth above management's own starting point, but the gap between expectations and guidance is now a source of risk as well as potential upside.

Morgan Stanley lowered its Affirm price target to $79 from $82 and kept an Equal Weight rating, according to the original report. The firm still expects fiscal first-quarter gross merchandise volume, or GMV, to grow about 34%, compared with roughly 31% at the high end of the company's implied outlook. The report says investors may be positioned for growth closer to 37%, which means a result above management's range could still disappoint if it falls short of the market's unofficial bar.

A second firm took a more bullish position. Monness Crespi initiated coverage with a Buy rating and a $90 target, arguing that buy-now-pay-later providers and neobanks have taken a larger share of spending and lending. The next phase depends on customers using the services more often and on changes in customer mix, not simply adding more users.

Affirm's own outlook calls for first-quarter GMV of $13.7 billion to $14.0 billion and revenue of $1.19 billion to $1.22 billion. The original report cites Fiscal.ai consensus revenue of $1.21 billion and earnings of $0.38 per share, compared with $0.23 a year earlier. Revenue consensus therefore sits inside the company's range, while the more demanding debate is about GMV growth and profitability.

The company also expects a GAAP operating margin of 11.5% to 13.5% and an adjusted operating margin of 28% to 30%. Its outlook assumes a promotional event with a large merchant will not repeat and that another merchant's move to its own wallet will remain a year-over-year headwind for part of the quarter. Affirm also expects short-term benchmark rates to be higher in fiscal 2027 than in fiscal 2026, adding pressure to funding economics.

Affirm reports after the market closes on November 5, with its conference call scheduled for 5 p.m. ET. A clean beat requires more than strong transaction volume. Investors should compare GMV with revenue less transaction costs, which the company guides to $575 million to $590 million, and check whether credit performance and funding costs support the margin forecast.

Amazon's third-quarter report on October 29 has a narrow test at its center: can Amazon Web Services accelerate again without making the returns on the company's AI spending harder to see? Bank of America's forecast says the cloud business can clear that bar, but the rest of the quarter leaves little room for a weak margin result.

The bank forecasts Amazon revenue of $202.0 billion and operating income of $26.4 billion. The revenue estimate is just below the $202.2 billion consensus cited in the original report, while operating income is slightly above the cited $26.1 billion consensus. Amazon's own guidance calls for revenue of $197 billion to $202 billion and operating income of $22.5 billion to $26.5 billion. In other words, the analyst's case sits near the top of management's ranges rather than assuming a broad companywide beat.

Bank of America expects AWS revenue to grow 40% from a year earlier, two percentage points above the consensus rate cited in its note. It estimates AWS will add roughly $4 billion of revenue from the second quarter and produce a 39% operating margin. The note attributes that growth to added capacity, price increases, AI laboratory contracts and enterprise demand for inference, while greater use of Amazon's Trainium chips and Bedrock platform could help absorb higher depreciation.

The starting point is already strong. In the second quarter, AWS sales rose 37% to $42.2 billion and AWS operating income reached $16.6 billion, according to Amazon. Advertising grew 26%, and total operating income rose to $27.5 billion. Yet Amazon's trailing free cash flow was a $7.6 billion outflow after property and equipment spending increased by $66.1 billion, mainly for AI infrastructure. That gap between reported operating profit and cash generation is the central counterweight to the cloud growth story.

The retail and advertising businesses could make the comparison less clean. Bank of America expects North American retail growth to slow to 6%, while advertising growth eases to 15% because Prime Day timing becomes less favorable. It forecasts fourth-quarter revenue guidance of $239 billion to $246 billion, with a midpoint below the consensus cited in the note, and operating income guidance of $29 billion to $34 billion.

A 40% AWS result would be encouraging, but investors also need evidence that new capacity is being used efficiently. The most revealing figures will be AWS operating margin, total capital spending, depreciation and the fourth-quarter profit range. Rapid cloud growth matters more if it can convert Amazon's unusually heavy AI investment into durable cash returns.

Roku is asking advertisers to treat the television home screen as the start of a measurable customer journey, not simply as a place to buy reach. The strategy fits the company's operating-system position, but the most important claims still come from Roku and its partners rather than independently audited campaign data.

Lauren Benedict, Roku's vice president of global ad sales and partnerships, told Beet.TV that Roku reaches 100 million households globally and meets viewers when they first switch on their televisions. She said users spend nearly 16 minutes searching for content and return to Roku on roughly 25 days each month, compared with about eight days for the next most-used app. Those figures explain why Roku believes menu, home-screen and in-content ads can work as a sequence rather than isolated impressions.

The commercial pitch is moving toward outcomes. Benedict cited Roku and Omnicom research in which a home-screen placement followed by sequential streaming messages produced nearly six times the purchase intent. A Roku case study said Toyota's multi-format campaign lifted ad recall by 8.1%, with 98% of campaign reach coming exclusively from sponsorship media. These are useful examples, but investors should treat them as company-reported evidence until methodology and control-group details are available.

Industry infrastructure is becoming more supportive. IAB Tech Lab introduced standardized signals for connected-TV formats such as pause, menu, overlay and squeeze-back ads. Common definitions can reduce manual work and make new inventory easier to buy programmatically, which may help Roku scale formats that appear before or between programs.

Roku's financial results show why this matters. Reuters reported that first-quarter 2026 platform revenue rose 28% to $1.13 billion, beating the estimate it cited, and Roku raised its full-year platform revenue forecast to $5 billion, or 21% growth. Devices revenue fell 16% to $118 million, reinforcing that advertising and other platform services, not hardware sales, are the larger economic opportunity.

The next proof point is not another household milestone. Roku needs to show that home-screen access, sequential formats and data partnerships increase advertiser spending and platform revenue without overwhelming viewers or relying on one-off case studies. Standardization makes the inventory easier to transact, but repeatable campaign performance will determine whether the operating-system advantage becomes durable cash flow.

A broad retail selloff pushed Lululemon and Under Armour down about 4% on October 23, while Nike fell roughly 2%. The similar share moves concealed three different operating problems, and investors should not treat the group as a single bet on consumer spending.

The macro backdrop was not uniformly weak. The U.S. Census Bureau estimated September retail and food-services sales at $745.0 billion, up 0.8% from August and 6.3% from a year earlier. The report was adjusted for seasonal and trading-day effects but not for price changes. With the SPDR S&P Retail ETF down about 2% that day and no fresh announcement from the three companies, the stock moves looked more like a repricing of discretionary risk than a response to new company results.

Lululemon has the clearest near-term demand problem. Second-quarter revenue fell 4% to $2.4 billion, Americas revenue declined 8%, and Americas comparable sales dropped 12%. Reported gross margin rose 200 basis points to 60.5%, but a $134.5 million tariff refund added 560 basis points. Operating margin still declined 190 basis points to 18.8%. Management guided third-quarter revenue to $2.29 billion to $2.32 billion, a 10% to 11% decline, and expects full-year revenue to fall 5% to 7%.

Under Armour's weakness is spread across channels and categories. Its first-quarter fiscal 2026 revenue declined 4% to $1.1 billion, with North America down 5%, wholesale down 5% and direct-to-consumer revenue down 3%. Footwear fell 14%, a sharper contraction than the 1% decline in apparel. The company reported $13 million of restructuring charges and held $911 million of cash, including $400 million raised in senior notes to help address $600 million of notes due in June 2026.

Nike is further along in parts of its turnaround, but the recovery is uneven. Fiscal first-quarter 2027 revenue fell 4% to $11.2 billion, while North America grew 2%. Performance categories grew at a high-single-digit rate, helped by running and global football, but Sportswear, Jordan and Greater China remained material drags. Nike ended the quarter with $8.4 billion in cash and short-term investments, giving it more financial room than the other two companies.

The October 23 selloff therefore did not prove that shoppers had suddenly stopped spending. It highlighted different tests: Lululemon must stabilize Americas comparable sales without relying on one-time tariff refunds, Under Armour must make restructuring produce durable demand and margin improvement, and Nike must turn strength in performance products into a broader brand recovery.

Exelon enters its October 30 earnings report with a larger growth opportunity than its current capital plan can capture, but the utility still needs to show which data center requests are likely to become paying load.

The company's second-quarter presentation identified about 36 gigawatts of projected data center demand across its service territories. Exelon classified roughly 11 gigawatts as high-probability load: about 9 gigawatts at ComEd in advanced design or backed by transmission security agreements, plus about 2 gigawatts in the Mid-Atlantic, where roughly 40% had signed agreements supported by about $1 billion of collateral. Another 25 gigawatts remained in current or future study clusters. That distinction matters because an interconnection request is not the same as a contracted customer.

Exelon's existing plan is already large. Management affirmed a $41.7 billion capital program through 2029, expected to produce 7.9% rate-base growth, and said it sees another $12 billion to $17 billion of transmission opportunities beyond the plan. Most of that additional opportunity is expected in 2030 or later. The company also highlighted more than 11,000 circuit miles of transmission and several awarded or proposed projects, including a $1 billion-plus MISO project being developed with Invenergy.

Near-term earnings are steadier than the load pipeline. Exelon reported second-quarter adjusted operating earnings of $0.43 per share, up from $0.39 a year earlier, while GAAP earnings were $0.39 in both periods. It maintained 2026 adjusted operating earnings guidance of $2.81 to $2.91 per share and still expects annual adjusted earnings growth near the top of its 5% to 7% range through 2029. Management had completed about 86% of its 2026 debt issuance needs and priced about 37% of the $3.4 billion of equity expected through 2029.

The financing progress reduces immediate interest-rate risk, but it does not remove execution or regulatory risk. A proposed $15.3 billion ComEd grid plan awaits an Illinois order expected by December 15, while BGE's $156.1 million electric rate request is scheduled for a January 2027 decision. Storm costs, a July PECO employee strike and the timing of large-load commitments can all shift quarterly results without changing the longer-term transmission thesis.

The most useful October 30 update will not be whether Exelon repeats the headline 36-gigawatt figure. Investors need to know how much of the 25-gigawatt study pipeline has advanced, how much collateral protects customers from abandoned projects, and whether incremental transmission awards can move into the capital plan without weakening affordability or credit metrics.