Notes from the Desk
Detailed market analysis of equity events. Each note focuses on one stock and the catalyst behind a notable price move: an earnings result, a regulatory ruling, or a shift in sentiment. Notes on closed classifications from the Historical Signal Board examine what the market had already priced in and how the classification resolved; notes marked Live cover an open event and withhold its classification. Published for general circulation; educational, not actionable.
Meta Q3 2026: Revenue Up 26%, a $9.89 Billion Settlement Charge, and Non-Voting Class C Shares
Meta's third quarter of 2026 reads three ways: revenue above the top of guidance, operating income down 37% after a $9.89 billion settlement charge, and capital spending larger than operating cash flow. The same filing creates a non-voting share class, to be paid as a one-for-one stock dividend in January 2027. This note separates the headline figures from the business impact and sets the cash flow next to the narrative.
Bull Trap: Anatomy of a False Breakout, and How to Measure One
A bull trap is a false breakout — price clears resistance, draws in buyers and short-covering, then reverses back through the level and traps the latecomers. This note gives a plain-language definition, decomposes the microstructure of the failure, sets out testable measurement rules for the tells, reviews the academic evidence on short-horizon reversal and technical levels, and works through a retrospective SELL classification of the 25 November 2025 Merck event, written in June 2026, against the realized tape.
Post-Earnings Announcement Drift: When a Beat Sells Off
PEAD is the tendency of a stock to keep moving in the direction of its earnings surprise for weeks after the report. This note defines and measures it — standardized unexpected earnings, drift magnitude by surprise decile, and why the anomaly survives — then works a retrospective BUY classification of the 17 July 2025 Abbott event, written in June 2026: a genuine beat that fell 8.5% on a guidance non-raise and reverted. That case is an overreaction-reversion, the mirror image of textbook drift, and the note keeps that distinction explicit rather than pretending the two are the same trade.
Dead Cat Bounce vs Relief Rally: Anatomy of a Failed Bounce
A dead cat bounce is a brief recovery in a downtrend that fails. This note defines the pattern — and how to tell it from a real bottom or a relief rally — then walks a quantified Microsoft SELL against the realized tape: a retrospective classification of the 9 April 2025 event, written in June 2026, in which a +10.1% tariff-pause pop round-tripped to a -8.0% trough.
Defensive Stocks: The Low-Beta Anomaly and the Sentiment Discount
Defensive stocks — consumer staples, utilities, healthcare — sell products whose demand barely tracks the business cycle, which shows up as steady cash flows and a beta below one. That same insensitivity means a several-percent one-day gap in a defensive name is mostly a repricing of mood, not of earnings power. This note defines the category, quantifies the low-volatility anomaly and the guidance-sandbagging incentive that feed these dislocations, and works through a retrospective BUY classification of the 14 February 2025 Procter & Gamble event, written in June 2026: a low-beta staple that fell ~4.8% on a sell-side guidance-skepticism note and mean-reverted to a +6.0% gain in seven sessions.
What Does "Priced In" Mean? A Beat That Moved a Stock Once
"Priced in" describes an outcome the market has already discounted, so its arrival adds no new information and moves price little. This note defines the term through semi-strong-form efficiency, shows how to measure whether news is embedded in a quote, separates a neutral efficient repricing from an active "sell the news" unwind, and works a JPMorgan HOLD — a retrospective classification of the 11 October 2024 event, written in June 2026, on a genuine Q3 beat that re-rated once and then held a range — with the classification methodology, the falsifiable invalidation, and the empirical literature made explicit.
Patent Cliff: Why Loss of Exclusivity Reprices a Pharma Stock
A patent cliff is the steep drop in a drug's revenue once market exclusivity ends and generics or biosimilars enter. This note quantifies the erosion curve from the empirical literature, shows why a scheduled step-down compresses a pharma multiple long before the date arrives, and dissects a retrospective SELL classification, written in June 2026, of the 26 July 2024 Bristol-Myers Squibb event — an 11.4% beat-and-raise pop, driven by costs rather than a revenue inflection, that faded into an unresolved cliff.
Buy the Rumor, Sell the News: Why a Record Quarter Sank the Stock
"Buy the rumor, sell the news" is the mechanism by which a fully anticipated event repriced in advance becomes a selling catalyst when it arrives. This note defines the pattern against semi-strong efficiency, shows how to quantify "priced in" ex ante — run-up magnitude, options-implied move versus realized move, implied-volatility crush, positioning proxies — and walks through a retrospective SELL classification of the 20 May 2026 NVIDIA event, written in June 2026, in which a ~20% pre-earnings run faded into a record quarter, reading the setup against the realized path and the empirical literature.
Operating Leverage: The Formula, the Math, and a Margin Inflection
Operating leverage is the multiplier a fixed-cost base places on revenue growth: once the base is covered, each incremental dollar of sales falls through to operating income at close to the contribution margin. This note derives the degree-of-operating-leverage identity, backs out the ~2.4x realized multiple from Uber's Q1 2026 print, quantifies the subscription flywheel behind it, and sets out a short-horizon BUY as a retrospective classification of the 6 May 2026 event, written in June 2026 — with the direction of the priced-risk evidence stated explicitly.
Biotech Catalysts: Pricing a Binary Pipeline De-Risking
A biotech catalyst is a scheduled, usually binary event — a trial readout, an FDA decision — that resolves a large block of a drug's uncertainty in one print. This note quantifies probability of success from published phase-transition data, prices a binary as the probability-weighted average across its two branches, treats a pipeline as a portfolio of staged real options, and works a retrospective BUY classification of the 3 February 2026 Pfizer event, written in June 2026: a Q4 sold on fading COVID revenue while a concurrent obesity-pipeline de-risking went under-weighted.
Turnaround Stocks: Scoring an Inflection vs. a Value Trap
A turnaround stock is a struggling company whose fundamentals are bending from decline toward durable improvement — and the money is made by telling that apart from a value trap before the market does. This note builds the distinction from the accounting-signal and profitability-factor literature, quantifies what a segment-level margin inflection does to consolidated earnings, and walks a retrospective BUY classification of the 3 February 2026 Disney close, written in June 2026: a mixed print the tape sold 7% while under-weighting a streaming-profit inflection that recovered the stock inside two weeks.