Capstone Deep Dive: US

Learning objectives

By the end you can:

1. Execute the entire program's system, unaided, on a US company you have never studied, going from a blank spreadsheet and a set of primary filings to a complete, institutional-grade thesis package, with no worked trail, no template prompting, and no AI, drawing only on the certified skill and the reusable artifacts you keep on your desk forever.
2. State precisely what "institutional-grade" means and hold your own output to it: not "clears the certification bar" but "would survive a hostile portfolio manager's read-through and could be published under your name": every claim sourced, every number tied to a filing, every qualitative judgment turned into a countable metric, and the whole thing conclusion-first.
3. Select the right capstone target from an approved candidate list: a company whose accounting is clean enough to learn on, whose disclosure is rich enough to work with, and whose price embeds a genuine, judgeable question. Then confirm honestly that you have never studied it.
4. Produce every component of the required thesis package to the deliverable spec: the ten-year spreadsheet history with incremental ROIC; the industry map (five forces + profit pool + share-stability test); the four-lens moat verdict proven in the numbers; the three-way valuation (DCF + comps + reverse DCF) triangulated to a defensible range; the management, capital-allocation, and incentives read from the DEF 14A; the scuttlebutt findings; the steelmanned bear; the one-page thesis; and the pre-mortem wired to a monitoring dashboard.
5. Triangulate a three-way valuation into an honest range whose width is itself information, reconciling what the DCF says the business is worth, what comparable multiples imply, and what the current price is already assuming (reverse DCF), then convert the gap between them into a falsifiable variant perception.
6. Grade your own package against the institutional rubric like a hostile examiner: scoring each deliverable component 0/1/2 against printed anchors, reading the total as a map of what to strengthen rather than a verdict on your worth, and catching the specific capstone failure modes (the book report, the false-precision valuation, the un-steelmanned bear, the missing variant question) before a grader does.
7. Run the two-week capstone as a project: sequencing the ten working days so that the numbers precede the narrative, the valuation is built on a rebuilt history rather than a screener export, and the qualitative edge (scuttlebutt, incentives, candor) is gathered after the numbers have told you what to look for.

Prerequisites & connections

Builds on: essentially the whole program, put to work at once. There is almost no new technique here; the entire content is the unaided synthesis of everything, held to a professional standard. Most immediately it rests on Phase 8: the source stack and 45–60-minute filing skim (M8.01), the twelve teardown questions and eight-line output (M8.02), the ten-year history and incremental ROIC (M8.03), the reverse DCF (M8.03), and the scuttlebutt, DEF 14A incentive read, steelman, one-page thesis, and pre-mortem (M8.04), all certified in M8.05. It leans just as hard on the machinery those modules assume: the three-statement linkage (M1.05) and full filing anatomy (M1.10); NOPAT, invested capital, and incremental ROIC by hand (M2.03), FCF and cash conversion (M2.04), the quality-of-earnings reflexes and red-flags checklist (M2.06–2.08); the WACC build (M3.03), FCFF forecasting and terminal value (M3.04–3.05), relative valuation and the driver algebra behind each multiple (M3.06), value drivers and the g = reinvestment rate × ROIC discipline and the five uses of capital (M3.07), and the three-statement model (M3.08–3.09); the four-lens moat protocol and share-stability test (M4.04–4.05); the sector KPI panels (Phase 5); the mental-model and personal checklists (M6.05); the macro overlay (Phase 7); and the sizing/sell/journal discipline (Phase 9). The reusable artifacts are your working kit: the [2–3 day deep-dive workbook](../artifacts/playbook-2-3-day-deep-dive.md), the [45–60-minute statement-read checklist](../artifacts/checklist-statement-read-45-60min.md), the [red-flags / quality-of-earnings checklist](../artifacts/checklist-red-flags-quality-of-earnings.md), the [valuation workflow](../artifacts/valuation-workflow.md), the [sector KPI cheat-sheets](../artifacts/sector-kpi-cheatsheets.md), the [mental-model & moat checklist](../artifacts/checklist-mental-models-and-moat.md), and the [macro dashboard](../artifacts/macro-dashboard.md).