P2B-C2 · 4-Dimensional Thesis Framework¶

Key Insight

A 4D thesis is missing nothing — but whichever dimension is missing, that's where the thesis breaks.

Layer 2 · Analyzing a Single Stock — After learning the industry, now apply it to one stock

L2-C2 (5 chapters total). After this chapter, you can expand C1's 30-character one-liner into a complete YAML to discuss with friends.

1. The Problem: One Sentence Works for Yourself, Not for Discussion¶

In C1, you wrote:

I'm bullish on NVDA because hyperscaler 2026 capex +60% + market worries about ASIC substitution but Blackwell is locked in for 12 months

Good enough for yourself. But when you show a friend:

"Where did you see +60%?" → You can't cite a specific source
"What does 'locked in for 12 months' mean? Is it RPO or backlog?" → You freeze
"You say the market worries about ASIC — but if ASIC actually takes off, when do you cut losses?" → You never thought about it

→ One sentence can't survive 3 follow-up questions. Discussion / verification / review all need a more structured vehicle.

2. The Solution: 6-Field YAML, Each Field Corresponds to One of the 4 Dimensions¶

Expand C1's one-liner into:

view: bull             # Single-point judgment
confidence: medium     # Your conviction in the view
core_thesis: |         # 1-2 sentences, an extension of C1's one-liner
supports:              # 3-5 specific pieces of evidence (with numbers and sources)
red_flags:             # 2-3 items + each with a trigger condition
catalysts_90d:         # Time-anchored observable events
price_outlook:         # base / bull / bear 90-day range

Key point: These 6 fields aren't random — they map to the Thesis 4 Dimensions taught in C1:

4 Dimensions	YAML Field	One-Sentence Answer
WHAT	`view` + `supports`	What's the current narrative + what evidence supports it
WHY	`core_thesis` + `confidence`	What are you more confident about than the market + how confident
SO WHAT	`catalysts_90d` + `price_outlook`	When will it be validated + what's the price impact
RISKS	`red_flags` (with trigger)	What would cause you to close the position

Missing a dimension → that's where the thesis breaks:

Missing supports → your view is a guess; you panic when earnings come out
Missing confidence → you don't know whether to take a starter or full position
Missing red_flag trigger → a 10% drop leaves you unsure if the thesis is broken or it's just noise
Missing catalyst → the thesis can never be validated — same as not writing it at all

3. How It Works: Design Trade-offs Between 4 Dimensions ↔ 6 Fields¶

graph LR
    Q[C1 One-liner] -.Expand.-> Y[YAML 6 Fields]
    Y --> W[WHAT<br/>view + supports]
    Y --> Why[WHY<br/>core_thesis + confidence]
    Y --> SW[SO WHAT<br/>catalysts + price_outlook]
    Y --> R[RISKS<br/>red_flags + trigger]

Design Principle 1: View Forces a Single Point¶

There's no "I'm not sure" option — you must choose bull / bear / neutral / watching. But the confidence field lets you say "I chose but I'm not confident."

❌ Bad example: "NVDA is definitely good long-term but uncertain short-term" — vague judgment, not a thesis ✅ Write it as: view=bull, confidence=low, red_flag="short-term valuation compression"

Design Principle 2: Supports Must Be Specific + Verifiable¶

✅ "NVDA Data Center segment Q4 FY26 revenue $35.6B YoY +93%"
❌ "The AI industry is growing fast" (this is an opinion, not a support)

Design Principle 3: Red Flags Must Have Triggers¶

✅ "If NVDA fwd PE > 35x AND hyperscaler capex guidance is cut, valuation compression risk triggered"
❌ "Valuation is high" (always true, never actionable)

A trigger tells you when the thesis breaks — this is the core difference between a thesis and a "feeling."

Design Principle 4: Catalysts Are Time-Anchored¶

Each catalyst must have a date + observable outcome:

"2026-05-20: NVDA Q1 FY27 earnings. Look for: Data Center segment YoY +60%+, FY27 guide >$200B"

If a thesis has no catalyst within 90 days, it's almost impossible to validate — you should set confidence=low or view=watching.

4. vs. What You Already Know from C1¶

Dimension	C1 You Can Do	C2 You Can Do More
Audience	Talk to yourself	Discuss with friends
Length	30 characters	30-50 lines of YAML
Verifiable	No	Yes (catalyst + price_outlook)
Reviewable	No	Yes (90 days later, check if supports still hold)
Risk Management	Implicit	Explicit (red_flag + trigger)

C1 is the entry ticket — without that one-liner, you shouldn't build a position. C2 is the battle plan — without that YAML, you shouldn't add to a position or make moves after adding.

5. Try It: Expand C1's One-Liner into a Complete YAML¶

Task: Using the ticker you wrote in C1, fill in the template below (~15 minutes)

ticker: ___
view: ___           # bull / bear / neutral / watching
confidence: ___     # high / medium / low
core_thesis: |
  (1-2 sentences, expand C1's 30-character one-liner into complete sentences)

supports:
  - "___"           # Must include numbers
  - "___"           # Must include source
  - "___"

red_flags:
  - text: "___"
    trigger: "___"  # What happens = thesis breaks
  - text: "___"
    trigger: "___"

catalysts_90d:
  - date: ____-__-__
    event: "___"
    look_for: "___"  # What numbers to watch on earnings night

price_outlook:
  current: $___
  base_90d: $___ - $___
  bull_90d: $___ - $___ (if ___)
  bear_90d: $___ - $___ (if ___)

NVDA v91 Real Thesis Example (Understand Before Copying)¶

ticker: NVDA
view: bull
confidence: medium
core_thesis: |
  NVDA benefits from continued growth in AI infrastructure capex. Q1 FY27 earnings will validate demand strength,
  but watch for customer quality risk (CoreWeave circular financing).

supports:
  - "2026 hyperscaler AI capex expected to reach $725B+, NVDA is the primary beneficiary"
  - "CoreWeave order backlog approaches $100B, an all-time high"
  - "327 large institutional 13F filings show NVDA remains the AI core holding"
  - "H200 China export licenses approved for approximately 10 companies"

red_flags:
  - text: "Samsung strike restarts 5/18; if an 18-day strike materializes, it could cut 3-4% of global DRAM supply"
    trigger: "Strike starts + lasts >7 days"
  - text: "CRWV-NVDA capital/commercial circular financing concerns (NVDA holds $36B stake in CRWV)"
    trigger: "CRWV customer concentration worsens OR next disclosure shows NVDA's CRWV exposure expands"

catalysts_90d:
  - date: 2026-05-20
    event: "Q1 FY27 earnings"
    look_for: "Data Center segment >$30B, FY27 guide >$200B, gross margin >74%"
  - date: 2026-06-19
    event: "GTC Asia"
    look_for: "Rubin platform shipping timeline, Blackwell Ultra mass production update"

price_outlook:
  current: "$225"
  base_90d: "$215-235"
  bull_90d: "$245-260 (Q1 earnings beat + China revenue confirmed)"
  bear_90d: "$190-205 (Q1 earnings miss OR Samsung strike impact OR continued 13F selling)"

Self-Check (All 5 Must Be Yes for C2 Completion)¶

view is not "not sure" — you chose a specific direction
Each support has numbers + a source
Each red_flag has an observable trigger
At least 1 catalyst is within 90 days
price_outlook is anchored to the current price

Any no → go back and fix it.

6. What's Next¶

Your YAML will throw up a bunch of terms:

Data Center segment — what is it?
fwd PE > 35x — what's fwd? Is 35x high?
RPO $553B (ORCL example) — how is this different from backlog?
gross margin >74% — what's the industry benchmark?

Without knowing these, you can't write supports with numbers.

→ L2-C3 · Valuation/Earnings Glossary — First categorize these terms into which of the 4 dimensions they belong to, then read real numbers.

7. Deep Dive (Optional): Why Force Triggers / Force Catalysts?¶

Click to expand the design trade-offs for triggers and catalysts

Why must red_flags have triggers?

A red_flag without a trigger is as good as unwritten — you'll never know when it's triggered. Example:

"Valuation is high" (always true, never triggers, equals a disclaimer)
"If fwd PE > 35x AND hyperscaler capex guide is cut, valuation compression risk triggered" (observable, actionable, can stop out)

A trigger turns risk from emotion into executable alert.

Why must there be a catalyst within 90 days?

No catalyst = the thesis is never validated. Looking back 3 years later, you won't know if you were right or wrong — you learn nothing.

Forcing one catalyst within 90 days pushes you to get slapped or rewarded by the market every quarter, so:

Within 90 days, you see that 1 outcome → you learn which supports were real and which were fake
Compound over 4 quarters → 1 year of calibration, your investing ability actually grows

Without this feedback loop, years of investing just mean repeating the same guesses.

Why force a single-point view instead of a "range view"?

"I see NVDA as 50% bull / 50% neutral" — this is avoiding the view. You're either bull (build a position) or watching (don't build) — there's no third state. confidence: low already covers the "not sure" feeling — no need for a "range view."