P2B-C1 · What Problem Does Your Stock Solve¶

Key Insight

First WHAT, then WHY — 90% of stock mistakes come from not clearly stating what problem it solves.

Layer 2 · Analyzing a Stock — After Industry, Now Down to One Stock

This is Chapter 1 of Layer 2 (5 chapters total). After this chapter, you can explain in 30 characters why you're looking at a stock.

1. The Problem: Shouting "NVDA bull" Without Knowing What You're Betting On¶

KOL says "NVDA is bullish." You ask: "Bullish on what?" Him: "AI" You: "Which part of AI?" Him: "...just AI"

→ This is failing to articulate WHAT. When it drops 5%, you don't know if the thesis is broken or it's noise. When it jumps 20%, you don't know if you were right or if hyperscaler capex exceeded expectations (you weren't betting on capex at all).

90% of wrong calls come from this — not bad analysis, but never clearly defining what you're analyzing.

2. The Solution: Write One Sentence First¶

Before opening any earnings report / reading any KOL / running any valuation, write 1 sentence in your draft:

I look at [TICKER] because [WHAT problem is solved] + [WHY I'm more confident than the market]

Examples:

✅ I look at NVDA because hyperscaler 2026 capex +60% + market fears ASIC substitution but Blackwell is locked in for 12 months
✅ I look at ANET because Meta/MSFT data center east-west traffic explosion + market treats it as NVDA side dish at forward PE of 28x
❌ I look at NVDA because AI is the future (WHAT unclear, WHY lacks differentiation)
❌ I look at ANET because its backlog is strong (doesn't say WHY buy now vs 12 months ago)

This 1 sentence is the seed for everything in C2-C5. Can't write it = no thesis yet, shouldn't open a position.

3. How It Works: Thesis 4 Dimensions — The Backbone of Layer 2¶

Starting from C2, this 1 sentence expands into Thesis 4 Dimensions:

Dimension	Position in 1 Sentence	Full Form After C2
WHAT	"Problem solved"	Core narrative of the company/sector right now (product / customer / financial event)
WHY	"You're more confident than the market"	Why market pricing hasn't fully reflected — your edge
SO WHAT	(Implied in whether you dare to buy)	Price landing — base/bull/bear 90-day range + trigger conditions
RISKS	(Not yet in 1 sentence)	Anti-thesis trigger conditions — what happens that makes you close

graph LR
    Q[Your 1 sentence] --> W[WHAT narrative]
    Q --> Y[WHY difference]
    W --> S[SO WHAT price]
    Y --> S
    W --> R[RISKS anti-trigger]
    Y --> R

The next 4 chapters each touch only 1 specific detail within these 4 dimensions. The backbone stays unchanged, content grows. This is the invariant of Layer 2 — don't forget it.

4. vs Layer 1 You Already Know¶

Layer 1 taught you:

AI industry chain 5 roles (upstream wafer / midstream accelerator / downstream compute / application customer / capital)
NVDA's position in "midstream accelerator"
Its relationships with TSMC / ASML / OpenAI / Meta

But Layer 1 doesn't tell you why this stock is worth buying / not buying right now. "Industry position" ≠ "investment thesis". NVDA has been in that same position — $14 in 2022, $230 in 2026, same position, completely different price.

Layer 2 fills exactly this 1 gap: from "I know what it is" to "I have 1 sentence explaining why I act now."

5. Try It: Write Within 30 Characters¶

Task: Pick 1 stock you know best, fill in the template:

I look at ___ because ___ + ___

Hard rules:

Total length ≤ 30 characters
WHAT must be a specific event / data point — no generic terms like "AI" / "growth stock" / "good cash flow"
WHY must include "how the market is pricing it" — not just your own opinion

Self-check (if passed, proceed to C2; if not, revise):

Check Item	Pass Condition
WHAT specificity	You can follow it with 1 number or 1 specific time
WHY differentiation	Show to 1 friend — they can repeat the "market vs you" disagreement
Length	≤ 30 characters

Can't write it = you're still in Part 1, shouldn't move to C2. Go back and review P1-C6 · Industry Chain 5 Roles.

6. What's Next¶

1 sentence is too rough — in practice you need:

WHAT broken into 3-5 specific pieces of evidence (supports)
RISKS broken into trigger conditions (red_flags)
SO WHAT calculated as 90-day price range (price_outlook)

→ L2-C2 · 4-Dimensional Thesis Framework expands this 1 sentence into a full yaml — this becomes your standard format for discussing any stock.

7. Deep Dive (optional): Why Not DCF / Multi-Factor Models?¶

Click to see the design trade-offs of the 4-dimensional framework

Textbooks / sell-side / quant each have their own frameworks. AI investment thesis doesn't use them directly because:

vs DCF: DCF requires 10-year cash flow assumptions + WACC. AI companies are invisible even 3 years out — the precision from DCF is false. This framework replaces it with 90-day price range + trigger conditions.

vs Multi-factor quant: Quant models (Fama-French / Barra) are cross-sectional explanations, not "why this stock right now." This framework pursues "why this stock is different right now."

vs Sell-side ratings: Buy/Hold/Sell are single points — they don't say what happens if wrong. This framework forces RISKS to list trigger conditions.

Core trade-off: This framework sacrifices "numerical precision" for "people can align in discussion + know where they went wrong when wrong." In investing, the latter is more valuable.