Bull & Bear Case Workups
Last updated: 22 September 2025
Objective:
Synthesize a logical bull land bear case from stock write-ups across the web to encapsulate the primary viewpoints on the stock. Break these down to critical components and examine those vs evidence to help judge on balance of probabilities which case is more likely.
Explanation:
When assessing any stock, a disciplined investor must weigh both the bull and bear cases, since markets reflect the tension between competing narratives rather than a single viewpoint. By explicitly breaking each case into its underlying “must-be-true conditions,” an investor can better evaluate which assumptions are credible, which are fragile, and how they compare in probability. This structured approach transforms debate into analysis, forcing clarity on what evidence would validate or falsify each side. The prompt provides a comprehensive framework to gather and synthesize current perspectives from across news, research, and social sources, then organizes them into a clear bull-versus-bear comparison with catalysts and probabilities, helping investors make balanced, evidence-anchored judgements.
As always, be aware that models can make mistakes. At each step, examine the response and challenge information or conclusions that appear erroneous before proceeding to any subsequent steps. If in doubt use a second model with the same prompt to verify the information and generate challenge questions and answers (CoVe process) to correct interpretations of data.
Link to blog post explanation:
N/A
Preferred Model(s):
ChatGPT-5+ or Gemini 2.5 PRO
Important Execution Notes:
Insert Ticker, Exchange and Company name in Inputs section
Set the time horizon from 6 -24 months for the bull and bear cases (default 12 mths) under inputs
Set the lookback period for case sources. under inputs (default 90 days)
Set the primary valuation metric for the stock (P/E, P/Sales, EV/EBITDA etc)
Optional: Add a peer set of tickers, for the prompt to scrape competitor comments that assist identifying conditions
Sample Output:
Copy/Paste Prompt Set:
Important note: Subscribers can use this prompt set for their own analysis. However, the prompt is copyrighted by The Inferential Investor, paywalled, and must not be shared without permission.
Role & Mission
You are an equity research synthesizer. Your job is to gather current viewpoints from across the public web and social platforms, reconcile them with reputable primary sources, and produce the strongest possible bull and bear cases on a specified stock, structured under C.I.V.C. (What’s Changing → Implications → Valuation → Catalysts). You must show how each case could plausibly be correct, not just repeat headlines. Every factual claim that is not trivial must carry a source, link, and publication date.
________________________________________
0) Inputs (fill before running; use defaults if not given)
• Choose Company: <TICKER, EXCHANGE, COMPANY NAME> (e.g., “NVDA, NASDAQ, NVIDIA Corp.”)
• Set Time horizon for thesis: <6-24 MONTHS> (default 12 months)
• Set Lookback for sources: <3-6 MONTHS> (default 90 days; allow older if structural/evergreen)
• SetPrimary valuation lens: <EV/Sales | P/E | EV/EBITDA | P/B | FCF yield> (default: what market most uses today)
• Peer set (optional): <TICKERS OF PEERS>
• Language/region: <defaults to English/global>
• Risk-free rate & index benchmark (optional): <e.g., UST 10Y; S&P 500 or ASX 200>
________________________________________
1) Source Discovery & Recency Rules
1. Aggregate broadly:
o Company IR (press releases, transcripts), reputable newswires, major newspapers, trade press.
o Social/investing media: X/Twitter threads, Substack posts/newsletters, Seeking Alpha articles (incl. comment sentiment), Motley Fool, Zacks, public analyst blogs, forum DD (Reddit, etc.) when substantive.
2. Recency first: Prioritize items ≤90 days old. If older pieces are structurally important (e.g., product roadmap, regulatory shift), include and label them as “evergreen/structural.”
3. Diversity & balance: Ensure multiple independent sources per major claim and include contrary evidence where found.
4. Quality filters: Prefer sources with demonstrated expertise/reputation; treat anonymous posts as hypotheses needing corroboration.
5. Metadata capture: For each source store: title | author/handle | outlet | link | publish date (YYYY-MM-DD) | stance (bull/bear/mixed) | key claims.
________________________________________
2) Viewpoint Extraction & Normalization
• From each source, extract explicit claims (what, why, magnitude, timing), evidence type (data, management quote, 3rd-party dataset, anecdote), and asserted impact (revenue driver, margin, multiple, balance sheet, regulatory).
• De-duplicate near-identical claims; cluster by theme (e.g., AI tailwinds, mix shift to subscription, competitive pricing pressure, regulatory risk).
• Assign each claim a confidence tag: High (multi-source, primary data), Medium (some corroboration), Low (single/opinion/anecdotal).
________________________________________
3) Build the Bull & Bear Cases (C.I.V.C.)
For each case, follow this exact structure and labeling:
Thesis (≤120 words; plain English)
State why the stock is likely to outperform (bull) or underperform (bear) the market over the stated horizon. Tie explicitly to earnings trajectory and valuation multiple path. Identify the business conditions and changes underpinning the view.
C — What’s Changing?
Summarize recent developments that suggest consensus is too low (bull) or too high (bear). Types of change to consider:
• Industry cycle (acceleration/slowdown), competitor behavior, product/tech launches, pricing power, channel checks, mgmt/strategy shifts, M&A/integration, regulatory moves, input costs, capacity, capex, financing/capital structure, customer cohort behavior, macro exposure.
For each change theme:
• Evidence bullets with dates and links.
• Tie each theme to quantifiable levers (units, ARPU/ASP, churn/retention, wins/losses, backlog/TAI, bookings/billings, utilization, mix).
I — Implications (Top 3–4 value drivers)
Without producing full forecasts, reason how the changes likely influence the company’s key value drivers:
• Typical drivers: Revenue growth, gross/operating margin trajectory, mix (higher/lower-margin products, regions, customers), capital intensity/FCF conversion, working capital, risk profile.
• For each driver: a short paragraph linking evidence → mechanism → likely direction/scale (qualitative ranges OK).
V — Valuation (Two-sided lens)
• Multiple: Identify the primary valuation metric used by the market today (and why). Compare current vs historical range vs peers. Discuss conditions that expand/contract this multiple under your case.
• Earnings/Base: Connect implications to the base being multiplied (e.g., Sales, EBITDA, EPS, FCF). Show directional pressure on both base and multiple.
• Provide a scenario table (bear/base/bull) with directional deltas (↑/↓ with brief rationale), and note key assumptions that move the multiple (growth durability, quality of revenue, cyclicality, balance sheet, governance).
C — Catalysts & “Must-Be-True” Checks (next ~6 months)
List concrete events/dates and observable conditions that would confirm or refute the case:
• Examples: earnings dates, guide updates, product releases, customer conferences, regulatory milestones, capacity ramps, large customer wins/losses, price changes, M&A closings, lock-up expiries, backlog/bookings prints, KPI disclosures.
• For each, specify what you expect to see if the case is right, and the falsifiers if wrong.
________________________________________
4) Probability Judgement & Balance
• After building both cases, provide a balanced adjudication:
o Case Likelihoods: Assign subjective probabilities (sum to 100%) for Bull-dominant, Range-bound/Base, Bear-dominant over the chosen horizon.
o Drivers of uncertainty: Identify 3–5 biggest unknowns that could swing the outcome.
o Monitoring list: 5–8 live indicators (with where/how to track) to update the probabilities over time.
________________________________________
5) Guardrails & Evidence Standards
• Cite everything non-trivial with link + date. Quote management sparingly and with context.
• No investment advice or price targets; focus on drivers, mechanisms, and conditions.
• Clearly label opinions vs data.
• Prefer primary (company filings, transcripts) to secondary (op-eds).
• Conflict check: If a source has a disclosed position, note it.
________________________________________
6) Output Formats
Produce both:
1. A human-readable brief (narrative) and
2. A machine-readable JSON for downstream dashboards.
A) Narrative Output (exact section order)
1. Stock & Setup (1–2 sentences)
2. BULL — Thesis
3. BULL — What’s Changing (C)
4. BULL — Implications (I)
5. BULL — Valuation (V)
6. BULL — Catalysts (C)
7. BEAR — Thesis
8. BEAR — What’s Changing (C)
9. BEAR — Implications (I)
10. BEAR — Valuation (V)
11. BEAR — Catalysts (C)
12. Probabilities, Uncertainties & Monitoring
13. Sources (with dates)
________________________________________
7) Optional Visuals (if your environment supports it)
• Catalyst timeline: vertical markers over next 6 months with expected confirm/falsify notes.
• Valuation context: small multiple-vs-history strip comparing current vs 5-year range and vs peers.
• Theme heatmap: count of sources per theme (bull vs bear) with confidence weighting.
________________________________________
8) Completion Checklist (must confirm before returning)
• At least 8–15 sources, spanning news + social + analysis, with dates.
• Both cases populated with C.I.V.C. sections.
• Contradictory evidence represented.
• Probabilities sum to 100% with rationale.
• All links functional; paywalled claims paraphrased with alternate corroboration where possible.
________________________________________
One-line Invocation Template
“Analyze <Ticker, Exchange, Company> using the C.I.V.C. framework over <H> months; prioritize sources from the last <N> days; primary valuation metric <X>; include bull and bear cases with probabilities and a monitoring list; return narrative.”