This group is not a place to recommend promising foreign stocks or trade together. It aims to become an analytical community that systematically examines the basis for an investment judgment, the conditions that would disprove it, and the appropriate investment size. It connects three tools—Why Chain, Falsification, and 5-ring Prior Bayesian—to form and continually update investment hypotheses grounded in primary sources. Instead of judging only whether an outcome succeeded, the goal is repeated practice in making better decisions under uncertainty. 📈

1. The Group's Purpose and Three Investment Questions

The group defines itself as:

"A primary-source-based analytical community that quantifies the three essential stages of investment decision-making and continually corrects them with peers."

The article views investing as answering three questions. How an investor answers them defines that person's investment process. Rather than supplying the right answer, the group lets members share how they refine their own processes.

The first question is: "Why should I invest in this company?" It is not enough to conclude that revenue will rise or margins will improve. The investor must explain in stages why that outcome will occur, presenting the investment point, the scenario in which it is realized, and a direct chain of causality leading to the result.

The second question is: "Under what conditions does this scenario break?" Analysis must not seek only evidence that the investor is right. It should define in advance the signals that would establish that the investor is wrong.

"Do not collect only evidence that I am right; define in advance the signals that I am wrong."

Without predetermined data thresholds that would weaken the hypothesis, analysis can turn into an untestable belief or faith. Investors should continually read reliable sources and refine the conditions that would break the hypothesis.

The third question is: "Given that, how much can I bet at most?" Quantify the first two answers to calculate confidence in the hypothesis, then set an upper bound on position size. Even an attractive idea can produce a large loss if the investment is too large, so the validity of the logic and the actual amount invested must remain separate.

What the Group Will Not Do

This is not a stock-tip room or a collective trading group. Information can be shared, but each person makes and remains responsible for buying and selling decisions. Short-term trading ideas are not the main subject.

Instead, the group focuses on changes in industry structure and the fundamental changes companies undergo within them. Members share not a simple conclusion about a ticker, but the analytical tools and thinking used to reach it.

2. Three Core Analytical Tools

The group uses three tools to structure investment judgment. Each is independent but connects into one continuous flow in actual analysis.

Why Chain: The Causal Chain Behind an Outcome

Why Chain asks why something happens. It does not stop at observable results such as revenue growth, margin expansion, or a rising stock price, but divides the causes producing that result into stages.

If Nvidia's revenue rises, for example, the analysis should go beyond "AI is growing." Data-center GPU shipments increased because demand for AI training expanded; that demand connects to higher capital expenditure by big technology companies; that capital is then invested in leading AI companies and ultimately improves productivity for enterprise customers and consumers. The causal chain can be made explicit all the way through.

Drawing the chain reveals its weakest link. Questions missed by the vague belief that "AI looks good" become visible. Why Chain sharpens the investment logic and helps identify the thesis's central risks.

Falsification: Conditions That Show a Hypothesis Is Wrong

Falsification records in advance the conditions under which each Why Chain link breaks. This differs from general pessimism or a vague risk review. It sets a specific threshold at which the investment hypothesis will be revised or discarded.

For example, an investor might define in advance that if advertising revenue growth at big technology companies falls below 5% for two consecutive quarters, downward pressure may emerge in their capital-expenditure cycle. Observable data and criteria let the investor judge the thesis objectively instead of clinging to belief.

"The central goal is to keep the analysis a living hypothesis."

Falsification prevents investment analysis from becoming a fixed assertion and keeps it as a hypothesis continually revised with new information.

5-Ring Prior Bayesian: Connecting Confidence to Position Size

5-ring Prior Bayesian answers, "So how much can I buy?" Divide one investment thesis into five independent hypotheses and assign each a confidence level between 0 and 1. Then calculate total confidence by multiplying, not adding, the five probabilities.

Multiplication is used because the whole thesis may weaken substantially if any essential hypothesis fails. The resulting composite confidence sets the upper bound for the position.

An important lesson is that a strong investment argument does not automatically justify a large position. A thesis may look 70% likely, but if that 70% arises from several combined hypotheses, the investable size may be far less than 70% of an average position. Separating confidence from size helps avoid major long-term losses.

How the Three Tools Connect

The tools proceed in order:

1. Use Why Chain to explain why to invest and create five causal links.
2. Use Falsification to define the condition and threshold that breaks each link.
3. Use 5-ring Prior to multiply confidence in the five hypotheses and calculate total confidence and the position-size ceiling.

The five Why Chain links become the five hypotheses in the 5-ring Prior. Falsification conditions become the criteria used to update priors in the next quarter. These are not separate assignments, but one investment analysis expressed in three dimensions: causality, disproof, and position management.

3. Priority of Learning Materials

The article assumes that analytical reliability is proportional to source reliability. The group follows a three-tier hierarchy and centers its work on primary and official company sources.

Tier 1: Legal Filings

The highest-priority materials are legal filings such as 10-K, 10-Q, annual reports, and material-event reports. They are the starting point for every analysis and deserve the most reading time.

AI may help read and organize them, but the investor must personally read and verify the originals. Presentations and analysis should clearly cite the page and exact original language.

Tier 2: Official Company Materials

The second tier is material directly provided by the company, including conference calls, earnings-call transcripts and Q&A, IR materials, and Investor Day presentations.

These sources are particularly useful for Why Chain. They reveal the causes management gives for revenue growth, the structural reasons margins may improve, and changes expected in the next quarter.

Tier 3: Industry Data and External Research

The third tier includes industry data and outside analysis: industry news and research from sources such as the Financial Times, TrendForce, IDC, BloombergNEF, Wood Mackenzie, Counterpoint, and Semianalysis, as well as sell-side reports and deep analysis.

They help identify price trends, market-share changes, and global capital flows companies do not state directly. The key principle is to read from the top of the hierarchy downward.

"If you look at outside analysis before sufficiently reading primary sources, you will be pulled into someone else's narrative."

Analysis from outside experts and communities can be useful for cross-checking, but should not become the investment basis itself. The group does not share other people's summaries; it shares analysis members have drawn directly from primary sources.

4. Presentation Format

A presentation is not simply a time to declare, "This stock is good." Its purpose is to expose the presenter's reasoning and find weaknesses in the process.

It follows four stages:

1. Conclusion and multiple anchoring
   State an opinion of OW (Overweight), N (Neutral), or UW (Underweight), then present the investment basis and valuation. This answers, "What is the conclusion?"

2. Why Chain
   Divide the investment logic into five causal links. This answers, "Why invest?"

3. Falsification
   State the condition and threshold that breaks each link. This clarifies, "What would have to break for me to be wrong?"

4. 5-ring Prior
   Estimate a prior probability for each of the five hypotheses, then derive total confidence and a maximum position size. This answers, "How much can I buy?"

Across repeated sessions, presenters should show how their analytical process develops. The goal is not receiving confirmation that the analysis is correct, but finding vulnerabilities and applying them to real investment decisions.

5. Recommended Books and the Foundation of Bayesian Thinking

The appendix recommends books supporting the group's way of thinking. Their shared subject is distinguishing signal from noise, decomposing hypotheses, recording priors, and updating with new information.

Nate Silver, The Signal and the Noise

This book is first about how to define a problem. Its central message is that predictions fail not chiefly because a model is insufficiently complex, but because people overfit data without distinguishing real signal from random noise.

Through baseball statistics, weather forecasting, seismology, poker, and political prediction, it shows how difficult it is to identify which variables are genuine causal drivers and which are accidental noise.

It helps explain the direct foundation of Why Chain. If an investor cannot distinguish whether a company's revenue growth comes from a true causal link or temporary quarterly noise, short-term data fluctuations will control the judgment.

It also explains the Bayesian concepts of prior, likelihood, and posterior intuitively without equations, making it suitable as an introduction to the mathematical background of the 5-ring Prior.

Philip Tetlock, Superforecasting

This book asks what kind of person forecasts more accurately. A large forecasting tournament conducted over 20 years revealed consistent differences in accuracy even among people with similar information, education, and expertise.

The difference was not innate intelligence or information quantity, but repeatable habits of thought. The group emphasizes two habits of superforecasters:

• Break large questions into small hypotheses.
  Divide a vague judgment such as "This company will improve" into revenue-growth probability, margin durability, and competitor-response probability, assign each one a probability, and combine them.

• Record in advance the conditions that would establish you are wrong.
  If data reaches a defined threshold, update the prior instead of clinging to the existing view.

The book is also important because it shows that these habits can be trained rather than being innate talents. Repeated practice of those habits is likewise the group's goal.

Annie Duke, Thinking in Bets

Annie Duke explains how to embody this way of thinking in everyday judgment and real decisions. A World Series of Poker champion and decision-theory scholar, she gives practical explanations of the psychological traps people face under uncertainty.

The most important insight is that decision quality must be separated from outcome quality. A good decision can produce a bad outcome, and a bad decision can produce a good result through luck. Evaluating judgment only by the result can make us mistake lucky success for skill or random failure for a lack of ability.

"A good decision can create a bad result, and a bad decision can create a good result."

The remedy is to record the prior and reasoning at the time of the decision. This is why presenters document Why Chain, Falsification, and 5-ring Prior. Later, they can evaluate which links behaved as expected and which assumptions were wrong, improving the next analysis.

Duke also recommends forming a truth-seeking group with peers. It evaluates decision processes rather than only outcomes and requires members to disclose priors and reasoning honestly. That aligns with the peer-review culture this study group seeks.

"The only way to create a consistent edge in decisions under uncertainty is to systematize the habit of explicitly externalizing your reasoning and evaluating it separately from outcomes."

Additional Recommendation: Damodaran on Valuation

For a quantitative foundation in valuation, Aswath Damodaran's Investment Valuation is also recommended. It may be burdensome to study deeply at the beginning, so the group plans to introduce the textbook's core concepts gradually through future presentations.

6. Conclusion

The global-equities study ultimately seeks not the correct answer for a specific ticker, but a better decision system for handling uncertainty. Why Chain externalizes the logic, Falsification manages the possibility of being wrong concretely, and 5-ring Prior connects confidence to actual position size.

The three recommended books emphasize the same principle: explicitly record priors and grounds for judgment, then update humbly as new information arrives. That habit is the foundation for becoming a better long-term investor.