A structured framework for distinguishing between “good stock at a fair price” and genuine Alpha

I've been actively investing for going on 20 years. Over that time I've developed an analytical process that has significantly outperformed double the S&P 500. The core insight isn't complicated – most of the value in a portfolio comes from a small number of positions where you understand something structural that the market hasn't properly priced. Everything else is risk management.

The problem is that most analytical processes, including every stock screener I've ever used, are good at eliminating bad investments but structurally unable to identify exceptional ones. A screen can tell you that a company has a high P/E. It can't tell you whether that P/E is high because the stock is overvalued, or because the market is comparing it to the wrong peer set and hasn't grasped the magnitude of what's happening.

That distinction is the difference between what I call a Plus position (a good company at a fair price, identified through standard fundamental analysis) and an Alpha position (a company where multi-level analysis reveals a specific gap between structural reality and market perception).

I formalized the process I use into a structured analytical framework. It covers six phases:

1. deeply learning a company before forming any view, 
2. assessing every data point at multiple meta-levels (surface, pattern, structural, market perception), 
3. identifying structural catalysts that would force repricing, 
4. determining whether a gap actually exists between your understanding and the market's price, 
5. building what I call a "load-bearing framework" that tests each critical assumption before you commit capital, and 
6. managing the position through its lifecycle (what to monitor, when to reduce, when to exit).

A few things it addresses that I don't see discussed much in value investing communities:

The wrong peer set problem. If a company is one of only two entities on Earth that can deliver a particular capability, comparing its valuation to a broad industry average is analytically meaningless. The correct comparison is to the other entity in its structural category. This mistake alone causes more misjudgments than almost any other analytical error.

The meta-level problem. A product delay can mean "execution failure" (Level 1) or "engineering discipline in a company that consistently delivers eventually" (Level 2) or "timing shift that doesn't change the addressable market or competitive position" (Level 3). Most investors read Level 1 and stop. The gap between Level 1 and Level 3 is where some of the best opportunities hide.

The load-bearing test. Before committing conviction-level capital, every structural support in the thesis needs to be tested individually. You identify which open questions are load-bearing (thesis breaks if the answer goes wrong) vs. secondary, research each one from primary sources, and maintain a running inventory of what's resolved and what isn't. If a critical element is still unresolved, you don't have an Alpha thesis yet. You have a hypothesis.

Position management. Sizing is about probability-weighted risk-return, not just "how much can I afford to lose." The load-bearing work you've done directly informs your probability estimate: more resolved elements means a tighter distribution. And exit decisions are driven by structural changes (a load-bearing element breaks, the gap closes, the thesis mutates), not price movements or technical signals.

I've made the full framework available as a free download in two formats:

PDF (works for anyone, also functions as an instruction set if you paste it into ChatGPT, Claude, or Gemini for AI-assisted analysis): Download PDF

Claude skill file (installs directly into Anthropic's Claude and loads automatically when you ask it to analyze a stock): Download .skill file

Interested in feedback, pushback, and how others approach the Plus vs. Alpha distinction in their own value investing selection process.

Disclaimer: Nothing in this framework constitutes investment advice. It's an analytical process, not a recommendation to buy or sell any security. This is not a commercial product. There is nothing for sale, no paywall, no email signup, and no monetization of any kind. The framework is free and complete as downloaded.