Despite the common focus on bottom-up fundamental analysis, statistical evidence shows two-thirds of an investment manager's relative performance is determined by macro factors, such as whether growth or value stocks are in favor. Ignoring top-down signals like Fed policy is a significant mistake, as it means overlooking the largest driver of returns.

While human analysts think linearly (e.g., higher oil -> inflation -> higher rates), LLMs process repercussions simultaneously across many dimensions (e.g., impact on ethanol, drillers, producers, yield curve). This allows for a much faster and more comprehensive understanding of market events.

Investment gains often come from "multiple expansion," where the market's perception of a business improves, causing it to trade at a higher valuation. This sentiment shift is frequently more impactful than pure earnings growth, and underestimating it is a primary reason for selling winning stocks too early.

A company can beat earnings and still see its stock fall if its actions (e.g., high CapEx) contradict the prevailing market narrative (e.g., the AI bubble is popping). Price is driven by future expectations, not just present-day results.

Widespread use of similar AI models by average investors will likely lead to herd behavior and crowding in certain securities. This pushes prices away from fundamental value, creating predictable inefficiencies and new alpha opportunities for sophisticated investors who can model these effects.

An average stock's return is dictated more by external forces than company performance: 40% by the market and 30% by its sector, with only 30% attributable to idiosyncratic factors. This means correctly identifying a winning sector is nearly as valuable as picking the best stock within it.

CFM operates on the belief that in the short-to-medium term (up to a year), market prices are driven primarily by investor flows, not fundamental value. This "inelastic market hypothesis" means their strategy focuses on predicting what people will buy and sell, rather than analyzing company balance sheets.

An asset's price is ultimately determined by what someone is willing to pay, making the market a game of predicting collective human emotion, much like trading baseball cards. Even fundamentally sound assets can crash if sentiment turns negative, meaning investors are gambling on the emotional state of others.

As AI masters the analysis of financial filings and transcripts, the source of investment alpha may shift to information that is difficult for models to process. Qualitative insights from attending conferences, judging a CEO's character via a handshake, or other forms of scuttlebutt could become increasingly valuable differentiators for human investors.