How Investment Decisions Actually Get Made

Most investors, if asked to describe how they make investment decisions, will produce a coherent and rational account. They identify an opportunity. They analyse the fundamentals. They assess the risk. They size the position appropriately. They monitor it against the original thesis.

This account is not wrong. But it is incomplete — and the gap between the account and the reality is where most investment errors are generated.

The research on decision-making under uncertainty is extensive and consistent. Human judgment, operating under conditions of incomplete information, time pressure, and emotional stakes, deviates from the rational model in predictable ways. The deviation is not random. It follows patterns that have been documented across thousands of studies, across different cultures and professional backgrounds, and across different levels of experience and expertise.

Investment professionals are not exempt.

The most well-documented pattern is the tendency to weight recent information more heavily than its relevance warrants. A period of strong performance creates an expectation of continued strong performance. A period of poor performance generates pessimism that outlasts the underlying conditions that drove it. Both directions produce systematically skewed assessments — and both tend to manifest in portfolio decisions that are made at the wrong time.

The second pattern is the asymmetry between how gains and losses are experienced. Losses register more powerfully than equivalent gains. This asymmetry distorts the decision to hold or sell: investors hold losing positions longer than they should, in the hope of recovering to breakeven, and sell winning positions too early, to secure the psychological benefit of a confirmed gain.

Neither of these patterns is the result of ignorance. They persist in experienced professionals who are fully aware of the research. They are features of the cognitive architecture that human decision-making runs on — not bugs that can be patched by knowing they exist.

What can be changed is the structure within which decisions are made.

A well-designed investment process addresses these patterns not by eliminating the psychology — which is not possible — but by removing the conditions that allow it to determine outcomes. Pre-commitment to criteria before a position is entered. Documented reasoning that is referenced at review, not reconstructed from memory. Separation of the decision to hold from the price at which the position was established. Structured review processes that evaluate outcomes against the original thesis rather than against the current market price.

Each of these structural features does the same thing: it introduces a version of the investor's own reasoning — captured at a point of relative clarity, before the position was exposed to the market — as a counterweight to the real-time psychological pressures that tend to distort subsequent decisions.

The difference between investors who improve over time and those who merely accumulate experience is not intelligence. It is the willingness to build and maintain this kind of structure — and the intellectual honesty to use it when it produces conclusions that are uncomfortable.

Experienced investors who have operated across multiple market cycles tend to develop informal versions of these structures through hard-won pattern recognition. The cost of learning these lessons informally is the tuition paid to the market over years of suboptimal decisions.

The alternative is to design the process before the lessons are learned — to build, in advance, the structure that would otherwise only be understood in retrospect.

That is what a genuine investment process is. Not a set of rules that govern decisions, but a structural environment that makes better decisions more likely than worse ones — not because the investor is always rational, but because the architecture of the process does not depend on them being so.