Hoc narratives from genuine signals · Steravindal

Why the quality of your research process matters more than you might think

Markets are extraordinarily good at producing noise that sounds like meaning. Every session ends with a price, and within minutes there is usually a confident account of why that price landed where it did. A sector fell because of a policy announcement. A company rose because sentiment shifted. An index dipped because traders were cautious ahead of a data release. These explanations feel authoritative because they arrive quickly, they use real events as their raw material, and they are delivered by people who sound as though they were watching closely. But there is a structural problem buried inside this process. The explanation is almost always constructed after the fact, assembled from whatever happened to be visible on the day and shaped to fit the outcome that actually occurred. If the price had moved in the opposite direction, a different explanation would have been assembled from the same pool of events, and it would have sounded equally plausible. This is not dishonesty on the part of commentators. It is a natural feature of how human beings make sense of complex systems. The difficulty for an investor is that a story built to explain yesterday's movement carries very little reliable information about tomorrow's.

A signal, by contrast, is something different in kind rather than just in quality. It is a piece of information that has a plausible, testable relationship to future outcomes, and that relationship exists independently of whether a price has already moved. A genuine signal might be a persistent divergence between what a business is earning and what the market appears to be pricing in. It might be a structural change in how an industry is organised that has not yet been widely discussed. It might be a pattern in how a particular type of company tends to behave during a specific phase of an economic cycle, observed across many instances rather than just one or two. What makes these things signals rather than stories is that they can be examined before the outcome is known, they can be challenged with contrary evidence, and they carry some logical reason to expect a particular direction of effect. A story, by contrast, usually only becomes visible after the outcome has arrived. It explains rather than anticipates, and it tends to dissolve under pressure if you ask whether it would have predicted the opposite result equally well.

One practical way to sharpen this distinction in your own research is to ask a simple question about any piece of market commentary you encounter: could this explanation have been written before the price moved? If the answer is no, or if you suspect the same writer would have produced a convincing alternative explanation had things gone differently, then you are almost certainly reading a story rather than a signal. This is not a reason to dismiss the commentary entirely. Stories can still be useful. They sometimes surface genuine risks or structural changes that deserve further investigation. They can help you understand how a broad community of investors is currently interpreting events, which is itself a form of information about sentiment. But they should be treated as starting points for your own analysis rather than conclusions. The discipline of separating what you know from what has merely been narrated to you is one of the more demanding habits to build, because the narratives are often fluent and the signals are often quiet and incomplete.

Uncertainty is the environment in which all of this takes place, and it is worth treating it as a feature of the landscape rather than a temporary inconvenience. Markets do not resolve into clarity once you have read enough. The most experienced researchers and the most rigorous institutional processes still operate under conditions where many important things are genuinely unknown. What changes with experience and method is not the elimination of uncertainty but the ability to be honest about which parts of your reasoning rest on evidence and which parts rest on inference or assumption. When you encounter a market narrative that feels compelling, the most useful thing you can do is identify what would have to be true for it to be correct, and then ask whether those conditions are actually observable or whether they are themselves part of the story. This habit of testing the foundations of an argument, rather than evaluating how convincingly it is told, is closer to what careful investment research actually involves. It does not make decisions easy or outcomes certain. It does make the basis for your thinking more visible to you, which is where honest independent research has to begin.