AI Wealth Truth (22): Why Gut-Level Investment Decisions Can Be Better

I. We usually assume: the more careful the decision, the better. The more information, the better. The more complex the analysis, the more accurate. Behavioral economics overturns that intuition.

II. German psychologist Gerd Gigerenzer ran a famous experiment. He asked two groups to predict which stock would perform better. One group was finance experts using complex financial models and detailed data. The other group was ordinary people using one simple rule: "pick the company you have heard of". Result: the ordinary group's predictions were just as accurate, sometimes even more accurate.

III. How is that possible? The answer is information overload and overfitting.

IV. When you have too much information, you start seeing patterns that do not exist. You treat noise as signal. Your model can perfectly explain the past, but fails to predict the future. This is overfitting. Complexity is a double-edged sword.

V. In contrast, simple rules are often more robust. "Pick the company you have heard of" is crude. But it contains an implicit signal: companies you have heard of are usually larger, more stable, and more established. Those traits often correlate with performance. Simple rules are rough, but they capture the core factor.

VI. This is the less-is-more effect. In some situations, using less information and simpler rules produces better decisions. Not because simple is "good enough". But because simple can be better.

VII. Gigerenzer called these rules fast and frugal heuristics. They use minimal information to produce an adequate answer. They do not chase the theoretical optimum. They chase good enough and robust.

VIII. Some examples:

IX. The 1/N rule. How should you allocate a portfolio? Complex approach: Markowitz mean-variance optimization, which requires estimating returns, risks, and correlations. Simple approach: split money equally across N assets. Research shows: in many cases, a simple 1/N performs as well as complex optimized models. Because optimization needs many estimates, and estimation errors compound.

X. The recognition heuristic. When choosing between two options, if you recognize one and not the other, pick the one you recognize. It is blunt, but it works surprisingly well for predicting city populations, company size, and sports team performance. Because what reaches you is often what is large enough or important enough.

XI. The default option. Do not know what to choose? Choose the default. Defaults are usually set by designers as the most reasonable choice. Not deciding can be a rational decision.

XII. Why can over-analysis be harmful?

XIII. Reason 1: analysis paralysis. With too much information, you do not know where to start. You spend huge time collecting and analyzing, and still cannot decide. Opportunities die in hesitation. Perfect is the enemy of good.

XIV. Reason 2: overconfidence. The more complex your analysis, the more confident you feel. But confidence has no necessary link to accuracy. Complex analysis gives you false certainty. Then you size up and take larger risks.

XV. Reason 3: ignoring unpredictability. Markets, economies, and societies are complex systems. They are fundamentally not precisely predictable. No matter how complex your model is, it cannot forecast the next black swan. You are trying to precisely predict a system that cannot be precisely predicted.

XVI. Reason 4: information has a cost. Collecting information takes time and energy. Analyzing information takes cognitive resources. Those resources could be used elsewhere. If the marginal benefit of extra information is below its marginal cost, you lose money.

XVII. The AI-era paradox: more information makes judgment harder. AI can give you infinite information. Every financial metric, every news article, every social sentiment signal. But you cannot process it all. Information overload makes it easier to drown and harder to decide well.

XVIII. How do professional analysts and quant funds perform? Most underperform the index. They have the best information, the most complex models, the smartest people. Yet they still lose to the "mindless" strategy of buying an index fund. Complexity did not translate into returns.

XIX. How do you apply less is more?

XX. 1. Set a few simple rules, and obey them. For example: "Only invest in companies I understand." "No leverage." "Do not chase hot themes." These rules are crude, but they keep you away from most traps.

XXI. 2. Limit information intake. You do not need every headline. You do not need to dissect every earnings report. Pick a few reliable sources and ignore the rest. Reduce noise and let signal emerge.

XXII. 3. Decide periodically, not continuously. Do not check your account and decide every day. Set a fixed cadence, quarterly reviews for example. Lower decision frequency and raise decision quality.

XXIII. 4. Accept uncertainty. You cannot know everything. You cannot make the "optimal" decision. The goal is "good enough and robust". Chasing perfection can produce worse decisions.

XXIV. "Gut feeling" sounds unprofessional. But if your "gut" contains the right simple rules, a quick decision can beat a day of overthinking. Wisdom is not complexity. It is knowing when simple is enough. In the AI era, the barrier to complex analysis is lower, but the traps of complexity do not shrink. Sometimes, the smartest move is: do not try to be too smart.