Risk Savvy: How to Make Good Decisions: Summary & Key Insights

When people hear about probability, their minds usually turn to randomness and complexity—concepts reserved, perhaps, for mathematicians. But in reality, the misunderstanding of risk has concrete, often painful consequences. Much of my research has shown that even experts, such as doctors or financial analysts, routinely misinterpret statistical information. For example, when physicians are told that a test has a sensitivity of 90% and a specificity of 91%, most fail to compute what those terms mean for a real patient sitting in front of them. It’s not because the doctors are careless or uninformed, but because statistical education rarely connects abstract percentages to human realities.

One of the most powerful sources of confusion lies in relative risk. When headlines scream that “a treatment doubles your chance of survival,” it sounds dramatic. Yet without knowing what the original probability was, the statement is meaningless. Consider two scenarios: a drug increases survival from one in a thousand to two in a thousand—that’s a 100% relative increase, but an absolute gain of only one person out of a thousand. Conversely, a change from fifty in a hundred to seventy-five in a hundred may represent a smaller relative increase but a profound difference in real outcomes. I always urge readers to translate relative risk into absolute numbers. Doing so changes decisions. When patients see the actual probability instead of the relative comparison, their anxiety often drops, and their choices become more balanced.

Another illusion arises from conditional probabilities—the way information is presented in complex forms like ‘the probability that a person with a positive test result actually has the disease.’ In medicine, people typically overestimate the reliability of screening tests because of this framing. If a disease affects one in a thousand people and a test has a 99% accuracy rate, intuition falsely suggests that a positive test almost certainly means the patient has the disease. In reality, the chance could be closer to one in ten. Represented as natural frequencies—"Out of every thousand people, one has the disease. Of those, nearly all test positive. But among the 999 healthy people, about ten also test positive"—the signal becomes clear. Represent numbers as counts, not percentages, and your brain processes risk more intelligently.

When faced with uncertainty, we often assume that complex algorithms must yield better decisions. Yet I have spent decades showing the opposite: simple heuristics—rules of thumb—can outperform elaborate statistical methods in real-world settings. Humans evolved to make quick judgments in uncertain environments where time and information are limited. The cognitive shortcuts we use—like the recognition heuristic (if you’ve heard of one option and not the other, pick the recognized one)—often reflect adaptations that work efficiently, not flaws in reasoning.

In *Risk Savvy*, I explain how doctors, investors, and entrepreneurs often succeed when they rely on well-calibrated intuitions rather than exhaustive analysis. Take the example of emergency medicine. A physician deciding whether to admit a patient with chest pain doesn’t always calculate every variable; instead, she might use a simple tool: if the patient’s ECG and history match certain patterns, she admits without further testing. Such heuristics save time and lives. Studies have shown that algorithmic overreliance—trusting computer models without understanding their assumptions—can lead to worse diagnoses than relying on practiced intuition.

The key lies in knowing which environments favor which type of thinking. In highly uncertain contexts—where the future cannot be precisely predicted—heuristics provide robustness. In stable, well-defined environments, complex calculations may shine. Understanding this distinction is part of being risk savvy. You learn to distinguish between chance and ignorance, between genuine randomness and misplaced precision. I often remind readers that our brains were never designed for perfect rationality—they were designed for adaptive rationality, capable of thriving amid uncertainty.

To become confident in decision-making, we must respect the power of simplicity. Good heuristics are not anti-scientific; they are empirical distillations of experience. When I trained managers to make quick decisions about investment portfolios, the most successful ones followed a heuristic known as the ‘1/N rule’: divide resources equally among the most promising options instead of relying on elaborate optimization. Over decades, this simple rule outperformed complex models that depended on data the future rendered obsolete.