âWe are notoriously poor judges of risk. We are terrified by dramatic, rare events while remaining largely indifferent to the statistical killers that surround us daily.â â Vaclav Smil
One of Smilâs most powerful and accessible chapters concerns risk - specifically, the systematic gap between the risks that frighten us most and the risks that actually kill us most. This gap is not random. It follows predictable patterns rooted in human psychology, media dynamics, and the architecture of the human nervous system.
Understanding actual risk requires the same quantitative literacy that Smil applies throughout the book: actual numbers, actual probabilities, actual fatalities - not vivid anecdotes or dramatic narratives.
Psychologists Daniel Kahneman and Amos Tverskyâs Nobel-winning research revealed systematic biases in how humans assess probability and risk:
We estimate the likelihood of an event based on how easily we can bring examples to mind. Dramatic, memorable events (plane crashes, terrorist attacks, shark attacks) are heavily covered by media and thus easily recalled - making us massively overestimate their frequency. Chronic, statistical killers (heart disease, car accidents, falls) are so common they barely register as news.
The result: We devote enormous emotional energy and public resources to managing risks that kill relatively few people while underinvesting in preventing the causes of death that are actually most common.
Psychologist Paul Slovic found that perceived risk is driven partly by âdreadâ - a sense of uncontrollability, catastrophic potential, and unfamiliarity. Nuclear energy scores extremely high on dread despite causing relatively few deaths. Driving a car scores low despite killing over 1.3 million people globally per year.
Annual global deaths (approximate):
The math is stark. The risks that dominate media coverage and public anxiety cause a tiny fraction of the deaths caused by the chronic, lifestyle-related, mundane killers that rarely make headlines.
Perceived risk: Nuclear energy is associated with catastrophic disasters (Chernobyl, Fukushima), radiation fear, and uncontrollable technology. It scores very high on the dread scale.
Actual risk data: Per unit of electricity generated:
Nuclear energy is statistically one of the safest energy sources ever developed by humans. Our intuitive fear of it is almost perfectly inverted from what the data shows.
Before COVID-19, pandemic preparedness received far less attention and funding than terrorism - despite the historical record showing that infectious disease has killed orders of magnitude more people than warfare or terrorism throughout history:
Pandemics were chronically underestimated as risks precisely because they are not dramatic, localized events with identifiable perpetrators - even though they are among the most dangerous threats humanity faces.
Smilâs data-driven analysis of risk points to several genuinely serious risks that receive insufficient attention relative to their actual threat level:
Smil is not arguing that we should be indifferent to terrorism, plane crashes, or nuclear accidents. He is arguing that the allocation of public attention, media coverage, policy resources, and individual anxiety should roughly track actual risk rather than the inverse of it.
This requires cultivating what he calls âquantitative literacyâ - the ability to reason about magnitudes, probabilities, and statistics rather than being guided primarily by vivid anecdotes and media narratives.
What risks do you spend the most energy thinking about or trying to avoid? What would your actual mortality statistics say about those choices? If you allocated your risk-reducing behaviors in proportion to your actual risks, what would you do differently?