Thinking, Fast and Slow Chapter 30

Rare Events

Key Takeaway: Rare events are either completely ignored or massively overweighted — never accurately weighted — depending on whether they attract focal attention: vivid imagery, concrete frequency formats ('1 of 1,000' vs '0.1%'), emotional intensity, and explicit description all trigger overweighting, while diffuse alternatives and experiential learning (choice from experience) produce neglect; denominator neglect means that '1,286 out of 10,000' sounds more dangerous than '24.14%' despite being half the risk.

Chapter 30: Rare Events

← Chapter 29 | Thinking, Fast and Slow - Book Summary | Chapter 31 →

Summary

This chapter explains when rare events are overweighted and when they're ignored — resolving a tension in prospect theory's original formulation. The answer is focal attention: rare events that capture attention are overweighted; rare events that don't are ignored. There is no middle ground of accurate weighting. "When it comes to rare probabilities, our mind is not designed to get things quite right."

Kahneman opens with a personal confession: despite knowing the risks were negligible, he avoided stopping next to buses in Israel during a period of suicide bombings. "I was avoiding buses because I wanted to think of something else." The vivid imagery of explosions, constantly reinforced by media coverage, made the possibility feel present and urgent — a textbook availability cascade from Chapter 13 operating through the possibility effect from Chapter 29. The actual probability was irrelevant; what mattered was whether the threat was salient.

The chapter's most important empirical contribution is #denominatorneglect — our systematic failure to attend to the denominator when risks are expressed as frequencies. "A disease that kills 1,286 out of 10,000" is judged more dangerous than "a disease that kills 24.14% of the population" — even though the latter is twice as deadly. The frequency format ("1 out of 1,000") evokes a vivid image of a specific individual suffering, while the percentage format ("0.1%") remains abstract. Forensic psychologists evaluating Mr. Jones were nearly twice as likely to deny hospital discharge when told "10 of 100 similar patients commit violence" versus "a 10% probability of violence" — identical statistics, dramatically different decisions. Attorneys exploit this: saying "a false DNA match occurs in 1 of 1,000 capital cases" creates the image of a specific wrongful conviction, while "0.1% chance of false match" does not.

The basketball fan experiment by Craig Fox demonstrates how focal attention inflates probability estimates. When fans estimated each of eight NBA teams' chances of winning the playoffs one at a time, the estimates summed to 240% — absurd, but explicable: each team in turn became the focal event, triggering confirmatory imagination of how that team could win. When asked about broader categories (Eastern vs. Western conference), estimates summed to 100%. The lesson: "the probability of a rare event is most likely to be overestimated when the alternative is not fully specified."

The distinction between #choicefromdescription and #choicefromexperience resolves a major empirical puzzle. When people read descriptions of gambles ("5% chance to win $12"), they overweight the rare outcome (possibility effect). But when they experience outcomes through repeated trials (pressing buttons and observing results), they underweight rare events. The explanation: in experience, many people never encounter the rare event in their sample, so it gets weight of zero. Even those who have experienced rare events form "global impressions" of the options (like forming impressions of colleagues) where rare events fade into the background of typical experiences.

For the library, the denominator neglect finding has immediate implications for risk communication and persuasion. Hormozi's case study approach in $100M Offers works partly through vivid individual stories that make success feel concrete and available. Dib's marketing in Lean Marketing benefits from concrete rather than abstract presentation of benefits. Voss's negotiation technique in Never Split the Difference should frame risks in frequency format ("3 of your last 10 deals fell through") rather than probability format ("30% failure rate") when trying to heighten urgency.

Key Insights

Rare Events Are Either Overweighted or Ignored — Never Accurately Weighted — Focal attention determines which: vivid, concrete, emotionally charged rare events are overweighted; diffuse, abstract, unmentioned rare events are ignored. There is no cognitive mechanism for accurate processing of low probabilities. Denominator Neglect Makes Frequency Formats More Impactful Than Probability Formats — "1,286 out of 10,000" sounds worse than "24.14%" (which is twice the risk). "10 of 100 patients" is more alarming than "10% probability." The frequency format creates a vivid image of affected individuals; the percentage remains abstract. Focal Events Get Overestimated; Diffuse Alternatives Get Underestimated — When eight basketball teams are evaluated individually, probabilities sum to 240%. When evaluated as two conferences, they sum to 100%. Success of a specific plan is easy to imagine (focal); failure through myriad unspecified ways is diffuse and underweighted. Choice from Experience Produces Underweighting of Rare Events — In repeated experience (unlike verbal description), rare events are often never encountered and get zero weight. Even when encountered, they fade into the global impression of the option. This explains why Californians don't prepare for earthquakes and why bankers in 2007 didn't prepare for financial crises.

Key Frameworks

Denominator Neglect (Slovic) — The systematic failure to attend to the denominator when risks are expressed as frequencies. "1 of 1,000" evokes a vivid image of the 1; the 999 fade into the background. "0.1%" remains abstract and evokes no image. Consequence: frequency formats produce stronger emotional and behavioral responses than equivalent probability formats. Choice from Description vs. Choice from Experience — Two fundamentally different modes of evaluating uncertain options. Choice from description (reading about probabilities) produces overweighting of rare events. Choice from experience (observing outcomes over time) produces underweighting or neglect. Most real-world decisions are from experience, which means rare risks are systematically neglected until a vivid instance makes them focal. The Focal Event Inflation Principle — When a specific event is made focal (by asking about it, imagining it, or describing it vividly), its probability is overestimated because: (1) confirmatory bias generates scenarios making it true, (2) cognitive ease makes the scenarios feel plausible, and (3) the diffuse alternatives are not similarly elaborated.

Direct Quotes

[!quote]

"When it comes to rare probabilities, our mind is not designed to get things quite right."

[source:: Thinking, Fast and Slow] [author:: Daniel Kahneman] [chapter:: 30] [theme:: rareevents]

[!quote]

"These advocates want to frighten the general public about violence by people with mental disorder, in the hope that this fear will translate into increased funding."

[source:: Thinking, Fast and Slow] [author:: Daniel Kahneman] [chapter:: 30] [theme:: riskformat]

[!quote]

"I was avoiding buses because I wanted to think of something else."

[source:: Thinking, Fast and Slow] [author:: Daniel Kahneman] [chapter:: 30] [theme:: terrorismpsychology]

Action Points

[ ] Use frequency formats when you want to heighten risk awareness; use probability formats when you want to minimize perceived risk: "3 of your last 10 clients churned" is more alarming than "30% churn rate." Choose your format deliberately based on whether you want to increase or decrease the salience of the risk.
[ ] Specify alternatives explicitly to reduce focal event inflation: When evaluating any opportunity, force yourself to list the specific alternatives and assign probabilities to each. If they sum to more than 100%, you're overweighting focal events.
[ ] Distinguish between choice from description and choice from experience: When your assessment of a risk comes from reading about it (description), you're likely overweighting it. When it comes from personal experience without encountering the rare event, you're likely ignoring it. Neither mode is accurate — calibrate accordingly.
[ ] Beware of the disaster cycle: After a vivid event (financial crisis, pandemic, security breach), risk is overweighted and overreaction follows. As time passes without recurrence, the event fades and risk is neglected. Build institutional systems that maintain appropriate risk levels regardless of recent experience.
[ ] Frame risks concretely in presentations and proposals: When you need stakeholders to take a risk seriously, don't use percentages. Say "if we launch 20 products with this approach, we expect 4 to fail catastrophically." The concrete representation makes the risk vivid and harder to ignore.

Questions for Further Exploration

If choice from experience systematically underweights rare events, how should organizations ensure that "black swan" risks remain visible in decision-making despite never having been personally experienced?
Denominator neglect means that the format of risk communication changes behavior. Should regulators mandate standardized risk formats for consumer products, financial disclosures, and medical information?
The basketball fan experiment shows that individual team evaluations sum to 240%. Does the same inflation occur in business portfolio planning — are division-by-division forecasts systematically too optimistic because each is evaluated as a focal event?
If vivid imagery overwhelms probability assessment, can data visualization be designed to counteract this by making probabilities more vivid than outcomes?

Personal Reflections

Space for your own thoughts, connections, disagreements, and applications.

Themes & Connections

Tags in this chapter:

#denominatorneglect — Failure to attend to the denominator; frequency formats more impactful than probability formats
#rareevents — Either overweighted (when focal) or ignored (when diffuse); never accurately weighted
#choicefromexperience / #choicefromdescription — Two modes producing opposite probability weighting
#frequencyformat — Concrete representation ("1 of 1,000") that evokes vivid imagery and increases decision weight
#vividness — Rich imagery overwhelms probability in evaluation of uncertain prospects

Concept candidates:

Denominator Neglect — New concept: the mechanism behind format effects in risk communication
Rare Events — New concept: the conditions for overweighting vs. neglect

Cross-book connections:

$100M Offers Ch 10-11 — Hormozi's case studies leverage focal event inflation: each vivid success story makes the prospect imagine themselves succeeding, overweighting the probability
Never Split the Difference Ch 5-7 — Voss should frame risks in frequency format for maximum impact: "3 of your last 10 deals" not "30% failure rate"
Contagious Ch 1-2 — Berger's emphasis on vivid, emotional content connects to vividness overweighting: emotionally charged content makes the described outcome feel more probable
Influence Ch 6 — Cialdini's scarcity principle works through focal attention on the rare event of missing out