Personal Dev Uncertainty¶
10 cards — 🟢 3 easy | 🟡 4 medium | 🔴 3 hard
🟢 Easy (3)¶
1. What is calibration in probabilistic thinking and why does it matter?
Show answer
Calibration means your stated confidence matches your actual accuracy. If you say you are 80% sure about 100 predictions, roughly 80 should turn out correct. Most people are overconfident — they say 90% sure when they are right only 70% of the time. Calibration matters because uncalibrated confidence leads to poor decisions: you bet too heavily on things you are wrong about and fail to hedge against risks you underestimate.2. What is a Fermi estimation and why is it a useful thinking skill?
Show answer
A Fermi estimation is a rough order-of-magnitude calculation made by breaking an unknown quantity into smaller estimable parts. Named after physicist Enrico Fermi (famous for "how many piano tuners in Chicago?"). It is useful because: (1) it forces structured thinking about unknowns, (2) errors in sub-estimates often cancel out, (3) it gives you a sanity check before accepting claims, (4) it builds comfort with approximation rather than demanding false precision. Being within a factor of 3 is usually good enough.3. What is ambiguity aversion and how does it lead to worse decisions?
Show answer
Ambiguity aversion is the preference for known risks over unknown risks, even when the unknown risk might be lower. People prefer a known 40% chance of loss over an uncertain chance that might be 20% or 60%. It leads to worse decisions because: (1) it causes people to avoid options with uncertain but potentially better outcomes, (2) it makes the status quo feel safer even when it is not, (3) it penalizes novel options that lack a track record. The antidote is evaluating options on expected value rather than certainty of the probability estimate.🟡 Medium (4)¶
1. How does reference class forecasting improve predictions about uncertain outcomes?
Show answer
Reference class forecasting starts with "outside view" data: find similar past situations (the reference class), look at the distribution of actual outcomes, and use that as your baseline. Then adjust modestly for unique features of your situation. It works because it replaces optimistic inside-view planning with empirical base rates. Example: instead of estimating your kitchen renovation will take 4 weeks based on the plan, check how long similar renovations actually took (median: 8 weeks) and start from there.2. How can you apply confidence intervals in daily life, not just statistics?
Show answer
Express predictions as ranges instead of point estimates: "This project will take 3-5 weeks" instead of "4 weeks." Assign confidence levels: "I am 80% sure it will be done by March, 95% sure by April." This is useful because: (1) it communicates both your best guess and your uncertainty, (2) it prevents false precision, (3) it forces you to think about what could go wrong (the wide end) and what would go right (the narrow end). Tracking your ranges over time improves calibration.3. What is premature closure and why is it dangerous in uncertain environments?
Show answer
Premature closure is settling on an answer or diagnosis too early, before sufficient evidence has been gathered, because the first plausible explanation feels satisfying. It is dangerous because: (1) early hypotheses anchor thinking and bias subsequent evidence gathering, (2) confirming evidence is sought while disconfirming evidence is ignored, (3) in fast-moving situations, early errors compound. It is one of the most common diagnostic errors in medicine and incident response. Defense: explicitly generate at least two alternative explanations before committing to one.4. What are second-order effects and why do most people stop at first-order thinking?
Show answer
First-order effects are the direct, immediate consequences of an action. Second-order effects are the consequences of those consequences. Example: first-order effect of rent control is lower rents; second-order effects include reduced housing construction, deteriorating building maintenance, and black markets. Most people stop at first-order thinking because: (1) it is cognitively cheaper, (2) first-order effects are concrete and visible while second-order effects are abstract and delayed, (3) advocacy focuses on intended effects, not unintended ones.🔴 Hard (3)¶
1. What does calibration training involve and how quickly can it improve your judgment?
Show answer
Calibration training involves: (1) making probabilistic predictions (e.g., "80% confident that X is true"), (2) tracking outcomes, (3) comparing your stated confidence to actual accuracy, (4) adjusting future confidence based on the gap. Research shows significant improvement in as few as 5-10 hours of practice. Common findings: people start massively overconfident, learn to widen their confidence intervals, and develop better intuitions about what 70% vs 90% confidence actually feels like. The key is rapid feedback — you must see your results soon to adjust.2. What is precision theater and how do you recognize it in yourself and others?
Show answer
Precision theater is expressing estimates with false specificity to create an illusion of rigor. Examples: "There is a 63.4% chance of success" from a pure guess, "the project will take 147 hours," or a risk matrix scored to two decimal places from subjective judgment. Recognize it by asking: what data supports this precision? If the answer is "gut feeling" or "experience," the precision is theatrical. False precision is worse than honest uncertainty because it discourages questioning and creates overconfidence. Better: "roughly 60-70%" or "somewhere between 100-200 hours."3. What does it mean to update beliefs when evidence changes, and what makes people resist updating?