Separating a good decision from a good outcome, and the structures that help under pressure.
A compelling story is a claim to test, and the test for a technology is watching it work.
Venture and growth diligence, and any investment where the case rests on a founder's account of something that cannot yet be seen.
From The Inventor: Out for Blood in Silicon Valley · 2019
The market misprices what is hard to measure, so a consistent cheap signal can beat an expensive expert.
Quant interviews, any debate about discretionary against systematic investing, and understanding how you yourself are being assessed in recruiting.
From Moneyball · 2011
Judge a decision by what was knowable when it was made, and keep the outcome in a separate column.
Reviewing a trade or a deal that went wrong, answering a question about a failure, and any post-mortem.
From Thinking in Bets · Annie Duke
Clarify the metric, protect the guardrails, then decompose the drop by segment until it is isolated; show the process, not a guess.
Product management interviews, any analytics role, and any desk conversation that starts with a number going the wrong way.
From Google Product Manager Metrics Interview: A Worked Mock · Aakash Gupta
Separate a bad year into your errors and the market's mood, name the errors specifically, and say whether the method changes.
Reviewing your own record, reading any manager's bad year, and the interview question about a mistake.
From Giverny Capital's 2025 annual letter, on a year of underperformance · François Rochon
Before estimating, ask what number is anchoring you and what the base rate is for cases like this.
Valuation, deal pricing, forecasting, and the behavioural finance questions in hedge fund and asset management interviews.
From Thinking, Fast and Slow · Daniel Kahneman
Evidence about demand comes from what people have already done and paid for, not from what they say they would do.
Customer discovery, commercial due diligence, and any interview that asks how you would validate a product idea.
From The Mom Test · Rob Fitzpatrick
Check whether the metric is still the goal or has become the goal, and reserve the heavy process for decisions that cannot be undone.
Product and leadership questions, assessing a large company's culture, and deciding how much process a decision deserves.
From Amazon's 2016 letter to shareholders (Day 1) · Jeff Bezos
In a field where the rules keep changing, a late specialist with range usually outlasts an early one without it.
Explaining a non-standard background, choosing between a narrow and a broad early role, or answering why you changed direction.
From Range · David Epstein
Ask what would make this fail and who is paid for what before asking anything else.
Diligence, deal structuring, reading any management proposal, and preparing for any decision you cannot easily reverse.
From Poor Charlie's Almanack · Charles T. Munger
The fair test of a decision is the information available when it was made; hindsight is a different exam.
Discussing the Lehman decision, or any post-mortem where later knowledge is contaminating the judgement.
From Inside the Meltdown · 2009
Look for situations where the range of outcomes is wide but the chance of permanent loss is small, because the market prices both as risk.
Idea generation, special situations, and any pitch where the obvious objection is that the outcome is unknowable.
From The Dhandho Investor · Mohnish Pabrai
Ask how many people ran the same method and are not in the sample before treating a record as evidence of skill.
Reading any success memoir, evaluating a manager's record, or presenting your own.
From How I Made $2,000,000 in the Stock Market · Nicolas Darvas
Write the thesis with its test attached: the variant view, the evidence, and the event that would confirm or refute it.
Writing a stock pitch, reviewing a portfolio, and any investment committee that wants to know what would change your mind.
From Greenhaven Road Capital's Q2 2026 letter, on changing how the portfolio is built · Scott Miller
Find the subproblem and the recurrence first; the implementation is mechanical once the structure is clear.
Coding interviews, and any problem whose difficulty is in the decomposition.
From 5 Simple Steps for Solving Dynamic Programming Problems · Reducible
In a post-mortem, separate the direction from the pace and the explanation, and say which one failed.
Subscription pricing, crisis communication, and the behavioural question about a mistake you made.
From Netflix's third-quarter 2011 letter to shareholders · Reed Hastings and David Wells
In a disagreement, weight each view by that person's track record on the question, not by their rank or confidence.
Investment committees, group cases, and any time you are deciding whose forecast to trust.
From Principles · Ray Dalio
Decide which mistake is more expensive before tuning a model; the threshold is a business choice the statistics only express.
Any model used for screening, fraud, credit or hiring, and the interview question of how you would evaluate a classifier.
From Precision vs Recall, With a Clear Example · Kimberly Fessel