The final product of investing is not a stock; it is a chain of decisions. Yet the quality of a decision is forever hidden behind its outcome—win and you assume you were brilliant, lose and you disown it all. This week we shift attention from "what to pick" to "how to decide": lock in your judgment with a decision journal, plug known holes with a checklist, surface the reasons for failure early with a pre-mortem, and finally learn to separate luck from skill. These four tools have nothing to do with stock-picking, yet they decide whether you can truly learn from experience.
The Principle
The reasons for a decision must be written down before the outcome is known. Memory rewrites itself after the fact—once you know how it ended, you convince yourself you "saw it coming." A decision journal is your only honest control group.
Source · Quote
"A general limitation of the human mind is its imperfect ability to reconstruct past states of knowledge, or beliefs that have changed. Once you adopt a new view of the world, you immediately lose much of your ability to recall what you used to believe."
— Daniel Kahneman, Thinking, Fast and Slow (2011)
A universal flaw of the mind: it can barely reconstruct past states of knowledge or beliefs that have since changed. Adopt a new worldview, and you instantly lose the ability to recall what you believed before.
Interpretation
This is hindsight bias (the "I-knew-it-all-along" effect). It blocks genuine learning: if "I already knew," there is nothing to revise. The decision-journal practice—before any major buy or sell, write down four things: ① your core thesis (why buy); ② the scenarios you expect and their rough probabilities; ③ where you might be wrong; ④ what signal would prove you wrong. When you review later, you compare against a black-and-white record, not a memory contaminated by the result. It splits "was the decision right" from "was the outcome good"—the starting point of training judgment.
Case
Ray Dalio, in Principles, makes "recording your decision principles and the reasoning at the time" a core mechanism at Bridgewater: you review against the real record rather than argue from memory about who was right. His formula is "Pain + Reflection = Progress," and reflection, without a contemporaneous record as its anchor, decays into self-justification. For an individual investor, a one-page decision log costs almost nothing yet is one of the few tools that can resist memory's self-flattery.
Limits · Checklist
Limit: a journal only works if you are honest and actually reread it; written and locked in a drawer, it is worthless. Misuse: turning the journal into after-the-fact rationalization, or logging only wins and never losses.
- Before buying, did I write down "what would prove me wrong"?
- Is my thesis falsifiable, or vague enough to be "right" either way?
- Did I attach probabilities to key assumptions, not just "I think it'll go up"?
- When reviewing, do I compare against the record, or a memory rewritten by the result?
Essence · Reflection
Without a record from the time, every review is a conversation with a witness whom the outcome has already tampered with.
Take your proudest win. From memory, write down your reasons for buying—then ask: how many were truly there at purchase, and how many were added after the money came in?
The Principle
Even the sharpest mind drops the obvious under pressure and complexity. A checklist is not for the ignorant; it is for the expert—it hardens what you already know, but easily forget, into steps you cannot skip.
Source · Quote
"Good checklists are precise. They are efficient, to the point, and easy to use even in the most difficult situations. They do not try to spell out everything—a checklist cannot fly a plane. Instead, they provide reminders of only the most critical and important steps."
— Atul Gawande, The Checklist Manifesto (2009)
Good checklists are precise, efficient, and to the point, usable even in the hardest situations. They don't spell out everything—a checklist can't fly a plane; they remind you only of the most critical steps.
Interpretation
Gawande distinguishes two kinds of error: errors of ignorance (you don't know) and errors of ineptitude (you know, but fail to apply it). Most investment losses are the latter—you know to check leverage, ask about the moat, insist on a margin of safety, yet skip it in the excitement. A checklist turns these "know but forget" items into hard constraints. Munger repeatedly urges "a checklist to avoid predictable stupidity." The key: the checklist must be distilled from your own (and others') real failures, not copied from a generic template.
Case
Medicine: the WHO surgical safety checklist Gawande led, piloted at eight hospitals, cut major complications by about 36% and inpatient deaths by about 47% (published in the New England Journal of Medicine, 2009). Aviation is the birthplace of checklist culture—after the 1935 Boeing Model 299 crashed on a test flight and was mocked as "too much airplane for one man to fly," pilots invented the takeoff checklist that made it safe. In investing, Mohnish Pabrai studied his own and others' failures and built a checklist of about seventy items, which he credits with sharply reducing his blow-ups.
Limits · Checklist
Limit: a checklist guards against known errors, not the "unknown unknowns" (black swans); too long and it becomes box-ticking theater. Misuse: mistaking the ticking for due diligence itself, or letting a generic list replace independent thinking.
- Did my checklist grow out of real failures, or was it copied?
- Can each item be answered clearly "yes/no," rather than by vague feel?
- Is there a "red-card" item that triggers an automatic veto (accounting you can't clear, a business you can't understand)?
- When excited or short on time, do I still force myself through the whole list?
Essence · Reflection
The fatal wound in investing is rarely what you didn't know—it's what you knew and skipped at the critical moment.
Recall your worst loss. In hindsight, was it a "gap in knowledge," or a piece of common sense you actually knew but failed to check? Make it the first line of your checklist.
The Principle
Before making a decision, assume it has already failed completely, then work backward: what exactly happened to cause the failure? This is one step earlier than a post-mortem—and far more honest than optimistic forward planning.
Source · Quote
"Imagine that we are a year into the future. We implemented the plan as it now exists. The outcome was a disaster. Please take 5 to 10 minutes to write a brief history of that disaster."
— Gary Klein, via Kahneman, Thinking, Fast and Slow (2011)
Imagine we are a year into the future. We implemented the plan as it stands. The outcome was a disaster. Take five to ten minutes to write a brief history of that disaster.
Interpretation
The pre-mortem counters overconfidence and group consensus. Once you get excited in one direction, dissent gets suppressed. The pre-mortem inverts the question: not "will it fail," but "it has failed"—forcing the mind to search for reasons. This exploits a psychological fact: explaining something that has already happened is far easier than predicting something that hasn't; Klein found that "assuming the outcome is a foregone conclusion" makes people identify markedly more failure reasons. The move is simple: before buying, force yourself to write "three years from now this is down 70%; the three most likely reasons are…"
Case
Before the 2008 crisis, almost no one ran a pre-mortem on "a nationwide simultaneous fall in home prices"—mainstream models treated "national prices never fall together" as an unquestionable premise, and that assumption was the heart of the crisis. Conversely, the handful who shorted subprime early had, in effect, completed a thorough pre-mortem: they first assumed "these AAA bonds will default," then worked backward to the conditions required, and found the conditions already met. The difference wasn't intelligence—it was the willingness to imagine failure first.
Limits · Checklist
Limit: a pre-mortem lists failure paths but not their probabilities—an imagined disaster is not a high-probability one, and overuse leads to gun-shyness and missed opportunity. Misuse: going through the motions, jotting a few harmless risks and moving on. Force yourself to write the one "most likely to be fatal."
- Did I seriously write the most likely scenario for "this goes to zero or halves"?
- Is there a premise in that scenario I'm excitedly ignoring right now?
- Among the failure reasons, which can I monitor or even avoid in advance?
- If the worst scenario hits, does my position size still let me survive?
Essence · Reflection
Forward planning shows you the hope; the pre-mortem shows you the grave—look at both, but far fewer are willing to look at the latter.
Run a five-minute pre-mortem on your largest holding: assume that three years from now it is your biggest loss, and write the three most likely causes. Can you do anything about any of them right now?
The Principle
A good outcome does not prove a good decision, and a bad outcome does not prove a bad one. Conflating the two—what Annie Duke calls "resulting"—is the most common, and most expensive, thinking error investors make.
Source · Quote
"Thinking in bets starts with recognizing that there are exactly two things that determine how our lives turn out: the quality of our decisions and luck. Learning to recognize the difference between the two is what thinking in bets is all about."
— Annie Duke, Thinking in Bets (2018)
Thinking in bets begins by recognizing that exactly two things determine how life turns out: the quality of our decisions, and luck. Learning to tell them apart is the whole point.
Interpretation
Investing is a probabilistic game, and over the short run the variance of luck is enormous—a poor decision can make money by luck, a great decision can lose money by luck. Reason backward from the result to the decision's quality, and you learn the wrong lessons: rewarding flukes, punishing sound judgment. Split decision and outcome into two dimensions and you get the grid below—the truly dangerous cell is the top-right: a bad decision that lands a good outcome, quietly reinforcing your bad habits.
| Good decision | Bad decision |
|---|
| Good outcome | Deservedprocess and luck both right; continue | Dumb luckmost dangerous; reinforces bad habits |
| Bad outcome | Bad luckdon't change the process; luck owed you | Just desertsthe decision itself needs fixing |
Mauboussin offers a test: if you can "lose on purpose," skill dominates; if you cannot lose on purpose (like roulette), it's pure luck. Investing sits between the two and, in the short run, leans heavily toward luck. So the correct yardstick for judging yourself is process, not a single outcome.
Case
LTCM posted net annual returns of roughly 20%–40% from 1994 to 1997, with two Nobel laureates aboard, and was seen as the peak of skill. Then, when Russia defaulted in 1998, its highly leveraged models lost about $4.6 billion in some four months and nearly took down the global financial system. In hindsight, those glory years were laced with "luck not yet detonated"—they had taken on enormous tail risk that simply hadn't paid out. Treating those years' returns as pure skill is textbook resulting.
Limits · Checklist
Limit: separating luck from skill needs a large enough sample—a single trade or year won't do; in long-run, repeated domains, sustained good results genuinely reflect skill, and you shouldn't dismiss all of it as luck and deny ability. Misuse: pleading "bad luck" for every loss, or claiming "that's skill" for every fluke—guard both directions.
- How much of this result came from my judgment, and how much from factors I don't control?
- If this same decision were repeated a hundred times, what would the average be?
- Am I rewarding "lucky wins" and punishing "unlucky-but-correct" decisions?
- Do I judge myself by process quality, or by this one trade's P&L?
Essence · Reflection
Over the short run the market rewards and punishes luck; only over the long run does it reward decisions—don't mistake the former for your grade.
Take one winning trade and one losing trade, and honestly place each in the grid above. Is there any "winning" trade that actually belongs in "bad decision + good outcome"—a fluke?
Going Deeper
Decision journals, checklists, and pre-mortems all take time, and most people can't sustain them—how do you embed them in your process rather than a burst of enthusiasm?
By structure, not willpower. Don't start a separate long essay; compress it into your existing order-placing action: before every order, fill in three lines—thesis, where I might be wrong, what signal cuts the loss—using a template, not a blank page. Add a hard trigger: once a position exceeds a certain share, you must complete the checklist. The point is to make "not doing it" more annoying than doing it. The individual investor's biggest enemy is often not ignorance of the tools, but overestimating their own discipline.
These methods mostly come from poker, medicine, and aviation—transplanted into investing, where is the biggest distortion?
In the length and clarity of the feedback loop. Poker deals hundreds of hands a night, every takeoff and landing gives immediate, clear feedback, and skill iterates fast; investing's feedback is measured in years and heavily contaminated by luck—a bad decision may not surface for five years, a good one may go unrecognized by the market for a long time. So "learning from outcomes" is especially dangerous in investing; you must rely on the decision journal to manufacture honest feedback artificially, or you may spend twenty years learning all the wrong lessons.
AI can write your decision journal, run your checklist, and generate pre-mortem scenarios—does this make you a better decision-maker, or outsource your judgment away?
Both. The upside: AI is good at countering memory tampering and omissions—it can faithfully record your original thesis, automatically compare past judgments against results, and enumerate failure paths you didn't think of. The risk: the core of judgment—assigning probabilities to key assumptions, deciding how much risk to bear, whether to add when others are fearful—remains your responsibility. Outsource that too, and you lose the chance to calibrate your own judgment, and you can't tell when the AI is wrong. The ideal division of labor: AI handles recording, comparison, and divergence; the human does the pricing, the probabilities, and bears the consequences.
Checklists and pre-mortems both guard against "failure"—could they make you over-averse to risk and miss the great opportunities?
Misused, yes. The purpose of these tools is to let you bear risk with clear eyes, not to avoid risk. A checklist's job is not to veto at the first sign of danger, but to make sure that, having seen the risk clearly, you still choose to bear it and that your position size lets you lose and survive. The real discipline is distinguishing "bearable known risk" from "fatal unknown risk": embrace the former with a margin of safety, veto the latter outright.