Part 5 · Building defences · Chapter 19
The investing journal
A journal is the one honest witness to what you knew before the outcome — a learning instrument, not a place to decorate hindsight.
15 min
Prerequisites not yet complete
This module builds on Chapter 5: Confirmation bias, Chapter 8: Hindsight, Chapter 11: Process versus outcome, Chapter 16: The written thesis. You can read on, but the sequence is load-bearing.
The witness that cannot be bribed
By now you have a written thesis — the dated claim, with the line that would prove it wrong and the size you can carry calmly. This module is about the record that outlives it: what you do with that thesis after the price starts moving and the outcome starts arriving. Because a thesis tells you what to do; only a journal tells you, honestly, what you actually did and why — and what it taught you.
Start with a scene you have almost certainly lived. A stock you bought falls, and a few weeks later you explain it to a friend: "To be honest, I always had a bad feeling about it — the warning signs were obvious." It feels completely true as you say it. But go back and ask what you truly believed on the day you bought: excited, hopeful, confident enough to spend real money. The bad feeling did not exist then. Your mind manufactured it after the fall and back-dated it, so that the past would agree with the present. This is , and it is not lying — it is the ordinary, automatic way memory works.
Here is why that matters more than it seems. If your memory of your own past decisions is edited by their outcomes, then you cannot learn from experience by remembering. Every review you run from memory is partly fiction — the winners feel like foresight, the losers feel like things you "always knew" — and so the same mistakes come back wearing new clothes. The one defence is a witness that was in the room before the outcome and cannot be bribed by it afterwards: a dated note, written in your own hand, of what you knew, what you felt, and what you decided. That witness is the , and building it is the whole subject of this module.
Why a journal beats a memory
The journal exists for one reason: memory is an unreliable witness in exactly the moments that matter most. It is not a filing cabinet you retrieve from; it is a story you re-tell, and each re-telling is quietly rewritten to fit what you now know. After a win, memory inflates how sure you were and how clearly you saw it coming. After a loss, it invents warning signs you never noticed and confidence you never had. The result is that unaided experience teaches you very little, and sometimes teaches you the opposite of the truth — that a reckless bet which happened to win was "a good call", or that a sound decision which happened to lose was "a mistake to never repeat".
A journal breaks that loop by moving the record outside your head, onto a page, timestamped, before the outcome exists. Ink written on the day of the decision does not get more confident when the stock rises or more fearful when it falls. It preserves the one thing the outcome most wants to erase: your real, unedited uncertainty at the moment you acted. Months later you can set the memory beside the note and see the gap — and it is almost always a shock how much less you knew, and how much less sure you were, than you now remember being.
This is also why the journal is the natural partner of the written thesis from the last module. The thesis is what you commit to before buying; the journal is where that thesis lives, is dated, and is later reviewed against reality. Together they turn investing from a stream of impulses you half-remember into a series of decisions you can actually audit. And auditing decisions — not results — is the skill everything on this shelf has been building toward:
One reassurance about effort, because it decides whether you will actually keep this up. A journal that takes twenty minutes an entry will be abandoned by the third week. The whole design below is built to be short — a handful of lines you can write before an order and reread in a minute — because a defence you will not use is not a defence.
What one entry contains
An entry has seven short fields. Each one is chosen to capture something the outcome will later try to erase, so that your future self reviews the truth instead of a flattering edit of it.
-
The date. Not decoration — the entire point. A journal is only a defence against hindsight because it is timestamped before the result. The date is the proof that you knew this much, and no more, on this day.
-
The decision and its size. What you did, precisely: buy or sell, how many shares, at what price, and — the part beginners skip — how much of the portfolio it is. Size is where fear and greed do their real damage, so it must be on the record.
-
The evidence for it. The facts you were leaning on, in your own words. Cash tracking profit, a widening distribution, a clean balance sheet. If every line is an adjective — "great", "strong" — you have recorded a mood, and the review will have nothing solid to test.
-
The alternatives you rejected. What else you could have done with the money, and why you passed. This single field quietly defends against the illusion that your choice was obvious; it records that there was a real fork, and which way you turned.
-
Your emotion and confidence, honestly. A number out of ten, and a plain word for how you feel — "confidence 6/10, felt rushed", "excited after a friend's tip". This is the field hindsight most wants to delete, and the one that later reveals your patterns. Write it even when — especially when — it is unflattering.
-
What would prove you wrong — the invalidation. The single checkable fact at which the decision is dead and you act. This is your , borrowed straight from the written thesis: not "if it stops feeling right", but a number a stranger could look up, like receivable days above 90 for two quarters.
-
The review date. The day you will next open this on purpose. Naming it in advance is what stops the journal from becoming a price ticker you check hourly; it sets the instead of letting the price set it.
Notice what is not on the list. There is no field for the running price, no field for the gain or loss, no field for what the stock did after you bought. Those are the outcome, and the outcome is exactly what a journal must be protected from — the moment you start recording the price, you start grading the decision by its result, which is the reflex this whole entry is designed to defeat.
A filled entry, and the same entry reviewed
Fields stay abstract until you see them filled. Here is one complete entry for a made-up company — every figure invented to teach — written on the day of the decision, before a single rupee moved.
| Field | What is written — 12 Aug 2026 |
|---|---|
| 1 · Date | 12 Aug 2026, written before placing the order. |
| 2 · Decision + size | Buy a branded packaged-foods maker. 5% of the portfolio (₹40,000) at ~₹520/share. |
| 3 · Evidence | Cash profit has tracked reported profit for 5 straight years; net debt near zero; distribution widening. Entry at ~28× vs a 5-year median of ~34×, after a soft festive quarter I think is over-read. |
| 4 · Alternatives rejected | A cheaper commodity-foods peer (no pricing power — passed); staying in cash (fine, but I have real conviction here). |
| 5 · Emotion + confidence | Mild FOMO — a friend flagged it yesterday. Confidence 6/10. Honest note: I feel slightly rushed. |
| 6 · What would prove me wrong | Receivable days above 90 for two consecutive quarters, OR cash profit lagging reported profit for two years. Either kills the cash-quality case — exit, do not average down. |
| 7 · Review date | After Q2 results, ~5 Nov 2026 — and any day the invalidation trips. |
Now let three months pass, and let the news be bad — the way it usually is when a journal earns its keep. The soft quarter continued, the price is down 14%, and the gut says sell. This is the moment memory would step in and start editing: "I always thought that festive weakness was a real problem." But you do not have to consult memory, because you have the note. Here is the same entry, reopened on its review date.
| On review | What is written — 5 Nov 2026 |
|---|---|
| The result so far | Down 14% since I bought (about −₹5,600). The soft quarter continued. My gut wants to sell. |
| Did the evidence hold? | Receivable days are 64 — well under the 90 I named. Cash profit still tracks reported profit. My invalidation has NOT tripped. |
| Grade the decision | The case is intact; the price fell, the business did not. Process sound → hold. The 14% is noise against a thesis still standing. |
| What the old entry taught me | On 12 Aug I wrote 'confidence 6/10, felt rushed'. Good that I sized this at 5%, not 15% — the rushed feeling was real, and the small size is why this fall is survivable. |
| Pattern across recent entries | Third time I've bought within a day of a friend's tip. New rule: 48-hour cooling-off before acting on anyone else's idea. |
Read the fourth row of the review again, because it is the quiet magic of the whole practice. You could only write "good that I sized this small, because I noted I felt rushed" because the rushed feeling was on the record from 12 August. A memory would have deleted it — you would now recall a calm, considered purchase, and learn nothing. The journal let your past self hand a real lesson to your present self, one that no amount of remembering could have reconstructed.
And the fifth row is where the journal stops being about one decision and starts being about you. One entry teaches you about one company. Ten entries, read together, teach you about your own — the mistakes you make again and again that are invisible inside any single decision. Do you always oversize when excited? Always sell winners early and cling to losers? Always buy within a day of a tip? No single entry reveals it; the stack does. That is the deepest thing a journal does, and nothing else can.
Read it live
Watch the two sides of one entry directly. The card below shows the filled decision-day note, then flips to the same entry on its review date — the moment the practice actually pays. Try grading the decision by its result, then by its process, and notice how differently the identical −14% reads. illustrative
This is the whole entry — seven short lines, written while calm, before a single rupee moved. The two shaded fields are the ones hindsight most wants to erase: how much you risked, and how sure you honestly were. Now jump to the review date and watch what the entry lets you do that memory never could.
Illustrative. A made-up entry for a made-up company — a teaching sketch, not a claim about any real business. Nothing here is investment advice.
The point the card makes on its own is the one the audit asked for: the entry is not a list of fields, it is a witness. Because the decision-day note exists — with its honest "confidence 6/10, felt rushed" — the review is a comparison, not a memory. You can see the uncertainty you really had, grade the decision the evidence still supports, and refuse to let a red price rewrite a sound call into a regret.
Using it: results day, exits, and patterns
Writing the entry is half the practice; using it is the other half, and it has three moments.
The first is the review date. When the day you named arrives — usually tied to results — you deliberately open the entry and ask one honest question: has the evidence in fields 3 and 6 improved, weakened, or stayed unresolved? Not "how do I feel about it now", but "what has actually changed against what I wrote". If the evidence has genuinely weakened but the invalidation has not tripped, you note it and watch more closely. If the invalidation has tripped, you act — even, especially, when you do not want to. And you resist the temptation to reopen the entry on any day the price is merely red; a note you reread every time it hurts is being renegotiated, not consulted.
The second is every exit. When you close a position — win or lose — you write a short review before the memory sets: what did the original evidence do, was the decision sound regardless of the result, and what would you do the same or differently. This is where the pays its largest dividend, because exits are when hindsight is strongest and honesty is hardest. Grade the decision, then note the result separately; never let the result decide the grade.
The third, and the most powerful, is the periodic pattern read — every few months, sit with the last stack of entries and look not at any one decision but at the shape of all of them together. This is the only place your personal weaknesses become visible, because they are invisible one at a time. The reader who oversizes when excited never sees it in a single entry; across ten, the correlation between "felt very sure" and "lost money" is impossible to miss. Then you turn the pattern into a rule — a size cap, a cooling-off period — which is the subject of the next and final module in this part.
What a journal cannot do
A journal is a powerful instrument, and it is honest to be clear about its limits.
It does not make your decisions correct. A perfectly kept journal full of well-reasoned entries can still be a record of a reader who keeps misjudging businesses. The journal disciplines your learning; it does not supply insight. What it guarantees is not that you were right, but that when you were wrong you will find out truthfully, and see the pattern in time to change it.
It does not work if the entries are dishonest. An emotion field that always says "calm and rational", an invalidation set so loose it could never trip, a review that quietly moves the goalpost when the fact you named is crossed — each turns the journal into a comfort object that flatters you instead of teaching you. The value is entirely in the honesty of the writing, and honesty here often means recording things you would rather not admit.
And a journal can become a place to over-monitor, if you let the review cadence collapse into a daily habit. Opening every entry every time the market moves reintroduces exactly the fast, feeling-driven churn the journal was built to slow. Written for you to consult on a schedule and at your triggers, it becomes harmful the moment it turns into another screen you check compulsively.
Where people get fooled
The same few slips turn a real journal back into a diary. Named once, they are easy to catch on the page.
-
Writing the entry after the outcome. A note written once the stock has moved records your edited memory, not your original reasoning — it captures the hindsight it was meant to defend against. The entry has to exist before the result, or it defends nothing.
-
Grading the decision by its result. A loss is not proof the decision was bad, and a win is not proof it was good; both are one noisy sample. A journal that celebrates every winner and condemns every loser is training you to fear sound decisions and repeat lucky ones.
-
Leaving out the emotion field because it feels unimportant. "Confidence 6/10, felt rushed" is the single most valuable line in the entry, because it is the one memory deletes and the one that reveals your patterns. Skip it and you can never learn that you keep buying while rushed.
-
A soft invalidation. "Exit if the story changes" cannot be checked and quietly follows the price. Without a fact a stranger could verify, the review has nothing to hold the decision against.
-
Never reading the stack. Individual entries teach you about individual companies; only reading many together reveals the base rate of you. A journal you write in but never review as a whole is a cost with half its benefit thrown away.
Decide
Test your reading, not your memory — short decisions under incomplete information. The answer only shows after you commit.
All figures are illustrative — constructed to demonstrate a judgement, not reported as fact.
Carry forward
- Memory is edited by outcomes — after the fact you "always knew" — so you cannot learn from experience by remembering. A dated journal, written before the outcome, is the one honest witness that hindsight cannot bribe.
- One entry has seven short fields — date, decision and size, evidence, alternatives rejected, honest emotion and confidence, the checkable invalidation, and the review date — and it deliberately records no price, because the outcome is what a journal must be protected from.
- Use it at three moments: on the review date, at every exit, and — most powerfully — in a periodic read of the whole stack, which is the only place the base rate of you (always oversize when excited? always sell winners early?) becomes visible.
- Always grade the decision by the reasoning you wrote, not the result by how it feels — a good decision often loses and a bad one often wins over a single noisy sample.
Enables: 020 Rules for calm, used in panic
Write the entry before the outcome, in the same fields every time, and grade the decision — not the result. The memory will flatter you; the dated page will not.
The thinkers this chapter leans on.