Process versus outcome

Here is a sentence you have almost certainly said, or thought, or nodded along to: "It went up, so I was right." It feels like the most obvious thing in the world. The stock rose, you owned it, the market handed you a profit — what more proof of a good decision could anyone want?

That sentence is the single most expensive habit a beginner brings to the market, and this module exists to take it apart.

The problem is hidden in a small word: right. Right about what? The decision to buy was made at one moment, using what you knew then. The outcome arrived later, shaped by a thousand things you did not know and could not control — a global cue, an oil price, a stray tweet, a quarter that surprised everyone. Between the decision and the result sits a fog of pure chance. When you say "it went up, so I was right," you are letting the result reach back through that fog and grade a decision it had almost nothing to do with.

Annie Duke, a professional poker player turned decision scientist, gave this error a blunt and useful name: resulting — judging the quality of a decision solely by the quality of its outcome. It is the reason a lucky gamble feels like genius and a sound, unlucky call feels like a blunder. And until you can pull those two things apart — the decision and the result — you cannot honestly learn a single thing from your own record.

Ask yourself, plainly, what part of an investment you can actually steer. Not the price — that is set by everyone else. Not the news, not the macro, not whether the promoter turns out honest, not whether the monsoon is kind to the company's rural sales. All of that is outside your hands and arrives after you have already acted.

What you control is narrow and complete: the decision. What evidence you gathered before you acted. Which alternatives you weighed. How much you put at risk. What you decided in advance would tell you that you were wrong. That bundle — the reasoning and the discipline around a choice — is your process, and it is the only thing that is genuinely yours. It is also the only thing that repeats. You will never buy this exact stock, on this exact day, into this exact world again — but you will use your process on the next decision, and the next, hundreds of times over an investing life.

The outcome, by contrast, is noisy. In the short run a price is mostly other people's feelings, and feelings are close to random over weeks and months. This means a single outcome is a terrible teacher: it is your process plus a large, unpredictable dollop of luck, and you cannot tell, from the result alone, how much of each you are looking at. Grade yourself by outcomes and you are grading yourself partly on a coin you never flipped.

The whole later half of this shelf — the written thesis, the pre-mortem, the position sizing, the journal — is really one instruction repeated: build and protect a good process, because it is the only thing that is yours and the only thing that will still be there tomorrow. This module is where you learn to grade it.

Lay a decision out honestly and it has two independent dimensions, not one. There is the process — was the decision sound, given what could be known at the time? And there is the luck — did the world, afterwards, break your way or against you? Cross them and you get four boxes.

Look down the "grade" column and notice that it tracks the process, not the outcome. Two of the boxes are honest — the good process wins, the bad process loses, and the screen agrees with the truth. But the other two are where every learning error lives. A good process that draws bad luck shows a loss — and if you grade yourself by the screen, you will punish a decision that deserved praise and slowly teach yourself to make worse ones. A bad process that draws good luck shows a win — and if you grade yourself by the screen, you will pin a medal on recklessness and do it again, bigger, until the luck runs out.

That second trap is the deadly one, because it feels wonderful. This is where outcome bias does its quiet damage: a win is a win, the money is real, and nothing about the moment tells you the decision behind it was rotten. The market pays out the reckless and the careful alike, in the short run, with no note attached explaining which was which.

There is a sharper tool for the boxes that involve odds, borrowed from gamblers and worth owning: expected value. It asks not "did this bet win?" but "across all the ways this could have gone, what was it worth on average?" A one-in-four shot that risks ₹2,00,000 to make ₹40,000 is a losing decision even on the rare day it pays — because three times in four it costs you dearly, and averaged out it bleeds money. The single happy outcome hides an ugly average. Grading by outcome sees only the win; grading by expected value sees the decision as it truly was, before the dice stopped rolling.

Watch the two dials come apart. illustrative

Two friends, Meera and Dev, buy the same mid-cap stock on the same morning. Six weeks later both are up 9%. On the screen, their decisions look identical — same stock, same gain, same smiles.

But rewind to the morning of the purchase, before any outcome existed. Meera had read two years of the company's cash-flow statements, noticed receivables finally falling and cash conversion catching up to reported profit, decided the market had not yet priced the improvement, sized the position at a level she could hold through a bad quarter, and wrote one line: I'm wrong if cash conversion slips back below 70% for two quarters. Dev bought because a friend in a group chat was very confident, and because the chart "looked ready." He wrote nothing, risked more than he would admit, and had no idea what would tell him he was wrong.

Same 9%. Two completely different decisions. Meera's is repeatable — she can do that again on the next company, review whether her reasoning held, and improve. Dev's is not — he cannot repeat "a friend was confident" as a method, and the 9% has just taught him, falsely, that it works. This is process over outcome in a single frame: the result graded neither of them; only the reasoning behind it could.

Now take the two dials into your own hands. Set the process — sound or weak — and set the luck — kind or unkind — and read the honest grade the widget returns. Watch the one thing that matters: move the luck dial and the WIN or LOSS on the screen flips, but the grade of the decision does not budge, because you never controlled the luck in the first place.

If a single outcome cannot grade a decision, and your memory rewrites the past the moment a result lands, then honest learning needs something neither noisy nor editable. It needs a witness written down before the outcome exists.

That witness is a decision journal, and it is the quiet hero of this whole shelf. Before you act, you write a short, dated note — not an essay, four or five lines you will actually reuse:

1. The decision, and the evidence for it. What am I doing, and what specific facts — not feelings — support it?
2. The alternatives I weighed and rejected. What else could I have done with this money, and why not?
3. The size, and why. How much am I risking, and can I hold it through a bad quarter without panic?
4. What would prove me wrong. The single condition — a number, an event — that would tell me the thesis has broken.
5. When I'll review it. A date, so the review happens on schedule and not only after a shock.

This note does something no memory can. When the outcome finally arrives — win or loss — you open the record and compare what you knew with what happened. Now you can place the decision in the right box. A loss with a sound, well-evidenced note in front of it was a good decision that got unlucky; repeat the method. A win sitting on top of a note that says "a friend was sure" was a bad decision that got lucky; do not be fooled by the profit. The journal is the only instrument that lets you tell those two apart — because it preserves the one thing hindsight destroys: what you actually believed before you knew the answer.

Grading process over outcome is powerful, and like every powerful idea it can be turned into a comfortable lie. Three cautions keep it honest.

It cannot become a shield for a broken method. The most common abuse of this whole module is the reader who loses again and again and murmurs "just bad luck, good process" each time. One loss is noise; a pattern of losses is data. If the same process keeps meeting the same failure, the outcomes have earned the right to indict the process — and refusing to hear them is not discipline, it is denial wearing the costume of discipline.

It cannot manufacture insight you do not have. A calm, well-journaled process is a defence against your own errors; it is not, by itself, knowledge of a business. You can grade your decision honestly and still have decided on thin evidence about a company you did not understand. Good process and good analysis are two different skills — this module builds the first, and the rest of the syllabus builds the second.

And it cannot judge a decision on information that only exists afterwards. This is the whole point, but it cuts both ways: just as a loss does not condemn a decision that was sound at the time, a near-miss you avoided by luck does not mean your process caught it. Grade every decision only on what was knowable before it resolved — never on the answer sheet the outcome later handed you.

The same few moves catch beginner after beginner. Named once, they are far easier to catch in yourself.

1. Reading a win as a verdict on the method. A profit means the outcome was good; it does not mean the decision was. The reckless and the careful both get paid in the short run — the screen never tells you which one you were.

2. Abandoning a sound process after one unlucky loss. If you drop every method the first time it loses, you will churn forever and keep only whatever most recently got lucky — which is the worst possible thing to select for.

3. Letting hindsight grade the decision. Once you know the result, your memory quietly rewrites what you believed beforehand. Without a note written in advance, every "I always knew" and every "I should have seen it" is fiction, and you learn a lesson that never happened.

4. Grading by the payoff instead of the odds. A bet that won can still have been a terrible decision if it risked far too much for far too little on a long shot. Expected value, not the single result, is the honest measure of a bet.

5. Turning "process over outcome" into an excuse. The principle grades the decision first — it does not tell you to ignore a run of results. When "just bad luck" is your answer every single time, the outcomes are trying to tell you something you have decided not to hear.