Conviction Is Not How Strongly You Feel About a Stock

At some point I had 80 stocks.

The explanation I gave myself was diversification. The real explanation was simpler and less flattering: I had never pushed any single idea hard enough to earn the right to size it properly. So each new stock joined the pile at 1–2%, which made it feel like a decision without actually being one. Eighty positions meant eighty ideas I was interested in but hadn't genuinely interrogated. Eighty maybes dressed up as a portfolio.

And yet I had opinions on all of them. I could have argued for any of them in a conversation. What I didn't have — for almost any of them — was the specific, falsifiable understanding of a business that makes sizing feel like a natural expression of what you know, rather than a bet on what you feel.

That distinction — between a feeling and a framework — is what this article is about.

Why does feeling convinced about a stock lead to the wrong sizing decision?

There is a version of conviction that every investor recognises. It's the state of mind where a stock feels obviously right — where you've read enough, thought enough, and arrived at a conclusion that feels solid. That feeling is real. It's also completely useless as an operating framework.

The feeling tells you you believe in the stock. The framework tells you what you actually know about it — and what you're prepared to do with that knowledge.

These are different things. And the gap between them is where most self-inflicted portfolio damage lives.

The investor who feels high conviction and sizes at 2–3% has made a quiet confession to themselves. Not in words — words, during research, are easy to make confident. The confession is in the position. The portfolio knows the difference between a thesis you'd stake 15% on and one you'd stake 2.5% on, even when the investor talks about both with identical certainty.

But here's the important part: the undersized position is usually not a risk management choice. It is the subconscious registering that the research was shallow. You read the annual report. You liked the numbers. You formed a positive impression. But you never pushed through to the specific claim — the falsifiable, testable assertion about why this business will compound in this way because of this structural condition. So when it came to sizing, something in you pulled back. Not because you were being prudent. Because you didn't actually know enough to be confident.

This is why conviction needs to be operationalised — not because writing things down magically improves the outcome, but because the act of writing forces a confrontation with what you actually know versus what you merely feel.

What are the four components of a real conviction framework?

The conviction framework has four parts: the thesis, the KPIs, the exit conditions, and the position size. Most investors who build frameworks treat these as documentation requirements — things to write down because a disciplined investor is supposed to write them down.

They are not documentation requirements. Each one is answering a specific, hard question that you would prefer, if you're being honest, not to answer precisely.

To show how all four work together, one example will run through each component: CDSL — the depository business that sits at the infrastructure layer of the equity market.

Component 1 — The one-line thesis

The thesis exists to answer one question: Can I compress the entire reason I'm holding this stock into a single sentence, without notes, right now?

Not because short theses are better than long ones. Because the ability to compress is evidence of real clarity. When you cannot state the thesis in one sentence, what you have is a collection of impressions, facts, and positive associations — not a falsifiable claim that a specific business will compound in a specific way because of a specific structural condition.

A theme is not a thesis. "India's financialization story" is not a thesis for CDSL — it's a backdrop that justifies a dozen different investments, none of which it actually constrains. A backdrop cannot be falsified. A thesis can.

The correct thesis for CDSL starts with understanding how the business actually makes money. CDSL earns revenue primarily from issuers — the companies whose shares are held in demat form — based on the higher of a stable annual fee or a per-folio fee tied to the number of investor folios holding that company's shares. This means the revenue driver isn't demat account additions per se. It's the growth in active folios across CDSL-held issuers, combined with the stability of the issuer base itself.

So a falsifiable thesis might read: CDSL's revenue will compound as active folio counts across its issuer base grow — driven by new retail participation in equity markets — while the two-depository regulatory structure protects its economics from pricing pressure.

That claim has testable components. Every quarter, you can check whether it's holding.

Component 2 — The KPIs

The KPIs answer: What are the two or three numbers — not the price — that tell me every quarter whether the underlying business is delivering what the thesis requires?

But there is a deeper function that KPIs serve, one that most investors miss: KPIs are also a test of whether the thesis itself is falsifiable.

If you write a thesis and then cannot identify KPIs that connect directly to the core assumption, the thesis isn't specific enough. The inability to name KPIs is diagnostic of a vague thesis, not just a monitoring gap. You haven't failed to find the right metrics — you've revealed that the underlying claim doesn't have enough precision to be measured.

Go back to the CDSL thesis. The load-bearing assumption is active folio growth across the issuer base and the stability of the two-depository structure. The KPIs follow directly:

• Active folio count per issuer — growing or stagnating?
• Total issuer count on CDSL — any attrition or share shift to NSDL?
• Annual issuer fee realisations — are per-folio economics holding?
• Any regulatory signal on the two-depository exclusivity — the structural moat the thesis depends on?

If someone had been using "demat account additions" as the primary KPI for CDSL — which is a common mistake given how prominently that number is reported — they would have been measuring the wrong thing. Demat additions are a leading indicator of potential folio growth, not the revenue driver itself. A thesis built on demat additions would have the wrong exit condition, the wrong monitoring cadence, and the wrong framework for interpreting quarterly results.

This is why the KPI question forces a confrontation with thesis quality. If you find yourself writing KPIs that feel adjacent to the thesis rather than derived from it, go back to the thesis first.

KPIs serve one additional function: they keep you from managing by price. A stock falling 20% while KPIs are intact is a market judgment about sentiment, not a business judgment about trajectory. Knowing the difference — viscerally, in the moment, not just intellectually — requires having named the KPIs before the fall began.

Component 3 — The exit conditions

Exit conditions answer: What specific observable event would tell me the core assumption has failed — so that when the moment arrives, I'm not making a fresh judgment call under emotional pressure?

Notice that the CDSL exit condition follows naturally from the thesis and the KPIs. Because the thesis is specific, the exit condition is specific. Because the KPIs connect directly to the load-bearing assumption, the trigger for exiting is already implicit in them:

• If active folio counts stagnate across the issuer base for three consecutive quarters despite stable retail participation — the folio growth assumption is not playing out.
• If a regulatory change introduces a third depository or removes the compulsory two-depository structure — the moat assumption is broken.
• If issuer fee realisations decline materially due to regulatory fee caps — the economics assumption needs revision.

An investor who had built the thesis on "demat account growth" would write different, weaker exit conditions — conditions that don't actually test the business model, because they were never derived from it.

This is the feedback loop: a precise thesis generates precise KPIs, which generate precise exit conditions. A vague thesis generates vague KPIs, which generate no real exit conditions at all — just a general sense of things going well or badly, evaluated fresh each quarter under whatever emotional conditions happen to be present.

The Exit Conditions Framework covers the full mechanics of writing and using exit conditions. This article concerns itself with how conviction generates exit conditions — why an investor with genuine thesis-clarity finds them easier to write, because the load-bearing assumption is already named.

Exit conditions are the most important component and the most often skipped. Partly because writing them requires imagining scenarios you'd rather not inhabit. Partly because they make the stock feel more contingent — less like a confident position and more like a bet with a defined failure mode.

That discomfort is the point.

Component 4 — Position size

Position size is the last component, and often the most revealing one. It answers: Given everything I've written above — how much of my capital am I actually prepared to commit to this thesis?

But it has to be read carefully, because size alone is not a clean signal. More on that below.

What are the two honest tests of real conviction under pressure?

Test 1 — How you sized it

A 2% position and a 15% position are not just different risk allocations. They are often different confessions about conviction. Most investors don't treat them this way. They treat position size as a risk management decision — I'll keep this small to limit downside — rather than what it usually is: the subconscious registering how much the research actually justified.

The investor who has done the work — written the specific thesis, derived the KPIs from it, specified the exit condition — and still sizes at 3% is saying something specific: I know this well, but I'm not willing to stake much on it. That might be rational. Maybe it's a sector they don't fully understand yet. Maybe they're still building conviction through early tranches.

The investor who sizes at 3% without having done the work is making the same statement, but unconsciously. They feel conviction, but the position size has registered what the thesis hasn't made explicit: I'm not sure I've gone deep enough.

One important caveat: position size as a conviction signal requires that the investor has a deliberate, consistent sizing framework to begin with. A newer investor who concentrates heavily in a tipped stock hasn't demonstrated conviction — they've demonstrated inexperience. The diagnostic value of size is highest for investors who have a framework and deviate from it in ways that reveal their actual confidence level.

Test 2 — How you behave under pressure

Sizing at entry is a static signal. The more dynamic test is what happens when the stock moves sharply in either direction.

In a sharp rally: Does the idea of booking profits feel like relief? Do you find yourself calculating how much you've made? Genuine conviction doesn't feel this way in a rally. It raises a different question: has the price run past what the thesis can justify, or is the business continuing to deliver? One question is about the position's profit. The other is about the thesis.

In a sharp drawdown: How quickly does the urge to sell arrive? How does it feel — like fear, or like a genuine re-examination of the thesis? An investor with real thesis-clarity experiences a drawdown as a problem to diagnose, not a sensation to escape from. The first question is whether any KPI has changed, not whether the pain has become intolerable.

Neither reaction alone is conclusive. The behaviour is not the measure — the reasoning behind the behaviour is. But the felt sense that you want to act without being able to name a thesis-based reason for it is one of the most reliable indicators that the conviction you thought you had was more feeling than framework.

Together — sizing at entry and the texture of your reactions under pressure — these are the two most honest read-outs of what you actually believe, as opposed to what you think you believe.

How do you grade conviction before committing capital?

Before adding to or initiating any position, answer three questions directly:

Question 1: Can you write the one-line thesis right now, without looking at any notes — and would you be willing to have that sentence read back to you in six months if the stock has fallen 30%?

Question 2: Can you name the two or three KPIs that connect directly to the load-bearing assumption in that thesis — not adjacent metrics, not industry indicators, but the specific numbers that would tell you every quarter whether the core claim is holding? If you can't name them, the thesis isn't specific enough yet.

Question 3: Have you written a specific exit condition that is not a price level — an observable business event that would tell you the core assumption has demonstrably failed?

All three answered clearly: size proportionally. The work has been done. The position deserves real capital.

Any one unclear: smaller position, more quarters of monitoring. Not because the stock is worse — because you have less to act on, and the position size should reflect that honestly.

How does the conviction framework hold up through quarterly reviews?

A conviction framework written once and revisited never is not a framework. It's a snapshot.

Every quarterly result is a check against the thesis: are the KPIs moving in the direction the thesis requires? Did management say anything on the concall that shifts the load-bearing assumption? Has anything changed in the competitive landscape that the original thesis didn't account for?

The discipline trap is the update that rationalises rather than revises. A KPI moves in the wrong direction, and the investor explains it away — project timing, weather, a one-off charge. Sometimes that's legitimate. Often it's the beginning of thesis drift — the gradual rewrite of the story without acknowledging that a rewrite is happening.

The signs of thesis drift in real time are covered in the companion piece to this article. The mechanism to prevent it starts here: every quarterly update either strengthens the conviction, weakens it, or leaves it unchanged. Over time, the trajectory of that assessment tells you more than any single reading does.

Why I built this into Stockport

The conviction framework is not a personality type. It is not a property of patient or intelligent investors. It is a system — a set of constraints you build before the emotional pressure of ownership distorts your judgment.

The problem with keeping it in your head is that ownership changes how you think. Once you hold a stock, loss aversion, confirmation bias, and social commitment all begin operating on your reasoning. They are not obvious when they're active. They feel like analysis. The conviction framework's job is to preserve the pre-ownership version of your thinking — the version that had no stake in the outcome — so that the post-ownership version can be checked against it.

The 80-stock problem doesn't start with buying 80 stocks. It starts with the first position you never pushed hard enough to either confirm or exit — the first maybe that stayed in the portfolio because exiting felt like giving up, and adding to it felt like admitting you'd been lazy with the research.

Conviction is not how strongly you feel about a stock. It is how precisely you can describe what you know, what you're watching, and what would change your mind. And position size — read carefully, in context — is one of the most telling signals of that conviction. Not because size proves conviction exists, but because when conviction is real and the framework is built, sizing tends to follow naturally.