Position Sizing and Risk Management: How Much to Invest Per Stock

Direct Equity — Arc 3, Article One

The Arithmetic

Why a 50% loss is not half a problem

Gains and losses look like mirror images. They are not, and the gap between them is the entire reason position sizing exists.

Lose 10% of your money and you need 11.1% to get back to level. Annoying, not serious. Lose 30% and you need 42.9%. Lose 50% and you need 100% — the money that remains has to double simply to return you to where you began, having earned nothing. Lose 70% and you need 233%. The formula is one line: the gain required equals the loss divided by what is left. As the loss grows, the denominator shrinks, and the required gain runs away from you.

Now apply that arithmetic to two investors who own exactly the same shares. Both researched a company called Bharat Widgets Ltd carefully. The first puts 40% of the portfolio into it; the second puts 8%. The company then falls 60% — a fraud comes to light, a customer walks away, the reason hardly matters. The first investor’s portfolio is down 24% and needs 31.6% to recover. The second is down 4.8% and needs 5%. Identical research, identical judgement, identical mistake. The only difference was the size of the cheque, and it decided whether the year was a scratch or a wound.

There is a second cost that never shows up in the arithmetic. A position large enough to hurt makes you a worse decision-maker while it is hurting. You check the price at work. You look for reassuring news and skip the uncomfortable kind. You sell at the bottom because you cannot take another week of it, or you refuse to sell at all because admitting it has become too expensive. Position size is not only how much you can lose — it is whether you can still think clearly while losing it.

The Two Numbers

Position size and risk are not the same number

Most people asking how much to invest per stock are really asking one question when there are two. The amount you put in is your position size. The amount you can actually lose is your risk, and it is smaller — because you will not hold a collapsing position to zero. Risk is position size multiplied by how far you are willing to let it fall before you act.

Written as a line of arithmetic, with the numbers in rupees rather than percentages, it becomes usable:

Rupees at risk = shares held × (entry price − exit price). Turn it around and it tells you the only thing you need: shares to buy = rupees you are willing to lose ÷ (entry price − exit price). You decide the loss first. The position size falls out of it. That reversal — deciding the maximum loss before deciding the amount to invest — is the whole discipline in one sentence.

How much should the maximum loss be? The convention borrowed from professional trading is the 1% rule: risk no more than 1% of your equity capital on any single idea, sometimes stretched to 2% by people running fewer, larger positions. The logic is survival, not timidity. At 1%, ten losing ideas in a row cost you roughly 10% — bruising, entirely recoverable, and you are still trading. At 10% risk per idea, the same ten losses take about 65% of your capital and you need a 186% gain to get level. Same ten decisions. Different lifespan.

Take a portfolio of ₹10,00,000 and a 1% rule, so ₹10,000 is the most you will lose on one idea. Bharat Widgets Ltd trades at ₹500. Where you place the exit changes everything:

Exit set below entry	Loss per share	Shares you can buy	Position value	Share of portfolio
5% (₹475)	₹25	400	₹2,00,000	20.0%
10% (₹450)	₹50	200	₹1,00,000	10.0%
15% (₹425)	₹75	133	₹66,500	6.7%
25% (₹375)	₹125	80	₹40,000	4.0%
40% (₹300)	₹200	50	₹25,000	2.5%

Two things in that table surprise people. First, the risk is identical in every row — ₹10,000 — even though the position sizes differ eightfold. Second, and less intuitively, a tighter exit produces a larger position. A stop close to your entry price is not the cautious choice; it makes you bigger and it sits inside the range of ordinary daily wobble, so it gets triggered by noise rather than by anything meaningful. Meanwhile the top row exposes the limit of the formula: 20% of a portfolio in one company is reckless whatever the arithmetic says. The formula gives you one ceiling. The caps in the next section give you another. You always take the lower of the two. Our position size calculator does this arithmetic if you would rather not.

The Caps

The three-way cap that stops one idea sinking you

A single ceiling on each holding is not enough, because holdings fail in groups. Three caps, applied together, cover the ways a portfolio actually gets hurt. The percentages below are illustrative — the structure matters more than the exact figure.

One: the per-position cap. No single company gets more than a set share of your equity money at the time you buy. An investor holding twelve to fifteen shares might cap each at 5–8%; a deliberately concentrated investor holding six or eight might run 10–15% and accept the swings that come with it. Apply the cap at cost, not at market value: if a holding grows past the ceiling because it did well, that is a pleasant and completely different problem, and the portfolio article deals with it.

Two: the per-sector cap. Something like 20–25% of equity money in any one sector. Sector shocks arrive all at once and do not care how well you picked within the sector — a regulatory change, a jump in one input cost, a rate cycle turning. Four excellent lenders bought at four sensible prices are still four lenders when credit costs rise.

Three: the per-thesis cap — the one almost everyone misses. You can hold seven companies across five sectors and still own one idea. A lender, a housing finance company, a cement maker, a paints company and a building materials distributor look like five sectors on a pie chart. They are one bet on the property cycle. The test is cheap: write the reason you own each holding in a single sentence. If three sentences say the same thing in different words, that is one position wearing three shirts, and it should be sized as one. This is what correlation means in practice (correlation is simply the tendency of things to move together), and diversification that ignores it is decoration.

These caps are also the honest answer to a question the pillar keeps returning to: they are what makes a direct-equity satellite safe to hold alongside an index core. Not conviction. Not research quality. Arithmetic ceilings, decided when you are calm and applied when you are not.

Stops, Honestly

When a stop-loss is a strategy and when it is self-harm

A stop-loss strategy is not universally good or universally silly. Whether it helps depends entirely on what your reason for owning the share was in the first place.

If your evidence is the price, the stop is your thesis expiring. Someone buying because a price pattern or trend suggests something — the world described in the technical analysis article — has staked their case on the price behaving in a particular way. When it stops behaving that way, the evidence is gone and there is nothing left to hold. Here a stop is not a safety net bolted on afterwards; it is the strategy, and trading without one is trading without an exit condition at all.

If your evidence is the business, a price stop is self-harm. Suppose you bought after reading three annual reports and concluding the business earns well on its capital and is priced sensibly. The share then falls 20% because a foreign fund is liquidating, or because the whole market fell, or for no reason anyone can name. Nothing in your evidence has changed. A stop-loss sells you out of the position precisely when the thing you liked got cheaper. Do this repeatedly and you have invented the worst of both worlds: the transaction costs and taxes of a trader, the research burden of an investor, and the returns of neither.

It is also worth knowing that a stop is a weaker instrument in practice than it looks on a screen. You place it as a stop-loss order with a trigger price — either a limit version, where you also name the worst price you will accept, or a market version, which takes whatever is available once triggered. Neither guarantees the price you had in mind, for three ordinary Indian-market reasons. Shares gap: a company that closed at ₹500 with your trigger at ₹425 can open at ₹380 the next morning after bad news, and ₹380 is where you exit. Price bands bite: if the stock is locked at its lower circuit there are simply no buyers inside the permitted range, and your order waits, unfilled, while the band steps down day after day — the exact trap that makes manipulated small-caps so dangerous. And thin stocks punish size: in a company that trades a few lakh rupees a day, your own sell order pushes the price down, so the price you get trails the price you saw (that gap is usually called slippage). Broker features such as GTT instructions are convenient, but most of them sit on the broker’s system and send an order when triggered rather than resting at the exchange — useful, not magic.

So what does an investor use instead? An exit written in business language rather than price language, and decided before you buy, in the same research file the research process article asks you to keep. Sell if the promoter pledges a large slice of their holding. Sell if margins fall for three consecutive quarters and management cannot explain why. Sell if the auditor resigns, if the single largest customer leaves, if the reason you bought is factually contradicted. Add a time limit for good measure: if the thesis has not begun to show up in the numbers within a stated period, the money moves on regardless of the price. And underneath all of it sits the real protection — a position small enough that being wrong about it never forces an emergency decision.

Sizing For Uncertainty

Wider unknowns, smaller cheques

The caps set your maximum. Very few positions deserve the maximum. A useful working rule: size in inverse proportion to how much you do not know. Five situations where the unknowns are structurally wider, and the cheque should shrink accordingly.

Companies with no public history. A recently listed business has no record of how it behaves under a bad year, and its published figures were assembled by people who were selling. Half-size at most until a few reporting cycles have passed.

Small and thinly traded companies. The genuine advantage of the small end of the market — less institutional competition, more mispricing — comes attached to the genuine disadvantage that you may not be able to sell in size on a bad day. Illiquidity is not a rounding error; it is a reason to be smaller.

Businesses at the edge of what you understand. If you cannot explain how the company makes money to someone who does not work in it, you are not being modest by taking a small position. You are being accurate.

Theses that depend on a forecast. There is a real difference between “this business already earns well and is priced reasonably” and “this business will do well if the commodity price rises, if rates are cut, if the scheme is announced”. The second kind stacks conditions, and every condition is a chance to be wrong. Turnarounds belong here too.

Borrowed money — where sizing stops being yours. This one deserves emphasis. With a margin trading facility, or with derivatives, the exit is no longer your decision: fall far enough and the lender’s margin call sells you out at the worst possible moment, converting a temporary fall into a permanent loss. Leverage does not raise your returns and your risk in equal measure — it raises your risk of never getting to the recovery at all. The margin risk versus cash calculator shows the difference in rupees, and the derivatives article carries the regulator’s own data on how this works out for individual traders.

The mirror image is how a position earns its way up. Start at a third or half of the intended size, and add as the business confirms the thesis in its results — not as the price rises, and emphatically not as the price falls into a thesis that has broken. That last distinction, between adding to something working and averaging into something broken, is the single most expensive confusion in retail investing and it gets a whole section in the beginner mistakes article.

The Portfolio View

What everything can lose at the same time

One more addition, and it is the one that catches disciplined people out. If you hold twelve positions and each is risked at 1%, your portfolio is carrying 12% of total risk. On a normal day that number is meaningless because positions fail one at a time. In a market-wide fall they do not fail one at a time — that is what a market-wide fall is. So set a ceiling on total open risk as well, perhaps 5–6% across everything, or hold fewer positions, or knowingly accept the larger figure with your eyes open. What you should not do is arrive at 12% by accident and discover it during the week that tests it.

Two outer caps sit above all of this, and they matter more than any per-trade rule. The first is an emergency fund that is not in the market, so that a job loss or a hospital bill is never funded by selling shares in a bad month. The second is asset allocation — the share of your total net worth sitting in equities at all. That single decision moves your outcomes more than every stock choice you will ever make, and it is the largest position-sizing decision in your financial life even though nobody calls it that. Money you might need within five years should not be in shares regardless of how good the shares are.

The through-line of this article is worth stating plainly. You cannot control what a company’s shares do after you buy them. You can control, exactly and in advance, how much of your money is exposed to being wrong about it. Sizing is the only part of investing that is fully within your power, which is a strange thing for the least-discussed part of investing to be. If you want to go further, the regulator publishes its own free investor education material on securities market risk, and it costs nothing but attention.

Key Takeaways

---

Frequently Asked Questions

Your questions answered

How much should I invest in one stock?

Ask it as two questions instead. What is the most I am willing to lose on this idea — commonly 1% to 2% of equity capital — and where would I exit? Those two numbers give you the position. Then check the answer against a ceiling on any single holding, illustratively 5–8% for a portfolio of a dozen or so companies, and take whichever number is smaller. The habit that matters is the order of operations: loss first, position second. Anyone who starts with “I’ll put two lakh into this” has skipped the only step that protects them.

The 1% rule gives me tiny positions on a small portfolio. Is it still workable?

Honestly, often not. On ₹50,000 of capital, 1% is ₹500 of acceptable loss, which with a sensible exit distance buys a position of a few thousand rupees — and brokerage plus taxes then eat a visible share of it. Two workable responses. Use the per-position cap as your main tool at small sizes and hold fewer companies, accepting a higher percentage risk on each because the rupee amounts are small relative to your income. Or build the core in index funds while your capital grows and start direct shares with a genuinely small satellite. What does not work is quietly abandoning both limits because the numbers feel inconvenient.

Should I put a stop-loss on my long-term holdings?

Generally no, and the reason is consistency rather than bravery. If you bought a business because of what it earns, a price fall is not evidence against that reason, so selling on the price means acting on information you never used to decide in the first place. The long-term investor’s protections are different in kind: a position small enough to be survivable, written invalidation triggers about the business, and money that is not needed soon. If the thought of a 30% fall in a holding is unbearable, the position is too big. That is a sizing problem, and a stop-loss is the wrong repair for it.

My stop triggered and then the share recovered. Did I do something wrong?

Not necessarily. A rule that is right on balance will still look foolish on individual occasions, and judging a process by one outcome is how good processes get abandoned. The useful review is over twenty or thirty exits, not one. If most of your stops trigger and reverse shortly afterwards, the message is usually that they are set too close to the entry price — inside the range the share moves in on ordinary days — which, as the table above shows, is also making your positions larger than intended. Widen the exit and buy fewer shares, and the risk in rupees does not change at all.

Isn’t holding twenty stocks the same thing as position sizing?

Related, but not the same, and assuming they are is a common way to be badly exposed while feeling diversified. Twenty holdings tell you nothing if one of them is 30% of the money, if fourteen sit in the same sector, or if all twenty depend on the same interest rate cycle. Diversification asks how many different things you own; position sizing asks how much damage each can do and how much of it can go wrong at once. Twenty positions that are quietly one bet are less protected than eight capped, genuinely unrelated ones.

Keep Learning