Position sizing, and how many stocks to own
The answer
Position size is what percentage of your money sits in one holding. It is the single most important decision in investing and it gets the least attention.
The decision that destroys a portfolio is almost never the stock choice. It is the size of the position, and what the holder does under stress. A good analyst at the wrong size loses more than a poor analyst at the right one.
Why this costs you money
Here is the arithmetic that nobody does before buying.
A holding that is 5% of your portfolio can fall by half and cost you 2.5% of your money. That is an ordinary bad month and you will recover from it without changing anything.
A holding that is 30% of your portfolio can fall by half and cost you 15% of your money. To get back to where you were, the rest of the portfolio must rise by about 17.6%. That takes years at a normal rate of return.
Same company. Same mistake. Two completely different outcomes, decided by a number you chose in about 4 seconds while placing the order.
Now the part that makes this worse. The size was probably never chosen at all. Most large positions in retail portfolios were not bought large. They became large because they went up. The stock you bought at 6% doubled twice and is now 22%, and nobody ever decided that 22% was right.
There is a behavioural cost on top of the arithmetic. An oversized position changes how you behave. You check the price more often. You defend it in conversation. You find reasons not to sell. The size of a position determines how honestly you can think about it, so an oversized position damages your judgement before it damages your money.
How it works
Three questions decide the size of any position. Answer them in this order.
Question 1 — How wrong can I be?
Not "will it go down". Assume the position falls by half and does not come back. Multiply that by the weight. That number is the cost of being wrong, and you must be willing to pay it before you buy.
This gives the standard limit used across the industry: no single stock above 5% to 7% of the portfolio at purchase. At 6%, a total loss costs 6% and a halving costs 3%. Both are survivable. There is nothing magic about 6%. The magic is having any number at all, decided in advance.
Question 2 — How quickly can I get out?
This is the constraint most people never apply, and it is covered in detail in the article on market makers and liquidity. Take your intended position in shares, divide it by the average daily volume of the stock, and multiply by 100.
- Under 5% of a day's volume — you can exit in a day.
- 5% to 20% — 2 or 3 days, and you will push the price down as you go.
- Over 20% — assume you cannot exit in a bad week at a price you would accept.
Then assume liquidity halves when you need it. In a small company this constraint, not the 6% rule, is often the binding one. Your position size is the smaller of what your conviction allows and what the volume allows.
Question 3 — How many of these do I already own?
If you hold 4 companies that would all be damaged by the same event, they are 1 position for sizing purposes. Add their weights together and apply the limit to the total.
How many stocks, then?
Most of the benefit of diversification arrives within the first 10 to 15 holdings, with diminishing improvement after that. Practical experience adds a constraint the research does not: you must be able to follow what you own.
| Number of stocks | What it means in practice |
|---|---|
| 1 to 4 | Not a portfolio. One bad company is a life event. |
| 5 to 9 | Concentrated. Justifiable only if you genuinely analyse each one. |
| 10 to 20 | The range most individual investors should be in. |
| 21 to 35 | Acceptable if the extra names are distinct. Watch the workload. |
| Over 35 | You are running an expensive index fund by hand. Buy the index. |
The workload is real. A proper review of one company takes several hours and must be repeated 4 times a year. Twenty companies is 4 working weeks a year. Forty is not possible alongside a job, and what happens instead is that the review stops and you own 40 things you no longer follow.
Rebalancing is the maintenance operation. Once or twice a year, trim what has grown beyond its limit and move the money to what has shrunk below it. It feels wrong every time, because you are selling what is working. That is what makes it a rule rather than a decision.
What it tells you, and what it does not
Position size tells you how much damage any single error can do, and that is the only variable in investing you control completely. You do not control returns. You do not control the market. You control weight.
It does not tell you which stock to buy, and it does not improve a bad idea. Correct sizing on a portfolio of poor businesses produces a slow loss instead of a fast one. If every position is capped at 5% but 15 of your 20 companies are poor, the cap has done its job and you will still lose money. Sizing controls the range of outcomes. It does not control the average.
And there is an honest cost. A 5% cap means your best idea can never make you rich quickly. Large fortunes are usually built through concentration, held by people who were right and could survive being wrong. That argument is examined below.
The decision rule
Size against the loss, not against the conviction.
Before you place the order, answer 3 questions with numbers:
- If this falls by half and never recovers, what percentage of my total money
is gone? If that number would change my life, the position is too big.
- What percentage of one day's average volume is my whole position? Over 20%
means I cannot leave when I want to.
- What else do I own that would fall on the same news? Add those weights to
this one before applying the limit.
And one rule for what you already hold: no position may grow past twice its intended weight without being trimmed back.
Try this now
Five minutes. This is the calculation that this whole cluster exists for.
- Open your holdings and note the total portfolio value.
- Find your largest single holding by value. Divide it by the total, multiply by
- That is its weight.
- Multiply that weight by 0.5. That is the percentage of your entire portfolio you would lose if that one holding fell by half.
- Now write down, in dates, how long it would take to earn that back at 10% a year. Divide the loss percentage by 10 for a rough answer in years.
- Repeat step 2 for your top 3 holdings combined.
- Finally, take that largest holding, find its average daily volume, and divide your share count by it. Multiply by 100.
What you should see. Most people find their largest holding is between 12% and 30% of the portfolio, and almost nobody chose that number. Step 3 converts it into an honest sentence: "if this one company halves, I lose 15% of everything I have."
Step 4 is the part that changes behaviour. A 15% loss takes roughly a year and a half of average market returns to recover. A 25% loss takes about 2 and a half years. Written as time rather than as a percentage, the number stops being abstract.
Step 6 tells you whether you could act on any of this. If your largest holding is also more than 20% of a day's trading volume, the decision has already been taken from you.
Three real cases
1. Archegos Capital Management, 26 March 2021 (United States) — size, not selection Archegos was the family office of Bill Hwang. It held very large positions in a small number of listed companies, built using total return swaps arranged through several investment banks, which meant no single bank could see the whole picture. Reported exposure was in the tens of billions of dollars against far smaller equity. When the underlying share prices fell, Archegos could not meet margin calls, and on 26 March 2021 the banks began selling the collateral. Credit Suisse reported losses of about $5.5 billion, Nomura about $2.85 billion, Morgan Stanley about $911 million and UBS about $774 million. In November 2024, Hwang was sentenced to 18 years in prison for fraud and racketeering. The stocks were ordinary listed companies. The positions were not.
2. Pershing Square and Valeant Pharmaceuticals, 2015 to March 2017 (United States) — a very good investor, a very large position Bill Ackman is among the best-known activist investors of his generation, with a long record of detailed research. His fund built a large position in Valeant Pharmaceuticals. After the Philidor allegations of October 2015 the share price fell steeply, and Pershing Square exited in March 2017 at a reported loss of about $2.8 billion. The research was extensive. The conviction was genuine. Neither of those things reduces the cost of being wrong at that weight, and the fund's investors experienced years of underperformance as a result.
3. Melvin Capital, January 2021 (United States) — the exit was priced by somebody else Melvin Capital held a large short position in GameStop, meaning it profited if the share price fell. In January 2021 the price rose sharply. Melvin was reported to have lost about 30% of its value by 28 January and about 53% over the month, and received $2.75 billion from Citadel and Point72. The firm closed in May 2022. A short position has no upper limit on the loss, which makes size the only defence available, and the size was too large to allow a calm exit.
The question that resolves it
A novice asks: how much do I want to own of this?
An expert asks: how much of this can I own and still be able to think clearly, and still be able to leave?
The first question is answered by how good the idea feels. The second is answered by 2 numbers you can calculate before you buy.
What would make this wrong
If position size were secondary to stock selection, then the largest investment disasters would be concentrated among people who chose bad companies. They are not. Long-Term Capital Management, Archegos and Pershing Square's Valeant position were all run by people whose analysis was, on the facts available, serious. What they had in common was size relative to the ability to be wrong.
The honest limits, and they matter.
Concentration is how large fortunes are actually made. Almost every very wealthy investor got there by owning a great deal of one or two things for a very long time. A 5% cap guarantees you will never do that. The condition attached to that argument is strict: concentration is defensible only when there is no borrowed money, no chance you must sell at a bad time, and no position so large against daily volume that you cannot exit. Remove any one of those and concentration becomes the thing that ends the portfolio.
There is also a limit on the rules themselves. A 6% cap applied to 20 highly correlated stocks does nothing. The cap works only when applied to exposures, and grouping holdings into exposures requires judgement no rule can supply.
And rebalancing is not free. Selling triggers tax and brokerage in a taxable account, and mechanical rebalancing in a strong trend means selling winners early. It is a trade of return for stability, not a way to get both.
In India
Sizing in an Indian portfolio has to account for a market structure feature covered earlier in this wiki: there are no designated market makers in the Indian cash equity market. No firm is required to quote a price in an ordinary listed company. In a falling market a small company can go days with almost no trading, and price bands can stop trading entirely at the lower limit.
The practical consequence is that the volume test binds much harder in India than the percentage test. A 5% position in a company that trades ₹2 crore a day may be impossible to exit in a bad month, however reasonable 5% sounds.
A workable set of Indian limits, to be adjusted to your own situation:
- Large companies: up to 5% to 7% per name.
- Mid-sized companies: up to 4%, and check the volume test.
- Small companies: 2% to 3%, and the volume test decides, not the percentage.
- Total in small companies: a stated cap for the whole group, because they fall together.
Systematic investment plans help with sizing in a way that is rarely described as sizing. A monthly amount builds a position gradually and removes the decision about when to put a large sum in. The same logic works for individual stocks: build a position in 3 or 4 instalments rather than 1.
Rebalancing in India creates a tax event. Selling within 12 months attracts short-term capital gains tax at 20%; after 12 months, long-term gains are taxed at 12.5% with the first ₹1.25 lakh of such gains in a financial year exempt. Rebalancing after the 12-month mark, and using the annual exemption deliberately, reduces the cost of the discipline.
In the United States
The percentage limits are the same. The differences are in the tools and in one very common concentration problem.
Instant diversification is cheap. A single total-market index fund holds thousands of companies for a few hundredths of a percent a year. An American investor can hold a diversified core in 1 product and run a small number of individual positions around it. That enforces sizing by itself: the core is the default, and any individual stock must justify taking weight away from it.
Fractional shares remove a real obstacle. You can put exactly $600 into a position rather than buying whole shares, so the intended weight and the actual weight match.
Employer stock is the common failure. Company stock in a 401(k), restricted stock units and employee stock purchase plans mean that for many American employees the largest holding is the company that also pays the salary. If that company has trouble, the job and the savings fail together. Financial planners commonly suggest a limit of 10% to 15% of investable assets in employer stock.
Rebalancing inside a 401(k) or an IRA has no tax cost, so an American investor can enforce sizing rules mechanically in those accounts for nothing.
Where they differ, and what that tells you
The rule is identical in both countries. The binding constraint is different.
In the United States, an investor of ordinary size can usually exit any listed position in a day. Liquidity is rarely what limits the size of a retail position. The binding constraint is tax in a taxable account, and concentration in employer stock.
In India, liquidity is frequently the binding constraint, because there is no market maker obligation in cash equities and because small companies trade thinly in exactly the conditions that make you want to sell. The percentage rule can be satisfied while the position is still impossible to exit.
What that tells you is which test to run first in each country. An American investor should start with the percentage limit and check liquidity only for genuinely small companies. An Indian investor should run the volume test first for anything outside the largest companies, because it will often produce a smaller number than the percentage rule, and the smaller number is the answer.
Imported sizing advice usually assumes American liquidity. Applied unchanged in an Indian small-cap portfolio, it produces positions that look correctly sized on a spreadsheet and cannot be sold.
Carry this
- The stock choice decides whether you are right. The position size decides whether it matters.
- Size against the loss: weight × 50% fall = the honest cost of being wrong.
- Your size is the smaller of what conviction allows and what daily volume allows.
- No position grows past twice its intended weight without being trimmed.
- Between 10 and 20 holdings is where most individual investors belong. Above 35 you are running an index fund by hand.