What Percentage of My Portfolio Should Be in One Stock?
The common 5–10% rule of thumb, why it exists, what a big position does to your losses, and what “top 5 concentration” tells you about your portfolio.
Both camps have evidence on their side. Big winners drive most stock market gains, and the same winners have fallen 60% or more along the way. Here is how to weigh it, the rules investors use, and the tax angle.
What is your own portfolio exposed to? Check it with Exent →
The deciding question is not whether the stock will keep rising. It is what happens to your plans if it falls hard. Size the position by its worst case in dollars. Illustrative, a $1 million portfolio:
| Position size | Value | If it falls 66% | Loss as share of portfolio |
|---|---|---|---|
| 5% | $50,000 | −$33,000 | −3.3% |
| 10% | $100,000 | −$66,000 | −6.6% |
| 20% | $200,000 | −$132,000 | −13.2% |
| 30% | $300,000 | −$198,000 | −19.8% |
If a 13% hit to the whole portfolio would change your retirement date or make you sell in a panic, the position is larger than your plan can carry. If you could shrug it off, letting it run is a reasonable choice. Our guide to a single-stock concentration limit covers how people set the cap.
Stock returns are lopsided, and the big winners do most of the work. Hendrik Bessembinder’s 2018 study found that 58% of US stocks since 1926 failed to beat Treasury bills over their lifetimes, while 4.3% of companies accounted for all the net wealth the market created. Selling a true compounder early can cost more than any loss you avoid.
NVIDIA is the clearest recent case: a total return of 13,575% over the 10 years to September 2026 in Exent’s data. An investor who trimmed every time it doubled would have kept far less of that. See the NVIDIA data page.
Winners fall hard too, often more than once. Every stock below finished the decade well ahead, and every one lost more than half its value on the way.
| Stock | Largest drop, 10 years | Calendar 2022 |
|---|---|---|
| META | −76.7% | −64.2% |
| NFLX | −75.9% | −51.0% |
| TSLA | −73.6% | −65.0% |
| NVDA | −66.4% | −50.3% |
The table uses data as of September 30, 2026. Two further points. You only know a stock was a compounder in hindsight; at the time, a 60% drop looks like a broken story. And a large position concentrates more than price risk: one company’s earnings, management and regulators now drive a fifth of your wealth.
| Rule | How it works | Trade-off |
|---|---|---|
| Hard cap | Trim back to a set maximum, for example 10%, whenever the stock goes over | Simple; sells the best performer repeatedly |
| Band | Trim only when it passes a wider line, such as 15%, back to 10% | Fewer trades and taxes; more drift between trims |
| Sell the cost | Sell enough to recover what you paid; let the rest run | Keeps upside; does not cap size if it keeps rising |
| Stop adding | Keep the shares, direct all new money elsewhere | No tax now; shrinks the weight only slowly |
| Staged sales | Sell a fixed slice each quarter or year | Spreads timing and tax; slower to reduce risk |
Illustrative examples; the cap levels are not recommendations. What matters is that you choose the rule before the next big move. A band works the same way as a rebalancing band for a whole portfolio.
Selling a winner is not a mistake in itself. The documented mistake is selling winners while holding on to losers. Terrance Odean’s study “Are Investors Reluctant to Realize Their Losses?” (Journal of Finance, 1998) analyzed trading records for 10,000 discount-broker accounts. Investors showed a strong preference for realizing gains rather than losses, the later performance of the stocks did not justify it, and in taxable accounts it led to lower after-tax returns. This is called the disposition effect.
The fix is to judge every position by the same test: would you buy it today, at this size? Apply it to winners and losers alike. See when to sell a stock.
Taxes are a real cost of trimming in a taxable account, but rarely a reason never to trim. Under IRS rules, shares held more than one year qualify for long-term rates of 0%, 15% or 20%, depending on income; shares held a year or less are taxed as ordinary income. Some ways investors soften the bill:
A tax adviser can run the numbers on your lots before you sell.
Many investors trim when a single stock passes a set cap, because a large drop would hurt the whole plan. Whether to do it depends on what a 60% fall in that stock would cost you in dollars and whether you could hold through it.
Selling a winner is not a mistake by itself. Research on the disposition effect shows the costly habit is selling winners while keeping losers, which led to lower after-tax returns.
In Exent’s data, NVIDIA’s largest drop over 10 years was 66.4%, Tesla’s 73.6% and Meta’s 76.7%. Each still finished the decade well ahead.
It means holding stocks that have risen instead of selling them to lock in gains. It relies on the fact that a few big winners create most stock market returns.
Hold for more than a year to qualify for long-term rates, trim inside tax-advantaged accounts first, offset gains with harvested losses, or spread sales across tax years.
Upload a screenshot of your holdings and get concentration, overlap, crash exposure and a risk-matched benchmark in about a minute.
Educational content, not investment advice. Examples are illustrative; past performance does not guarantee future results.