Risk and selling · Rebuilt with the Oct 13, 2026 close
One stock too many: what concentration does to an account
No signal on one stock can protect an account that rides on it. The arithmetic of a big position, and how often single stocks fall hard.
A signal is about one stock. An account is about all of them together, and how much of it rides on each. No Buy, Hold or Sell, however good, can keep an account steady when most of it sits in one or two stocks and those stocks have a bad week. This piece is about the arithmetic, using the price history of every stock Be The Puck covers.
The arithmetic
An account moves by each position's share times that position's move. A stock that's half the account and falls 20% takes 10% off the whole account by itself, whatever everything else does:
| One stock is… | It falls 10% | It falls 20% | It falls 40% | It falls 70% |
|---|---|---|---|---|
| 10% of the account | −1.0% | −2.0% | −4.0% | −7.0% |
| 25% of the account | −2.5% | −5.0% | −10.0% | −17.5% |
| 50% of the account | −5.0% | −10.0% | −20.0% | −35.0% |
| 75% of the account | −7.5% | −15.0% | −30.0% | −52.5% |
The change in the whole account from that one stock alone.
How often single stocks fall hard
- A typical covered stock had a week of 10% or worse about 1.7 times a year since 2006. The S&P 500 had 3 such weeks in 21 years.
- 84% of the 344 stocks with that history fell 50% or more from a peak at some point, and 51% fell 70% or more. The median stock's worst fall was −71%. The S&P 500's worst, in 2008–09, was −57%.
- Over the last three years the median stock swung about 34% a year (annualised volatility), against 15% for the S&P 500.
And these are the survivors: stocks that went to zero aren't on the list, so the real history of single stocks is worse.
Two stocks in the same business are close to one
Owning several stocks only spreads the risk if they don't fall together. Within a sector they often do. The average correlation of weekly moves over the last three years (1 means they move together every week, 0 means no relation):
| Covered stocks in | Stocks | Average correlation |
|---|---|---|
| Semiconductors | 23 | 0.49 |
| Metals & mining | 51 | 0.56 |
| Big tech | 6 | 0.37 |
| Financials | 26 | 0.28 |
| Energy & power | 59 | 0.24 |
| Any two covered stocks | – | 0.12 |
Two chip stocks, two gold miners or two bitcoin-linked stocks tend to have their bad weeks together.
What Be The Puck does about it
- If you load your holdings into the watchlist with quantities or values, it now says when one name is a quarter or more of the list, and what a 20% fall in it alone would do. That stays in your browser.
- The signal now carries exit rules (when to sell), so a falling stock loses points. That helps with the stock, not with how much of an account it is.
How much to put in any one name depends on your goals, your timeline and what else you own, which Be The Puck doesn't know. It's the kind of question an advisor is for.
This analysis does not constitute trading advice. Please meet with an advisor or independently review sources before making any decision. Be The Puck reports what published rules and public data say; it doesn't know your goals, taxes or timeline.