Lesson 6 of 7
Surviving a drawdown
Work out the gain a fall demands before it happens, so a drawdown never forces a sale.
By Javier Sanz
Founder of ValueScreener. Built Ninety Nine, a securities services provider, and sold it to a tier-one global bank; ran operations and internationalization at Alpaca, the brokerage API.
Who writes thisEditorial processLinkedIn (opens in a new tab)X (opens in a new tab)
In one minute
- A fall and the gain needed to undo it are not the same size. The bigger the fall, the further ahead of it the recovery has to run.
- In calendar 2008 the S&P 500 fell 37.0 percent with dividends (Berkshire Hathaway 2025 annual report), and needed a 58.7 percent gain to get back.
- Planning for a fall in position size and in cash, before it happens, is what keeps a drawdown from turning into a forced sale.
A fall of 50 percent needs a gain of 100 percent to reverse, not 50 percent, because the gain is measured on the smaller amount left after the fall. A drawdown is survivable when a plan already accounts for that asymmetry, in what it holds and in what it can be forced to sell.
The gain needed to recover from a fall
- 10% fall11.1%
- 20% fall25.0%
- 37% fall (S&P 500, 2008, with dividends)58.7%
- 50% fall100.0%
Why the math is not symmetric
Gain needed to recover equals 1 divided by 1 − the fall.
Example1 ÷ 0.50 = 2.0x, so a 100% gain
Subtract one from the multiple to get the gain: 1 ÷ 0.63 = 1.587, a gain of 58.7 percent.
A fall of 50 percent leaves half of the starting amount. Getting back to the start means doubling that half, a gain of 100 percent, not 50. The larger the fall, the more the required gain outpaces it: a 20 percent fall needs a 25 percent gain, a 37 percent fall needs 58.7 percent, and a 50 percent fall needs a full 100 percent.
What a drawdown does to a plan
A price falling and a permanent loss are not the same thing, but a fall can be forced into a permanent loss if money is needed before the price recovers. A position sized so a loss does not demand a sale, and cash held so a bill does not either, are what keep a paper drawdown from becoming a realized one at the worst possible time.
When a fall runs across years
The market's worst stretches are not always one bad year. The table of S&P 500 returns with dividends in Berkshire Hathaway's 2025 annual report shows 1973 and 1974 together falling 37.3 percent, and 2000 through 2002 falling 37.6 percent, each a run of losing years rather than a single one. A plan built to survive one bad year is not automatically built to survive several in a row, since a bill or a margin call does not wait for the calendar to turn.
A drawdown at portfolio scale
Check yourself
A portfolio falls 50 percent. What gain gets it back to where it started?
- 50%
- 100%
- 150%
Show the answer
Answer: B.
Half the value is left after a 50 percent fall. Doubling that remaining half, a gain of 100 percent, is what restores the original amount.How much did the S&P 500 fall in 2008, with dividends included?
- 18.1%
- 37.0%
- 50.0%
Show the answer
Answer: B.
The Berkshire Hathaway 2025 annual report's table of S&P 500 returns with dividends shows a fall of 37.0 percent for 2008, one of the index's worst calendar years.Why does planning for a drawdown before it happens matter?
- It guarantees the portfolio will not fall
- It can keep a paper fall from being forced into a permanent, realized loss
- It means cash is never needed again
Show the answer
Answer: B.
A fall that would have recovered becomes a realized loss when a need for money forces the sale first. Sizing and cash held in advance remove that force. They cannot help a holding that never recovers.Sources
- Berkshire Hathaway 2025 annual report, table "Berkshire's Performance vs. the S&P 500", S&P 500 with dividends by calendar year, 1965 to 2025.
What this leaves out
- The recovery arithmetic is exact and universal. It says nothing about how many years a real recovery takes.
- The 37.0 percent fall in 2008 is a real, filed figure. The other falls in the chart are round, illustrative sizes, not forecasts.
- Some declines, in a single company or in an index, never recover. The math here assumes the price eventually does.
- Nothing here is advice about any security or a forecast of any future fall.