Drawdown Explained: Why a 50% Loss Needs a 100% Gain

You bought 500 shares at $48 in March. That was $24,000, and by June the stock printed $24. Half your money is gone, and the arithmetic you’re doing in your head says the stock only has to climb 50% to put you back where you started.

It has to climb 100%. From $24 back to $48 is a double, and no amount of wanting it to be a 50% move makes it one. The gap between the percentage you lost and the percentage you need is not a technicality — it’s the reason a drawdown gets harder to escape the longer it deepens, and it’s worth understanding before you’re inside one.

Why the two percentages can’t match

The loss and the recovery are measured against different numbers. That’s the whole thing.

Your 50% loss was measured against $24,000, the balance you had at the peak. The gain that undoes it gets measured against $12,000, because that’s the money you actually have now. Same $12,000 of movement, two different denominators, two different percentages. The dollars are symmetric. The percentages never are.

Written out, the relationship is:

Recovery gain needed = Loss ÷ (1 − Loss)

Feed a 10% loss into that and you get 11.1%. Feed in 20% and you get 25%. At the shallow end the two numbers are close enough that ignoring the difference costs you almost nothing. Then the denominator keeps shrinking, and the curve stops being polite.

All of it assumes 500 shares bought at $48, no deposits or withdrawals, no fees, no taxes, and no dividends. It’s arithmetic about one position, not a forecast about any market.

Look at the step from the 40% row to the 50% row. The stock falls another $4.80 a share — a move that would barely register as news — and the recovery bar jumps from two-thirds to a full double. One row. That’s usually where a holder stops privately believing they’ll see $48 again, even if they keep the position open and tell themselves otherwise.

The bottom of the table is where the numbers turn absurd. At $4.80 a share you’re down 90% and you need a ninefold gain. Not to make money. To get back to the price you paid.

Percentages compound off what’s left

There’s no trick hiding in the formula. Percentages always work on the balance in front of you, not on the balance you used to have. Lose half of every dollar and you are not managing a dollar anymore — you’re managing fifty cents, and fifty cents becoming a dollar is a 100% move by definition.

The same mechanism runs in the other direction, which is where it gets uncomfortable. Any strategy that doubles your money is one 50% decline from giving all of it back. A 25% gain is erased by a 20% loss. If you’re inclined to trust the arithmetic when it’s working for you, you’re obliged to trust it when it turns.

The same account, this time on margin

Now run the identical position through a margin account, because leverage doesn’t invent this asymmetry — it just moves you down the table faster than you expected to travel.

Under Regulation T, a brokerage can lend you up to half the purchase price of marginable stock, which the SEC lays out in Margin: Borrowing Money to Pay for Stocks. So the same 500 shares at $48 cost you $12,000 of your own cash and $12,000 borrowed. Exposure: $24,000. Equity: $12,000.

The stock falls to $36. That’s a 25% decline, the kind of quarter plenty of ordinary stocks have without anything structural going wrong.

Your position is now worth $18,000. The loan is still $12,000, because loans don’t shrink out of sympathy. Your equity is $6,000. The stock fell 25% and your account fell 50%, and you’re sitting on the row of the table where the recovery bar reads 100%.

Getting equity back to $12,000 requires the position to be worth $24,000 again, which requires $48 a share, which from $36 is a 33.3% gain in the stock. Your account has to double so the stock can rise a third. That ratio holds at every point on the way down and it never works in your favour.

There’s a second problem that the unleveraged version doesn’t have. FINRA’s maintenance requirement, also covered on that SEC page, obliges you to keep equity of at least 25% of the market value of the securities, and many brokers set their own house requirement higher. At $36 your equity is $6,000 against $18,000 of stock — 33%, so you’re clear. Work out where you stop being clear: equity of $500P − $12,000 has to stay at or above 25% of $500P, which holds only while P is at least $32. Four dollars a share below where you’re standing, the maintenance call arrives, and you either wire more cash or the broker sells to raise it.

That is the part the recovery table can’t show you. The table describes a path back to $48. The margin agreement describes who gets to decide whether you’re still in the position long enough to walk it. The SEC’s Day Trading: Your Dollars at Risk makes the same point about leveraged short-term trading in blunter terms: money you can’t afford to lose is money you can be forced out of at the worst possible price.

Drawdown as a number people actually use

None of this needs margin to bite. A cash account holding one stock rides exactly the same curve; it just tends not to reach the bottom rows, because a position you own outright can’t be liquidated out from under you and a diversified account rarely lets one holding drag the whole balance down 80%.

Maximum drawdown — the largest peak-to-trough decline over some period, measured before a new peak is set — is one of the standard ways funds and analysts describe how rough a strategy has been. It’s usually quoted next to volatility and return figures rather than on its own, and for a reason: it’s a single scar, not a history. It tells you how deep the worst hole went. It says nothing about when, nothing about why, and nothing about how long the climb out took.

That last omission matters more than people expect. Two accounts can report the same 40% maximum drawdown when one recovered in seven months and the other spent four years underwater. Identical number, completely different experience for whoever held it.

What this math doesn’t tell you

The formula is exact and its scope is narrow. Here’s what falls outside it.

Time. The table says a 40% drawdown needs a 66.7% gain. It has nothing whatsoever to say about whether that takes eight months or a decade, and duration is usually what determines whether someone actually holds on.

Survival. Margin calls, forced liquidations, stop-loss orders and delistings all end positions before the arithmetic gets its chance. The recovery path exists on paper. Whether it stays open to you is a separate question governed by your broker’s agreement and your own cash.

Cash flows. Deposits, withdrawals, commissions, margin interest, dividends and taxes all move the real balance. Your statement will part company with this table the moment any of them show up, and margin interest in particular accrues the entire time you’re waiting.

Behaviour. Most people don’t sit still through a 50% decline. They sell near the low, or add to the position at the wrong moment, or abandon the approach halfway down. Any of those changes the outcome, and none of them are in the formula.

Probability. The table tells you what gain is required. It is silent on whether that gain is remotely likely for the asset you happen to hold. A required 900% recovery and a plausible 900% recovery are different claims, and the arithmetic only makes the first one.

FAQ

Does drawdown math work the same for a whole portfolio as for one stock?

Yes — it’s a function of percentages, not of what’s being measured. A single trade, an account, a fund and an index all follow the identical curve. The only practical difference is that a diversified portfolio rarely reaches the deep rows, because it takes a lot of simultaneous damage to knock a spread of holdings down 70%.

Is a 50% drawdown basically unrecoverable?

Not unrecoverable, but you should be honest about what it asks. Doubling your money is a demanding outcome in any timeframe, and that’s now the price of merely returning to where you began. The deeper rows are worse still: 90% down needs a ninefold gain, which is why most risk management is aimed at never reaching that part of the table rather than escaping it.

Why doesn’t my brokerage statement show the gain I need to break even?

Statements are built to report what happened — current value, cost basis, unrealised gain or loss. The recovery figure is forward-looking and depends on which peak you’re measuring from, which is a choice the statement can’t make for you. Compute it yourself: divide the loss by one minus the loss.

Does averaging down change the recovery math?

It changes your cost basis, which lowers the bar, and it raises your exposure at the same time. Buy 500 more shares at $24 and you hold 1,000 shares at an average cost of $36, so break-even now needs a 50% move rather than 100%. You also have $36,000 committed to a position currently worth $24,000, so the same percentage decline from here costs you considerably more dollars than the first one did.

Does the formula work in reverse, for gains?

It does. Asking what loss would erase a gain is the same relationship solved the other way: a 100% gain is undone by a 50% loss, a 25% gain by a 20% loss. Worth sitting with when a position is up and starting to feel safe.

Why does leverage make a drawdown so much worse?

Because the borrowed money doesn’t share the loss. Your equity absorbs the whole decline in the position while the loan balance stays exactly where it was, so a 25% fall in a 2:1 position is a 50% fall in your account. You then need a much larger percentage gain in your equity than in the stock to get back — and you have to keep meeting the maintenance requirement the entire time you’re waiting.

Where to take this next

Pull up your own account history and find your actual largest peak-to-trough decline, not a hypothetical one. Run it through the formula and see what gain it implied at the bottom, and whether you’d have believed that gain was coming at the time. That number is more informative than any table, because it’s yours.

From there the useful topics are position sizing and where stops sit, since those are the two levers that decide how far down this curve a single bad trade is permitted to take you.

This article is general information, not financial advice. See our disclaimer.

More in Trading · all topics · calculators · how this was checked