← AlgoRishi Blog · algorishi.com

Drawdown explained: why a 50% loss needs a 100% gain, and how to size for the worst case

15 September 2026 · AlgoRishi

Ask a trader how his strategy is doing and he will tell you the return. Ask a professional the same question and he will tell you the drawdown first.

Drawdown is the single number that decides whether you are still in the game next month. Here is what it means, why the maths of recovering from it is so brutal, and how position size gets built backwards from the worst case instead of forwards from the best one.

−20%needs +25% to recover
−33%needs +50% to recover
−50%needs +100% to recover
20SENSEX option lot size

What a drawdown actually is

Drawdown is the fall from the highest point your account has ever reached, down to its lowest point after that, before it makes a new high.

Suppose you start with ₹2,00,000. Over six weeks you grow it to ₹2,60,000. Then a rough patch takes you to ₹2,08,000. Your drawdown is ₹52,000 from the peak — exactly 20%. You are still up ₹8,000 from where you started, and it still feels terrible, because your brain anchors to ₹2,60,000. That gap between "up on the year" and "down from the peak" is where most traders quit or double their size. Both are fatal.

peak equity trough this gap is the drawdown Account equity over time — measured from the highest point, not from the start
Drawdown is always measured from the peak. The starting balance is irrelevant once a new high is made.

The recovery maths nobody wants to hear

Losses and gains are not symmetric, and this is where most retail accounts die.

Lose 50% of ₹2,00,000 and you have ₹1,00,000. To get back to ₹2,00,000 you now need to double the smaller account — a 100% gain to undo a 50% loss. The deeper the hole, the steeper the ladder out.

DrawdownGain needed to get back to the peak
10%11.1%
20%25%
33%50%
50%100%
75%300%
Below about 20% the maths is annoying. Past 50% it stops being a trading problem and becomes an arithmetic one. This is the real reason professionals cap risk per trade at levels that look boringly small.

The auto-rickshaw way to think about it

An auto driver in Pune does not decide his day by his best fare. He knows roughly what a bad day looks like — slow morning, one breakdown, petrol up — and he makes sure that bad day still leaves him able to come out tomorrow. The good days take care of themselves.

Position sizing is the same trade-off. You do not size for the day the break runs 80 points. You size so that the worst realistic run of losses still leaves an account big enough to keep trading the strategy that produces the good days.

Sizing backwards from the worst case, with real SENSEX numbers

SENSEX weekly options: lot size 20, strikes in steps of 100, expiry on Thursday. Say a strike is trading at ₹400 premium. One lot therefore costs 20 × ₹400 = ₹8,000 to buy.

Now apply a fixed stop of 22 points on the premium. Per lot, that stop is worth 22 × 20 = ₹440. That ₹440 is your actual risk per lot — not the ₹8,000 you paid, as long as the stop is honoured and the market is liquid enough to fill it.

The sizing question then becomes arithmetic rather than emotion:

Notice what never entered that calculation: conviction, a tip, how sure you feel, or how much you lost last week. Size came out of the stop distance and the streak length, both of which you can count.

Write down your peak equity every Friday. Drawdown is impossible to manage if you cannot say what your high-water mark was.

Three things that quietly make drawdowns worse

Illiquid strikes. A stop is only as real as the bid on the other side. Far out-of-the-money SENSEX strikes can gap straight through a stop level, so the loss you planned and the loss you take are different numbers.

Expiry day decay. On Thursday, an option's time value drains fast. A position that would have been a small loss on Monday can be a total loss on expiry afternoon, because the premium has nothing left in it to give back.

Increasing size after losses. Doubling up to "get it back quickly" converts a routine drawdown into an account-ending one, because it is exactly when the hole is deepest that you have the least capital to absorb another loss.

A drawdown seen in a backtest is a floor, not a ceiling. Live trading adds slippage, missed fills, and the human decision to abandon the system halfway down. Assume the real worst case is deeper than the tested one. F&O carries a genuine risk of losing your entire capital.

None of this is a method for making more money. It is a method for still being here when the strategy's good stretch arrives — which, in break trading, is usually a small number of sessions hiding inside a large number of ordinary ones.

Disclosure: AlgoRishi publishes signal tracking on real market prices using virtual capital, for information and education only. Nothing here is investment advice or a recommendation to trade. Futures & options trading involves substantial risk of loss. Past performance is not indicative of future results. AlgoRishi is not a SEBI-registered investment adviser or research analyst.