You have had this day. The option you bought is up 30%. Your hand hovers over the exit button. You decide to be patient, because last week you sold too early and watched it double without you. Ten minutes later the position is red, your stop-loss hits, and you are staring at the screen wondering how a winning trade became a losing one.
That whipsaw is not a skill problem. It is a structure problem. And the boring fix for it is called a partial exit.
A partial exit means you sell some of your position at a planned level and keep the rest running.
If you bought 4 lots of a SENSEX weekly call — that is 80 quantity, because one lot is 20 — a partial exit might be selling 2 lots when the option is up a defined amount, and holding 2 lots with a trailing floor behind them.
Think of it like cricket. A batsman who tries to hit every single ball for six will occasionally win a match on his own, but he will also get out cheaply far more often. The batsman who takes the easy single first, and then looks for the boundary, ends up with a score. The single is your partial exit. It is not glamorous. It is what keeps you at the crease.
Say a SENSEX weekly option is bought at ₹180 for 4 lots (80 qty). Cost outlay: ₹14,400. The plan has three parts, written down before entry: a fixed stop-loss at ₹150, a scale-out of half at ₹230, and a trailing floor on the remainder.
Now look at what happens in the two futures that actually occur:
| What the market does | Scale out half at ₹230 | Hold all 4 lots |
|---|---|---|
| It reverses — runs to ₹230, then falls back to ₹165. Trailing floor exits the rest at ₹200. | ₹2,000 booked on the first half, ₹800 on the second half = ₹2,800 | Stop-loss hits at ₹150 = −₹2,400 |
| It keeps running — goes to ₹300 and you exit there. | ₹2,000 + ₹4,800 = ₹6,800 | ₹9,600 = ₹9,600 |
Read that table twice. It contains the trade-off most content on this topic hides.
So why do it? Because of the first row. The reversal case is not rare on SENSEX weeklies — it is routine, especially on Wednesday and Thursday when time decay accelerates and a move that looked like a trend turns out to be noise around a level. The partial exit converts the most common outcome from a loss into a small gain. You pay for that with a slice of your rarest outcome.
Here is the same idea drawn out. Two traders, same entries, same exits on the stop-loss. One books nothing until the end; one takes half off at a planned level.
Notice both lines climb. The difference is the depth of the dips. Those dips are where real traders break their own rules — where you double the size to "make it back", or skip tomorrow's valid setup because yesterday hurt. A smoother curve is not a cosmetic preference. It is the thing that lets a plan stay a plan.
Our own machine maps support and resistance levels on SENSEX, trades only confirmed breaks of those levels, and every position carries three fixed things: a stop-loss, a scale-out on part of the position, and a trailing floor behind the remainder. No discretion in the moment.
September has been a working month for that structure, and it looks exactly like the green line — up, with real dips. Month-to-date across 14 sessions the machine is at ₹1,30,310. The best session was ₹65,408 on 22 September. The worst was −₹26,200 on 10 September, and there were four losing days in all. We publish the red days for the same reason we publish the green ones: a month with no losses would mean we were hiding something.
A partial exit will make you feel stupid on a regular basis. You will book half at ₹230 and watch it print ₹400, and you will be certain the rule is costing you. Then a Thursday will come where a break fails inside four minutes, your booked half is the only reason the day is green, and you will not post about it anywhere.
That asymmetry in how the two outcomes feel — loud regret, quiet relief — is exactly why so many retail traders abandon scaling out. SEBI's own data shows roughly 9 in 10 individual F&O traders lose money. Very few of them lose because they booked half too early.
Partial exits are not an edge. They will not turn a bad entry into a good one. They are a way of making sure that when your method does work, you are still solvent enough to be in the seat.