Not yet listed. Not yet listed. Options cannot be backtested until we hold historical option prices, and nothing is listed without a test.
Walk into any room of Indian retail options traders and ask what they sell. A large share will say strangles. It is easy to see why: sell a call well above the index and a put well below it, and most weeks the index finishes between the two and both premiums are kept. The win rate is high, the income is regular, and for months at a time it feels like the closest thing in markets to being paid for nothing.
The strategy is an honest version of that trade. It takes the high win rate seriously, and it takes the other thing seriously too: that the loss on a strangle has no ceiling, and that one week in a few years will try to find out how far down it goes.
What you are paid for
Compared with an iron condor, a strangle collects nearly twice the premium for the same week. That extra money is not free. It is the price the market pays you for not buying insurance, and the market prices insurance well. Over a long enough stretch, the extra premium and the occasional uncapped loss roughly offset. What a seller is left with is the difference between how well they manage the bad week and how well everyone else does.
So the robot's rules are all about the bad week. It sells only when option prices are high against the past year, because selling cheap options is a poor trade even when the range holds. It enters mid-morning, after the opening noise. It skips the day before a scheduled event. It takes profit at half the premium and leaves. And it closes everything by the end of the day, every day, because the loss that has no ceiling is almost always an overnight one.
What the stops can and cannot do
There are three. If either sold option's delta rises to a set level, the index has moved towards it and that side is closed. If buying the whole position back would cost twice what it collected, it is bought back. And if the day's losses reach a small share of the account, trading stops until tomorrow.
On an ordinary day these work. A trending day closes the threatened side for a loss of a few thousand rupees per lot and that is the end of it. What they cannot do is act on a price that was never printed. If the index gaps three percent at the open on news, there is no moment at which the delta was at the exit level, and the first price available is the one you get. That is why the strategy never holds overnight, and it is also why no stop can make the worst case a known number. Only insurance does that, and insurance is the thing a strangle sells.
The stops make a large loss rare. The missing insurance is what makes it possible. Both sentences are true, and anyone selling strangles should be able to say both.
The adjustment the robot refuses to make
This is the strategy people adjust most. Roll the untested side closer to collect more. Roll the threatened side out to next week. Done well, over years, these smooth results. Done under pressure on the one bad day, they are the mechanism by which an open-ended loss becomes a very large one, because each roll adds a fresh position to a trade that is already wrong.
The robot closes. It does not roll, and it will not be talked into it at two in the afternoon. If you want the version of this trade where a bad week costs a number you have already seen, that is the iron condor, and the article on it explains the trade-off.
If you take one thing away
A strangle is paid well because it sells without insurance. The robot makes the bad week rare by selling only rich premium, exiting early and never holding overnight. It cannot make that week impossible, and it is unlisted until we can test it.
Think a stricter filter or wider stop fixes short strangle? Open the rule sheet, change the numbers, and run your own test.
Customize Short strangle