Short straddle: selling the ATM call and put, plus the 9:20 version

The 9:20 am straddle: why 9:20, and what the stops cost

The “9:20 straddle” is the version most Indian algo traders search for, and it is the template in the rules above: sell the ATM straddle five minutes after the open, give each leg its own stop, be flat by the close. The five-minute wait lets the messy first prints pass, with their wide spreads and unsettled premiums. Nobody has shown that 9:20 is special. It’s a convention that stuck.

In rupees, using the premiums from the example: sell the call at ₹180 and the put at ₹170. A 25% stop puts the call’s exit at ₹225 and the put’s at ₹212.50. If the call is hit, that leg loses ₹45 × 65 = ₹2,925 per lot. If the market then whips back and the put is stopped at cost, the day ends near −₹2,925 plus charges. Without the move-to-cost rule, a whipsaw that hits both stops costs about ₹5,690.

Which day you trade it changes the trade. The nearest Nifty weekly expires on Tuesday (as of November 2026), so a Tuesday 9:20 straddle sells options with a few hours of life left, while a Wednesday one sells almost a week of time value. Test them separately. Bank Nifty no longer has weekly options, only monthly contracts expiring on the last Tuesday of the month, so a Bank Nifty 9:20 straddle mostly sells longer-dated premium and reacts more to swings in implied volatility.

The loss side has no floor

Losses are open on both ends. In the example, Nifty finishing 600 points away costs ₹16,250 per lot; 900 points away costs ₹35,750. Anything held overnight also carries gap risk, because a stop-loss order can’t protect you from a price that opens beyond it. Weekend news and event days land on the seller.

Volatility is the quieter risk. A short straddle loses when implied volatility rises, so both legs can gain value on a day Nifty barely moves and your stop fires anyway. Watching India VIX at entry tells you what level you’re selling. It doesn’t tell you what happens next.

Then margin. On Zerodha’s SPAN calculator on 4 November 2026, with Nifty near 23,750, one lot of an ATM short straddle on the 17 November weekly needed about ₹2.1 lakh. Buying wings 500 points out on each side (which turns it into an iron fly with a capped loss) brought that down to about ₹95,000. On expiry day SEBI’s rules add an extra 2% extreme-loss margin on short index options, roughly ₹31,000 per short leg at Nifty 24,000. Margins move daily, so check your broker’s calculator before you size anything.

What one round trip costs

Take one round trip of the 9:20 version: sell both legs for ₹350 combined, buy them back for ₹290 at 3:15 pm. That’s ₹22,750 of premium sold and ₹18,850 bought. STT is 0.15% of the sell-side premium (₹34), NSE transaction charges come to about ₹15, brokerage is ₹80 for four orders at ₹20, GST about ₹17, and stamp duty is under a rupee. Call it ₹147 per lot before slippage, at rates as of November 2026. Your own broker’s numbers are in the brokerage calculator.

Slippage is usually the bigger number. Lose one point on each of the four fills and that’s ₹260 a lot. On a day the straddle decays by ₹60 you keep about ₹3,490 of the ₹3,900 gross. On a day a stop fires into a fast market, one fill alone can slip several points. If your backtest fills stops at the exact trigger price, it is flattering you.

Testing a straddle on NSE data without fooling yourself

You need minute-level prices for the actual option strikes you’d have traded, including expired weekly contracts, not an index chart with a pricing model on top. Make sure the history covers more than one volatility regime. A calm year tells you very little about a strategy whose risk lives in the rare wild day.

Change one parameter at a time and look for plateaus: entry time (9:20, 9:30, 9:45, 10:00), stop size (20%, 25%, 30%, 40%) and exit time. If 9:20 with a 25% stop looks great and 9:25 with a 27% stop looks poor, the good result is noise. Split results by weekday, since a Tuesday straddle and a Friday straddle are different trades. And distrust any test that ignores costs or has never been through a big gap day.

INDfolio’s options backtester runs this on NSE options history with the Indian cost stack deducted per trade (we build it, so weigh that accordingly). Whatever tool you use, ask what bar resolution its options data has and how it fills stops, then paper trade for a few weeks before real money goes in.

Variations worth testing

The iron fly is the one I’d test first: the same straddle with a call and put bought further out, giving up some premium for a fixed maximum loss and less than half the margin. A combined-premium stop (exit both legs when the total rises 25–30%) whipsaws less than per-leg stops on two-way days but reacts later to a one-way move.

There is also a positional version: sell the straddle at 10:00 am the day before expiry, exit if the combined premium rises 30%, take profit when half the credit has decayed. It holds overnight, so it carries the gap risk the intraday version avoids. Re-entries and adjustments are popular too, and each adds a parameter you can overfit. Move the strikes apart and you have a short strangle.