The 45-DTE short straddle on NIFTY: a famous rule whose profit target never fires

The idea

“Sell options with about 45 days to expiry, take profits at 50%, get out by 21 days” is one of the most widely taught option-selling routines in the world, popularised by Tom Sosnoff and the tastytrade research team in the US. The logic: 45 days out, option prices still carry plenty of time value, and that value decays fastest in the following weeks. Leaving at 21 days avoids the final stretch, when prices swing hardest with every move in the index.

We tested the rules exactly as they are usually stated, on NIFTY’s monthly options, with an at-the-money straddle: the call and the put at the same strike, nearest to the index.

The rules

• Entry: 45 days before the monthly expiry (the next trading day if that’s a holiday), at 3:15 pm, sell one lot of the NIFTY at-the-money call and put (strike = NIFTY rounded to the nearest 50).
• Monthly expiries only, one position at a time. With these dates the two never overlap: each position closes about a week before the next one opens.
• Exit, whichever comes first:
  • 50% profit: the call and put together fall to half of what we sold them for.
  • 100% loss: together they double.
  • Time: 21 days before expiry, at 3:15 pm.
• Monitoring: every minute. A stop that triggers on a gap fills at the gapped price.
• Cost: about 7.4 points per straddle, round trip: every charge from the published schedule (STT, exchange charges, stamp duty, GST, ₹20 per order) plus 1 point of slippage per option on each side. See Wrong costs, in either direction.

The backtest

Data: 1-minute NIFTY option prices, 61 monthly expiries, June 2021 to July 2026 (five years). Open positions are valued at every day’s close. All figures are per unit in NIFTY points; a lot is 65 units.

(t measures whether the average trade is distinguishable from zero; around 2 or more is needed before we call a result more than luck. ret/DD is the average profit per year divided by the worst drawdown.)

Year by year, per lot of 65 units:

*Trades closed from July 2021; 2026 up to 7 July.

The profit target never fired. In 61 trades, the straddle never once fell to 50% of its entry price before the 21-day exit; the best it managed was 56%. 59 trades ended on the time exit and 2 on the stop. Removing the profit target changes nothing at all. This is arithmetic, not bad luck: an at-the-money option loses its time value roughly with the square root of time, so going from 45 to 21 days leaves about √(21/45) ≈ 68% of the price if NIFTY doesn’t move. To reach 50%, the market would have to sit unusually still for three weeks. The 50% rule was designed for options sold further from the index, which lose value faster in relative terms.

What it takes in capital. A short straddle needs about ₹2.4 lakh of margin per lot. Because losses on sold options arrive during the day, we size the buffer on the worst intraday drawdown in the test: 1,252 points × 65 ≈ ₹0.81 lakh. That makes the capital about ₹3.2 lakh, on which the average profit of about ₹58,000 a year is a return of about 18% a year. That drawdown took about 25% of the capital.

The trades that hurt

The pattern is the usual one for selling at-the-money options: many modest wins (typically about 200 points), and a few losses of 400 to 850 points whenever NIFTY moved far in one direction. Over the five years, the 45 winning trades made about +8,900 points and the 16 losing trades gave back about 4,400 of them. The March 2026 stop filled at 2.01 times the entry price only because the gap happened to stop there; a larger gap would have filled worse.

Robustness tests

Changing one rule at a time:

• More data made the case. On the last three years alone (37 trades) the result was positive but not distinguishable from luck (t 1.65). With five years and 61 trades it clears our bar (t 2.28).
• Costs are not the problem. At about 1% of the premium collected, it is still profitable at triple cost.
• The timing settings matter a lot. Entering five days earlier halved the profit, and exiting three days earlier almost wiped it out. The profit-target and stop settings, in contrast, made little difference.
• Taking profit at 25% worked better than 50%. It actually fires (in two trades out of three) and gave the strongest result in the table. We report it, but we don’t score it: it’s a change found by looking at this data.
• Uneven over time. Only 14 of 22 quarters were profitable. 2023 made almost nothing, and the second half of the test made about three times as much as the first.

Risks

• Uncapped losses. Nothing is bought further out to limit a move. A crash or a gap of a few thousand points could cost far more than the −849 points in this sample, stop or no stop.
• Gaps through the stop. Positions are held for about three and a half weeks, through nights and weekends. The stop only works when the market trades through it; a gap jumps straight past it.
• No real crash in the sample. The five years include sharp falls, but nothing like March 2020.
• Few trades. One trade a month means even five years gives only about 60 trades.
• Margin calls. Margin requirements rise with volatility, exactly when the position is losing.

Pros & cons

Pros

• Simple, mechanical, and only two decisions a month.
• Profitable after costs over five years, about 18% a year on capital.
• Robust to costs, and to the exact profit-target and stop settings.

Cons

• Only 14 of 22 quarters profitable; some years made almost nothing.
• Very sensitive to the entry and exit days.
• The famous 50% profit target never fired.
• Uncapped losses, with stops that gaps can jump.

Our verdict

It passed five of our eight checks: profitable after costs beyond luck, better than a deposit, positive in both halves, still profitable at double costs, and not dependent on a few trades or one quarter. The best quarter came in at 33% of the profit, just under our limit of a third.

It failed on consistency (14 of 22 quarters profitable, against the 85% we expect from option selling), on sensitivity to its timing settings, and on sample size: 61 trades is still well short of the 100 we ask for.

The most useful finding is about the rules themselves: on an at-the-money straddle the 50% profit target is inert, so anyone following this routine is really trading a “sell at 45, exit at 21” rule with a stop. That works, modestly and unevenly, and depends heavily on the exact days chosen.