Long Straddle into a Policy Event: Cheap IV, Rich Skew, Brutal Theta
Nifty is at 24,800 with four sessions left in the weekly contract and the RBI policy decision two sessions away. Quotes: the 24,800 call is ₹133 and the 24,800 put is ₹158, so the at-the-money straddle is ₹291, or 1.17% of spot. Implied vol is roughly 13% for the ATM call, 15% for the ATM put, 17% for the 24,300 put and 12% for the 25,300 call. On the last eight policy days the index's absolute move was 0.6%, 0.9%, 1.2%, 1.3%, 1.5%, 1.9%, 2.4% and 3.1% - median 1.4%. A client wants to buy four lots (lot size 75) of the at-the-money straddle for ₹87,300, which is 30% of his trading capital, arguing: 'IV is cheap and the market always moves on policy day.' He intends to hold to expiry.
1. Compute the breakevens and the move required. Using the eight historical observations as equally likely outcomes, is the straddle a positive-expectation trade? What is the real problem with the plan?
Read the worked answer
Breakevens: 24,800 ± 291 → 24,509 and 25,091, i.e. a 1.17% move is needed by expiry, not merely on the event day.
Observation-by-observation payoff (points per unit = max(|move| − 291, 0); ×75 for one lot):
| |move| | points | payoff/unit | P&L per lot | |---|---|---|---| | 0.6% (149) | 0 | −291 | −₹21,825 | | 0.9% (223) | 0 | −291 | −₹21,825 | | 1.2% (298) | 7 | +₹525 | | 1.3% (322) | 31 | +₹2,325 | | 1.5% (372) | 81 | +₹6,075 | | 1.9% (471) | 180 | +₹13,500 | | 2.4% (595) | 304 | +₹22,800 | | 3.1% (769) | 478 | +₹35,850 |
Sum of payoffs = 1,081 points / 8 = 135 points per unit average = ₹10,125 per lot, roughly ₹40,500 on four lots, or 46% of the ₹87,300 premium.
But the edge is fragile: half the observations return nothing; the 3.1% outlier contributes 478 of the 1,081 points (44%); the sample is eight events; and the calculation assumes the event-day move is still there at expiry with no vol crush. Costs (brokerage, exchange fees, STT on the sell leg, ₹2-4 of straddle bid-ask) consume ₹300+ per lot per round trip.
The real problem: 30% of capital on a four-day trade that is a total loss in half the historical outcomes. Even a genuinely positive-expectation trade ruins an account at that size; a 1% capital risk budget would allow one lot, not four.
2. Two sessions later Nifty closes at 25,172, a 1.5% move. Implied vol drops 3 points on the news (straddle vega ≈ ₹21 per vol point, ₹1,575 per lot). Estimate the mark-to-market P&L that evening, and the P&L if he then holds to expiry and the index drifts back to 24,850.
Read the worked answer
On the event evening (two sessions left):
- The 24,800 call is 372 points in the money; the put is worthless.
- Remaining time value is small (two sessions, post-event vol): the straddle would be about 380 before vega.
- Vol crush: 3 × ₹21 = −₹63 per unit.
- Mark ≈ 380 − 63 ≈ 317 against a cost of 291 → +₹26 per unit ≈ +₹1,950 per lot, i.e. under 10% on a trade that was directionally right by 1.5%.
- After bid-ask and costs, roughly scratch.
If held to expiry and the index drifts to 24,850: settlement value = 50 points.
- P&L = 50 − 291 = −241 per unit = −₹18,075 per lot = −₹72,300 on four lots (−83% of premium).
Why this happens: a long straddle is a bet on realised volatility to expiry, not on the event-day move. Three things attack it simultaneously once the event passes: vega collapse as the event premium is released, theta accelerating into expiry, and the drift that gives back the move. The trade was 'right' on the event and still lost four-fifths of the premium.
Rule: with a catalyst inside the option's life, a long straddle is a trade to monetise, not to hold. Set an exit trigger before entry - for example, close if the straddle marks 15-20% above cost, or close on the first session after the event regardless of the mark - and size so that total premium at risk is a small fraction of capital.
3. Given the skew numbers (24,300 put at 17%, 25,300 call at 12%), how would you restructure the same 'big move' view? Price the alternatives approximately and state which view each one actually expresses.
Read the worked answer
The at-the-money straddle forces the buyer to purchase the downside at a 2-4 vol premium to the upside - he is paying for the skew, not just for volatility.
Alternatives (approximate, four sessions, 13-17% vol):
- Call only, ₹133. Breakeven 24,933 (+0.54%), maximum loss ₹9,975 per lot. Cheapest convexity, but only expresses an upside view; a crash loses the premium.
- Strangle: buy 24,800 call ₹133 + buy 24,300 put ≈ ₹30-35 → ≈ ₹165. Upside breakeven ≈ 24,965 (+0.67%); downside breakeven ≈ 24,135 (−2.68%); maximum loss ₹165 per unit (₹12,375 per lot). Cheaper than the straddle with a much better upside breakeven, but it gives up the 24,300-24,800 payoff zone and demands a much larger fall to pay on the downside.
- Risk reversal: buy 24,800 call, sell 24,300 put. Near-zero cost, but the sold put is the expensive option and the position is short the tail - the opposite of what a 'big move' buyer wants.
- Put-only or put-heavy structures are justified only if the view is specifically a downside shock; the 17% put is the market's priced insurance and is not cheap.
Judgement: if the client expects a large, roughly symmetric move, the strangle is the more cost-efficient expression and reduces the maximum loss from ₹87,300 to about ₹49,500 for four lots. If he expects a moderate move (1-1.5%) in either direction, only the straddle pays in the 24,300-24,800 band, and then the correct fix is size (one lot), not structure. Selling the skew by buying OTM puts is a bet that skew is overpriced - a different trade from buying event volatility, and it should be stated as such.