An iron condor is a beautiful structure — defined risk, time decay working for you, and a wide range of outcomes where you win. The problem is expiry day. On the last session of the weekly, gamma explodes, price can tag a short strike in one candle, and the "let it expire" dream becomes a 3R loss while you're at lunch. This note is the decision framework we use on Thursday mornings when a Nifty or Bank Nifty condor is still open.
As always: everything below is educational. We are a market-education community; Trading Arc LLP is not SEBI registered; nothing here is a recommendation. The goal of this piece is to show you how to think about condor management, not to tell you what to do with your own account.
When to leave it alone
The first adjustment decision is the one most traders get wrong: should you adjust at all? A large fraction of expiry-day condor losses come from over-management — traders who adjust a safely-positioned condor on every 20-point wiggle and pay transaction costs plus slippage for the privilege.
Our rule: do not adjust when both of the following are true:
- Spot is more than 0.8× wing width away from either short strike. For a 200-point wing in Nifty, that means spot is 160+ points inside the range.
- Realised intraday range is less than the expected move implied by the morning opening IV.
If both are true, the condor is simply paying you theta as designed. The correct action is to do nothing and check again at the next scheduled review time (we use 11:30, 13:30, and 14:30 on expiry).
Most expiry-day condor losses are self-inflicted. The trade works because you leave it alone; you don't win a condor by adjusting it into shape.
When to roll
Rolling means moving a threatened wing further out — buying back the close-to-the-money short and selling a new short at a safer strike. It costs premium (you're reducing credit), and it is only worth doing when the alternative (taking the loss) is more expensive than the cost of the roll.
We roll a side when all three of these are true:
- Spot is inside 0.5× wing width of the threatened short strike.
- Time to expiry is more than 2 hours. Rolling inside the final two hours is usually paying to be wrong slower.
- The cost to roll one side is less than 40% of the maximum remaining loss on that side.
If any one of these fails, we don't roll. We either close or hold — never adjust just because "something must be done".
How we size the roll
We roll the threatened short to a strike whose delta matches the delta we sold on the other (untouched) side. That keeps the structure delta-balanced rather than betting on a reversal. For example, if the put side is untouched at 22,400 (delta ~0.15) and the call side is threatened at 22,600, we roll the 22,600 short to whichever higher strike is currently showing a ~0.15 delta.
This is mechanical. No view on where price is going. The roll is a risk adjustment, not a directional bet.
When to close
Closing is the right answer when neither "leave alone" nor "roll" is attractive. In our framework, we close when:
- Spot has touched or breached a short strike and time to expiry is less than 2 hours.
- The remaining max-loss-if-assigned is within 1.2× of the cost of closing at the current mid.
- Realised IV is spiking fast enough that the rolling cost is climbing each minute we wait.
The emotional resistance to closing is real. You are, in effect, admitting that the trade didn't work. That is exactly what you are doing — and the alternative (hoping for a reversion inside the last two hours with gamma at a maximum) is almost always worse in expectation than locking in a known loss.
Decision tree
We keep this printed on the desk. It is a checklist, not a signal generator.
EXPIRY-DAY CONDOR CHECK │ ├─ Spot > 0.8× wing width from both shorts? │ └─ YES → LEAVE ALONE. Recheck at next window. │ └─ NO → continue. │ ├─ Threatened side: spot inside 0.5× wing of short strike? │ └─ NO → LEAVE ALONE (monitor closely). │ └─ YES → continue. │ ├─ Time to expiry > 2 hours? │ └─ NO → CLOSE threatened side. │ └─ YES → continue. │ ├─ Roll cost < 40% of remaining max loss on that side? │ └─ YES → ROLL to delta-matched strike. │ └─ NO → CLOSE threatened side.
Write down the output of the tree before you act. If the output is "roll" and your finger is on the "close" button, pause. You are overriding a rule with a feeling.
Real example walkthroughs
Example 1 — leave alone
Nifty condor: short 22,400 PE / long 22,200 PE / short 22,700 CE / long 22,900 CE. Wing width 200. On expiry morning spot opened at 22,540 and at 11:30 was at 22,555. Both shorts more than 160 points away. Realised range for the morning was 30 points against implied 55.
Checklist output: leave alone. At 14:45 spot closed at 22,590 and the condor expired for near-maximum credit. No action taken, no adjustment cost, no slippage.
Example 2 — successful roll
Bank Nifty condor: short 48,400 PE / long 48,100 PE / short 48,900 CE / long 49,200 CE. Wing width 300. At 11:45 on expiry, spot pushed to 48,830 — within 0.23× wing of the 48,900 short. Time to expiry 3h 45m.
Roll cost to move the 48,900 short out to 49,100 (delta-matched to the untouched put side): 22 points debit per lot. Remaining max loss on call side if assigned: 75 points. Roll cost is 29% of max loss — under the 40% threshold.
Checklist output: roll. We rolled, spot mean-reverted to 48,650 by 14:30, and the adjusted condor closed with a reduced but positive credit.
Example 3 — close and take the loss
Nifty condor: short 22,500 CE threatened by a 14:00 breakout to 22,520. Time to expiry: 1h 30m. Roll cost quoted at 38 points; remaining max loss 60 points. Roll cost is 63% of max loss — above the 40% threshold. Time is under the 2-hour floor.
Checklist output: close. We closed the threatened side for a known 1.1R loss. Spot continued to 22,570, which would have been a 2.4R loss on the untouched position. Closing locked in the smaller loss.
Note that the "closed" trade in Example 3 still felt like a mistake when we did it — the gut wanted to hold and hope. The tree said close. The tree was right.
Key takeaways
- Most condors don't need to be adjusted. Over-management is the leading cause of expiry-day losses.
- The three choices — leave alone, roll, close — each have specific, mechanical preconditions.
- Never roll inside the final 2 hours; gamma makes rolling more expensive than closing.
- When you roll, roll to a delta-matched strike. The roll is a risk rebalance, not a directional bet.
- Closing feels like admitting defeat. That is precisely what it is — and it is usually the correct answer when the math says so.
Condors are not a "set and forget" structure on Indian weeklies, especially in Bank Nifty where intraday ranges can be 2–3× the IV-implied move. But they are also not a structure that rewards constant tinkering. The edge lives in a small number of correctly-taken adjustment decisions per quarter — and the discipline is in having a written tree you can point at when the gut wants to override it.