The iron condor is usually the first option-selling structure a retail trader is taught, and for a sensible reason: the loss is capped. You sell an out-of-the-money call and put, then buy a further-out call and put as "wings", so that no market move can cost you more than the distance between the strikes minus the credit you took in.
What is taught less often is what those wings cost. This article looks at that cost on real NIFTY intraday data, why it is larger than it looks, and why the obvious conclusion — "so drop the wings" — is not the end of the story.
A condor is a strangle with insurance attached
Strip a condor down and it is two trades stacked together:
- a short strangle (sell an out-of-the-money call and put), which is where the premium comes from, and
- a long strangle further out (the wings), which is the insurance.
The wings are options you own, so they decay too. Every hour you hold the position, the short legs earn theta and the long legs pay it. The condor's gross profit is therefore mechanically lower than the strangle's — the difference is the wings' own decay.
What the data showed
We compared the two structures over the same set of roughly 320 NIFTY sessions, entering at 09:45 and exiting at 15:00, using real one-minute option prices and a full cost model (modelled bid-ask slippage plus statutory charges). The short strangle (0.25 delta) netted about +2.6 index points per unit of NIFTY after costs; the iron condor (0.25 / 0.10 delta) finished below it.
Two things happen at once:
- Gross falls. That is the decay of the 0.10-delta wings over roughly five hours.
- Costs rise. A condor has four legs instead of two, so it crosses the bid-ask spread eight times per round trip instead of four, and pays charges on every order.
The wings take a bite out of both sides of the ledger. In an earlier, separate study on 71 weeks of NIFTY options data, the same pattern appeared from a different angle: condors lost 31–60% of their gross profit to transaction costs, against 7–13% for strangles, under two independent cost assumptions.
None of this was a surprise in hindsight. Moving the wings further out had already improved the condor, because cheaper wings decay less and cost less to trade. Removing them entirely is just the end point of that trend.
Why "just drop the wings" is the wrong takeaway
If the article stopped here, it would be advertising naked option selling. It shouldn't stop here, for four reasons.
1. The improvement was not statistically significant. The strangle's advantage over the condor was consistent, but its own net result was not distinguishable from zero. Its bootstrap confidence interval included zero, and removing just the ten best sessions out of roughly 320 cut its headline risk-adjusted result by more than half. A result that depends on a handful of days is fragile, however good the average looks.
2. The return distribution is ugly. Short-premium returns are negatively skewed — many small wins, occasional large losses. The naked structures had meaningfully negative skew and fat tails. Once a performance statistic is adjusted for that shape and for the number of variants tried (a deflated Sharpe ratio), the strangle failed the test. The condor's bounded loss is exactly what keeps its distribution better behaved.
3. The tail is unbounded, and this sample didn't test it. In this sample the worst naked-strangle session was worse than the condor's worst, but not catastrophically so. That comfort is a property of the sample, not of the structure. A condor's worst case is written into its strikes; a strangle's worst case is whatever the market does next.
4. Margin changes the arithmetic. A naked strangle is margined on exchange risk models and typically ties up several times the capital of a defined-risk condor. A better absolute result can be a worse return on capital.
What helped, and what didn't transfer
Because naked positions need a stop, we tested stop-loss rules using each trade's maximum adverse excursion. The counter-intuitive finding: wide stops hurt and tight stops were nearly free. A stop at half the credit received reduced the average result by under 10% and cut the worst single-session loss nearly in half. Wide stops, by contrast, gave up profit without much protection. The robust lesson is "tight beats wide", not any specific level.
Two further checks are important:
- It did not replicate on BANKNIFTY. The same structures over eight years of BANKNIFTY data produced a negative result.
- Better cost data narrowed the result further. When we later re-ran the strangle-plus-stop rule with a spread model calibrated on our own recorded quotes, it came out roughly break-even. The earlier, more flattering figure owed part of its size to simpler fill and cost assumptions.
What to take away
- The wings of a condor are not free insurance. On intraday NIFTY they cost a large share of what the strategy earned, through both decay and transaction costs.
- Removing them improves the average but replaces a bounded loss with an unbounded one, worsens the shape of the return distribution, and raises the capital required.
- A structure comparison is not a strategy. "Strangles beat condors on this sample" is a statement about relative cost, not evidence that either one makes money after realistic costs, on another instrument, or through a crash.
If you sell premium, the useful question is not "condor or strangle?" but "what am I paying for protection, and is the protection I'm buying the protection I actually need?"