MyAlgoGuru Help
Education Glossary Open the app
Chapter 4 of 5

Per leg, or per basket?

A four-leg structure has a known maximum loss — as long as it has four legs. Risk measured only on the net hides the one leg that can hurt you.

A short iron condor has a defined maximum loss. That is the entire reason people trade it: whatever happens, the loss is capped by the long wings.

The sentence is true and it contains a condition most traders never examine. The loss is capped as long as the structure has all four legs. Remove one — because it did not fill, because you closed it, because the strike went untradeable — and the thing you are holding is not a condor with a cap. It is a naked short option, and its loss is not capped by anything.

break even all four legs: profit capped, loss capped three legs: nothing stops this long put short put short call long call The missing leg is the cheapest one in the structure, and it is the only thing holding the floor up.
One unfilled leg changes the kind of position you own. The green payoff is the structure you designed: a defined profit between the short strikes and a defined loss beyond the long ones. The red line is the same position with the long put absent — identical for every price above the short put, and unbounded below it. Note which leg it is: the long wing is the cheapest leg in the structure and the one most likely to be skipped, mis-filled or closed first to free up margin. It is also the only reason the position has a floor.

Why risk must be measured per leg

If you monitor only the basket's net, a position can look calm while containing something dangerous.

Consider a condor whose net P&L is flat. That net could be four quiet legs. Or it could be a short call losing ₹80,000 and a long call gaining ₹78,000 — a net of −₹2,000, in a position that is one gap away from the long leg being unable to keep up. The net tells you nothing about which it is.

This is why a risk engine worth having evaluates each leg against its own limit, not the structure against a single number. A basket-level stop at ₹20,000 does not fire while the legs offset each other, and by the time the net moves it has usually moved a long way. Per-leg limits catch the leg that is in trouble while it is still only one leg.

The same logic applies to how you read a P&L figure at all. A sum over three of four legs is not the basket's result — it is a smaller, unrelated number that looks exactly like a result. When one leg has no usable price, the honest display is the partial sum labelled as partial, so you never mistake three legs for four. A figure that quietly sums whatever it has will be confidently wrong on precisely the days something is wrong with the data.

All or none

A multi-leg structure should go on as a unit or not at all.

When legs are placed one at a time, there is a window between the first fill and the last in which you hold something that is not the strategy. Usually that window is a second and nothing happens. Occasionally the market moves inside it, a later leg fills at a materially worse price, and your structure's economics are not what you designed. And sometimes a leg does not fill at all, leaving you holding a position whose risk profile you have never evaluated.

Three things make this manageable:

Order the legs deliberately. Place the protective legs first. If you are going to be caught mid-structure, be caught holding the hedge without the risk, not the risk without the hedge. Buying the long wings before selling the short strikes costs you a few seconds of margin and removes the only genuinely dangerous intermediate state.

Put a timeout on the fills. If the structure is not complete within a few seconds, something is wrong — a strike is illiquid, the book has moved, a leg was rejected. A timeout that unwinds what filled is better than a position nobody designed.

Know what happens on a partial. Before you place it, decide: if leg three fills and leg four does not, do you complete it at the market, unwind the rest, or hold and manage? Decide this in advance, because you will have about ten seconds to decide it live, and the default — doing nothing — leaves you in the red line on the diagram above.

Legging out is the same problem in reverse

Exiting is where this bites most often, because the exit usually happens under pressure.

The structure is losing, margin is tight, and the long wings have barely any value left. Closing them releases margin and recovers almost nothing, so it feels like the obvious first move. It is also exactly the wrong one: you have just sold the floor out from under the position and kept the part that can run.

Close short legs first, long legs last. At every intermediate moment you are then holding something less risky than what you started with, not more. If you can only get part of the way out, you want to be left holding long options, which can lose at most what they are worth.

Illustrative

A short strangle with long protective wings, four legs. The market moves hard against the call side.

Exit A — wings first. The long call is worth ₹4 and the long put almost nothing; closing both returns ₹3,000 and frees margin. Now the position is a naked short strangle in a trending market. Over the next twenty minutes the short call doubles and the loss is bounded by nothing but the close.

Exit B — shorts first. The short call costs ₹46,000 to buy back; the short put closes for almost nothing. Now the position is two long options with no risk beyond their remaining value, and you can close them calmly or let them expire.

Exit B is more expensive at the moment you do it and it is the only one of the two that has a known worst case. Exit A felt cheaper because it looked at the cash and not at what was left behind.

The number to size on

Which brings us back to sizing. For a multi-leg structure, what is "risk per unit"?

If you can rely on the structure being complete, it is the basket's defined maximum loss, and sizing is straightforward. If you cannot — if legs are placed separately with no all-or-none guarantee, or if a wing sits on a strike that does not reliably trade — then your real risk per unit is the risk of the structure that survives a partial fill, which for a condor is the naked short.

That is a much larger number, and it is the one to size on until the execution is reliable enough to justify the smaller one. Sizing on a cap that depends on a fill you have not guaranteed is the same error as sizing on a stop you have not placed.

What to take from this chapter

  • A defined maximum loss is a property of the complete structure. One missing leg can make it unbounded.
  • The leg most likely to be skipped or closed first is the cheap long wing — the only thing holding the floor up.
  • Measure risk per leg. A flat net can hide one leg losing heavily against another gaining.
  • A sum over some of the legs is not the basket's P&L. Label a partial as partial or do not show it.
  • Place protective legs first, time out incomplete structures, and decide your partial-fill response before you place the order.
  • Exit short legs first and long legs last, so every intermediate state is less risky than the last.
  • Size on the risk of the position that survives a partial fill, until your execution has earned the right to the smaller number.