Portfolio margin, liquidity, and legging risk

Three practical matters decide how much of a strategy's theoretical result you actually keep. Portfolio margin changes what you are allowed to hold, which for most people means it changes what they do hold. Liquidity decides how much of each credit the bid-ask spread takes. Legging is the way multi-leg trades go wrong at the moment of entry. None of these appears in a payoff table.

Portfolio margin

Standard margin (Reg-T) uses fixed formulas: a spread's requirement is its max loss, a short option's is a percentage of the underlying plus premium. It does not care whether positions offset each other.

Portfolio margin (PM) replaces the formulas with a stress test. The broker moves the underlying through a range of prices, typically about ±15% for single stocks and a narrower range for broad-based indices, with IV shifts at each point, and holds the worst loss found as the requirement. Positions on the same underlying offset each other fully; correlated underlyings may offset partially. Eligibility usually starts at $100,000–$125,000 of account equity and requires an options approval level and, at many brokers, a test.

The strangle at roughly half the requirement is the reason larger accounts prefer strangles to condors; the same credit needs half the capital. It is also the hazard. PM lets you hold two to four times the position count of a Reg-T account for the same buying power. The sizing rules from the previous lesson are what stop that capacity from being used; the sizing rules do not change because the broker will let you.

Three PM realities:

• The stress range is not the worst case. March 2020 exceeded ±15% in a week. A PM account sized to its buying power limit is sized to a loss the stress test did not include.
• Requirements change intraday. Brokers add concentration charges for large positions in one name and raise house requirements in stressed markets. A book that was at 60% of buying power at the open can be in a margin call by noon with no trade placed.
• Margin calls are met at the broker's prices. Liquidation happens at the market, at the widest spreads of the year. That is not a risk to model; it is a risk to avoid by never being near the limit.

Liquidity and the four-leg bid-ask

The condor from Module 3 collected $0.78 at the mid. Each leg's market:

At the "natural" price (selling at each bid, buying at each ask) the condor collects $0.68, not $0.78: 13% of the credit gone. Exit at 50% profit means buying it back, paying the natural again, another $0.10. Round trip, about $0.20 of the $0.78 credit, 26%, and on a trade whose target profit was $0.39, the friction is half the target.

Rules that follow:

• Enter and exit at or near the mid. Place the order at the mid, then walk it toward the natural in $0.01–0.02 steps if it does not fill within a few minutes. Most liquid names fill within a couple of cents of the mid.
• Skip underlyings where the combined spread exceeds about 10% of the credit. The strategy cannot overcome that friction.
• Check open interest and volume on every leg, especially the far wings, which are the illiquid ones. Open interest in the hundreds and a spread of a few cents is the minimum for a $50 stock; index products and large ETFs are far better.
• Prefer strikes that are $1 or $2.50 apart on cheaper stocks; $5 gaps on a $50 stock force wide wings.
• Never use market orders on multi-leg spreads. The fill will be at the natural or worse.

Legging risk

"Legging in" means entering the legs of a spread as separate orders. It is done to get a better price, and it introduces a risk that the multi-leg order does not have.

Enter the condor by selling the put spread first for $0.40. Before the call spread is placed, XYZ rallies 1%. The 56/61 call spread now offers $0.30 instead of $0.38. The condor collects $0.70; or you wait for a pullback and now the put spread is losing. In the worse version, the trader sells the short leg of a vertical first and places the long wing second; for those minutes the position is a naked short, and a gap in between is the strangle's tail from Module 3 for a trade that was supposed to be defined-risk.

The same applies on exit. Closing the profitable side of a condor and leaving the other "to expire worthless" is legging out, and it converts a defined-risk condor into a single credit spread with its full gamma into expiration. If the plan is to close the condor, close the condor.

Use multi-leg orders, always. Every broker's platform supports spread, condor and butterfly order types with a single net price. If the complex order will not fill, change the price; do not split it.

Legging a vertical: the window that should not exist

  t=0     sell short leg          -> naked short option
  t=0+    (waiting for better fill on the wing)
  t=?     buy long wing           -> defined-risk spread

  Everything between t=0 and t=? is an undefined-risk position
  you did not size for.

Key idea: Portfolio margin replaces fixed formulas with a stress test that offsets positions and roughly halves the requirement for strangles; it lets you hold far more than the sizing rules allow, and the stress range is not the worst case. The four-leg bid-ask takes about a quarter of a condor's credit round trip unless you work the mid; skip names where the spread exceeds a tenth of the credit. Enter and exit every spread as one order; legging creates a naked position for the minutes it takes to complete.

Try it: For a condor on a stock you follow, write down bid, ask and mid for all four legs. Compute the natural and mid credits and the round-trip cost as a percentage of the mid credit. Then do the same on a broad index ETF and compare. If your broker offers portfolio margin, look up the requirement it would show for a strangle versus the Reg-T figure.

Recap

• Portfolio margin holds the worst loss in a stress range (about ±15% for stocks) with offsets; strangles need roughly half the Reg-T requirement, which is both the appeal and the danger.
• The stress range is not the worst case, requirements change intraday, and calls are met at the broker's prices.
• A four-leg condor loses about 13% of its credit to the bid-ask at the natural, about 26% round trip; work orders at the mid and skip names where the spread exceeds 10% of the credit.
• Check open interest and spread on the wings, and never use market orders on spreads.
• Legging in or out creates an unsized naked position for the gap between fills; use multi-leg orders for entry and exit.