Credit Spread Management

The playbook for selling put/call spreads without donating your max loss. Educational reference, not financial advice.

THE FIVE RULES - memorize these, skip the rest

Close winners at 50-80% of max profit.Never hold to expiry to scrape the last few percent.
Size on max loss: 1-5% of account per spread.Never size on the credit received. Write your exits down BEFORE entering.
Manage losers at 1-2x the credit against you.Max loss is the disaster you priced, not the outcome you accept.
Don't hold the endgame.Monthlies: manage by 21 DTE. LEAPS: be out 2-3 months before expiry.
Roll only for a credit, only if the thesis lives.Never pay a debit to extend a broken trade.

Motivation - why manage at all

A credit spread has capped loss, and that cap fools people. "Defined risk" means you know the worst case - it does not mean you should ever let it happen. The max loss on a typical spread is 2-3x the max gain, so one un-managed loser erases three winners. Amateurs do exactly backwards: they hold winners to expiry (squeezing pennies) and ride losers to max loss (hoping). Pros invert both. That inversion - not strike picking - is where most of the long-run edge in spread selling lives.

Theory - why each rule works

Rule 1 (close winners): risk/reward inverts as you win. Sold for $31, now worth $6? You're risking ~$75 of drawdown to earn the last $6 - a horrible new trade you re-enter every day you hold. Closing at 50-80% redeploys capital into a fresh, fairly-priced trade instead.
Rule 2 (size on max loss): max losses happen - crashes gap through stops. Sizing so any single max loss costs 1-5% means the disaster is an annoyance, not an ending. Sizing on credit ("I collected $3k") hides the $10k you actually bet.
Rule 3 (manage losers): a spread marked at 2x its credit against you is telling you the market disagrees with your thesis. Exiting there loses roughly one winner's worth - recoverable. The gap between "2x credit" and "max loss" is pure hope, and hope has negative expectancy.
Rule 4 (skip the endgame): near expiry, gamma explodes - a spread with the stock between strikes swings violently per $1 of stock move, and pin/assignment risk appears. All of the remaining "theta" you're waiting for is exactly compensation for that risk. Rule 1 has almost always fired before this matters anyway.
Rule 5 (roll for credit): rolling down/out for a net credit lowers your breakeven AND pays you - it's a fresh trade you'd plausibly take on its own. Rolling for a debit is buying back a loss and paying extra for more time to be wrong.

Approach - the lifecycle

Phase	Do this
Entry	Sell when IV is elevated (premium is fat). Ladder in tranches - never all-in on one day. Pick the short strike to match the sentence you believe ("won't fall past X"), buy the long leg where history says disasters bottom. Write down: profit target, loss trigger, thesis-break trigger.
Winning	At 50-80% of max profit: close it. Full stop. Redeploy or rest.
Losing, thesis intact	At 1-2x credit against you: roll down and/or out for a net credit, or cut size. Short strike breached on a fast move you predicted would recover: the long leg is doing its job - reassess, don't panic-close into the spike.
Losing, thesis broken	Take the managed loss. The event you sold insurance against is happening; stop being the insurer.
Near expiry	Monthlies: out by 21 DTE. Long-dated: out 2-3 months before expiry, win or lose.
Assigned early	Stay calm - the long leg still caps you. Same day: sell the assigned shares + sell (or exercise) the long put. P&L is unchanged; only the margin picture moved.

Worked example (GOOGL Dec-2028 330/230 put spread, $31 credit)

Max gain $3,100/contract (GOOGL ≥ $330 at expiry). Max loss $6,900 (≤ $230). Breakeven $299.
Plan written at entry: close at ~$7-8 remaining value (75% of max). Manage if marked -$60/sh against (2x credit) or if GOOGL breaks $280 with the thesis dead. Out by Oct 2028 regardless.
Stock rallies to $450 in 2027 → spread worth ~$8 → close, bank ~75%, done in half the time.
Stock crashes to $270 in early 2027 → marked against you, but thesis says recover-by-2028: the long 230 caps the margin requirement, so you can choose - that choice is what the long leg bought.
Funding crisis, thesis broken at $260 → roll for credit if possible, else take the ~1.5x-credit loss. You never donate the $6,900.

Why the long leg matters more than it looks

It converts open-ended margin into a fixed number. A naked short put's requirement floats and balloons exactly in a crash - the forced-liquidation-at-the-lows machine. The spread's requirement is capped at width-minus-credit and cannot grow. You keep the option to be patient.
It caps the tail below where historical disasters bottom - you pick the long strike from valuation floors, not round numbers.
It costs EV. Insurance always does. You're paying part of your edge for survivability; that's the correct trade whenever a margin call is the true catastrophe.

Built 2026-07-30. General reference distilled from standard professional practice (defined-risk premium selling; mechanical-management research popularized by tastytrade et al.). Numbers in the example are illustrative. Not licensed financial advice.