Butterfly Spreads 101: Profit If SPX Lands in the Middle, With Losses Capped on Both Sides
With SPX at 7520, you think it is going nowhere today and will probably settle near 7550 — buying a 7550 call costs real money, and selling one leaves you exposed if it really rallies. A butterfly turns that "sideways" view into a trade: buy the lower strike, sell two of the middle strike, buy the higher strike, and the four legs form a payoff that peaks in the middle and falls off on both sides; settle in the middle and you earn the most, settle outside either wing and you lose at most the net premium you paid up front plus fees. This page works out the math first, then shows how to practice it in the simulator — one thing up front: the simulator has no one-ticket butterfly button, so you practice it as two vertical spreads.
How a butterfly spread makes money
The structure is buy one, sell two, buy one: with SPX at 7520, buy one 7500 call (say, $2.00 of premium), sell two 7550 calls (say, $1.00 each), and buy one 7600 call (say, $0.40), all four legs on the same expiry; the bought legs pay, the sold legs collect, for a net debit of 2.00 − 2 × 1.00 + 0.40 = $0.40 per share — $40, the entire capital you put in up front.
If SPX settles exactly at 7550 at expiry: the long 7500 call is worth 50 points ($5,000), while the two short 7550 calls and the long 7600 call are worth nothing; minus the $40 paid up front and the $20 round-trip fee, the maximum profit is $4,940 — that is the peak of the butterfly, and landing in the middle pays the most.
There are two breakevens (premium only): the lower one is 7500 + 0.40 = 7500.40, the upper one is 7600 − 0.40 = 7599.60; settle between them and you profit; after the $20 round-trip fee ($0.20 per share), true breakeven is 7500.60 and 7599.40.
Time is the butterfly buyer's enemy: a 0DTE has only today left, and the two calls you bought lose premium faster than the two you sold — even if SPX doesn't move, the butterfly gets cheaper; this trade bets on "where it settles", not on "which way it goes", so don't use it to chase a rally.
Why the loss is capped
The worst case is SPX settling below 7500 or above 7600 at expiry: all four legs expire worthless, and you lose just the $40 net premium paid up front plus the $20 round-trip fee: a maximum loss of $60, no matter how far either way settlement goes.
Don't add the numbers wrong: the two vertical spreads on their own show maximum losses of $110 and $4,950, but those are worst cases for two separate tickets that can't happen at the same time; added together, the bull call spread half loses at most $110 while the bear call spread half makes $50 at the same settlement, for a net loss of $60 — the butterfly's maximum loss comes from the combined math, not from adding the two halves' maximum losses.
Margin locked is only $4,940: the bull call spread half is a debit spread and locks no margin; the bear call spread half locks (50 − 0.60) × 100 = $4,940; a $10,000 paper account can open it, but that $4,940 is just a deposit, not the maximum loss — the maximum loss is only $60.
Compare with a naked short call: one short 7550 call locks $151,000 of margin and a $10,000 account can't open it; even if it could, a 7600 settlement costs (7600 − 7550) × 100 − 100 + 1.30 = $4,901.30, and the loss keeps growing as settlement rises — no cap. The butterfly spends $40 of premium plus a $4,940 deposit to cap the loss at $60.
How to practice in the simulator
First, the plain truth: the simulator has no one-ticket butterfly — combos only support vertical spreads (two legs) and iron condors (four legs: one short and one long call plus one short and one long put); a butterfly with all four legs in calls gets rejected with "invalid combo: unknown strategy". Practice it as two vertical spreads: open a bull call spread first (buy the 7500 call, sell the 7550 call), then a bear call spread (sell the 7550 call, buy the 7600 call) — the middle 7550 is sold once in each ticket, and together they make exactly the butterfly's four legs.
Live-tested numbers (1 contract): the bull call spread half pays $100 net, costs a $5 opening fee, and locks no margin; the bear call spread half collects $60 net, costs a $5 opening fee, and locks $4,940 of margin; together they pay $40 net, cost $10 in opening fees, and lock $4,940 of margin; they use 2 of the 3 daily opens, leaving 1.
How to practice: pick three equally spaced strikes on the chain (say, 7500, 7550, 7600, 50 points apart), work out both breakevens first (net: 7500.60 and 7599.40), and write your exit plan before you submit; no new opens after 3:30 p.m. ET, and sort out positions before the 3:55 p.m. reminder — don't leave it to the close.
Spread fees: a vertical spread is charged $2.50 per leg per side — $5 to open one spread, $5 to close it; a butterfly split into two spreads costs $10 to open and $10 to close, $20 round trip — that $20 must come off when you work out the maximum profit and maximum loss; one spread open counts as 1 open against the daily limit, so two spreads are 2 opens.
Four mistakes beginners make
Adding the two halves' maximum losses to get the butterfly's: 110 + 4,950 = 5,060 is wrong — the two halves' worst cases happen in opposite directions and can't occur together; the butterfly's combined maximum loss is the $40 net premium paid plus the $20 round-trip fee, $60 in total.
Looking at the "$4,940 maximum profit" without seeing how narrow the landing zone is: net of fees, settlement must fall between 7500.60 and 7599.40 to profit, a window of only about 99 points; a 1% SPX day is 75 points on a 0DTE day, so one day can walk out of the window — ask "how likely is it to land inside" before "how much do I make".
Picking strikes too wide or too narrow: a 50-point-wide butterfly can make at most $4,940 and lose at most $60; a 10-point-wide one (say, 7540, 7550, 7560) can make at most (10 − net debit) × 100 − $20, with premium taking a bigger share and less room for error — decide the most you can lose first, then work backwards to how wide the wings can be.
Opening with no exit plan: a 1% SPX move is 75 points on a 0DTE day, and a small drift away from the 7550 peak sheds value fast — decide how much you'll lose and by when before you open, not at the close.
This simulator uses real CBOE option chains delayed about 15 minutes, starts with $10,000 in virtual funds, charges $0.65 per contract per side in fees, and shows P&L net of fees; short-sale margin is a simplified simulated rule at 20% of the strike, not a real broker margin. The simulator is for practicing orders, discipline, and review. It is not investment advice and promises no returns.