The spread is the commission nobody itemizes

Every option has two prices. The bid is what a buyer will pay right now; the ask is what a seller will accept. When you sell an option, you're paid something near the bid. When you buy it back, you pay something near the ask. The distance between them is the spread, and it's a cost you pay on the way in and again on the way out, with no line for it anywhere on the statement.

The size of it

A put quoted $1.00 bid, $1.10 ask has a $0.10 spread. Sell at the bid, buy back later at the ask, and you've paid $0.10 per share for the round trip: $10 per contract, 10% of the premium, before the stock has moved. On a $0.40 option with the same $0.10 spread, the cost is 25%. On a thin strike quoted $0.80 by $1.30, a round trip costs $0.50 against a mid-price of $1.05, nearly half the premium. The right measure is the spread as a percentage of the premium, because that's the fraction of your expected edge the market charges to let you in and out.

Where it's wide

Spreads are narrow where there's competition. Index products like SPX and ETFs like SPY and QQQ trade in pennies at most strikes. Large-cap stocks are close behind. Small caps, far-out expirations, strikes far from the money, and anything whose options were listed last week are wide. Open interest and daily volume are the quick checks. A strike with no open interest can look fine right up until you try to leave it.

The mid, and limit orders

The mid is halfway between bid and ask. Market makers will often fill a limit order at or near the mid, especially in liquid names, because the quoted spread is wider than what they'll actually accept. Selling at the mid instead of the bid on a $1.00 by $1.10 quote saves $5 a contract, half the round-trip cost. The habit is simple: never send a market order in options. Place a limit at the mid, and if it doesn't fill within a minute or two, move a cent or two toward the other side. A fill you have to chase is telling you something about the liquidity.

Twice, and four times

Opening and closing is two crossings. A roll is four: buy back the old option, sell the new one, each leg with its own spread. Sending the roll as a single combination order gets you a package price, which usually costs less than two separate crossings but never nothing. Multi-leg positions pay per leg, so an iron condor pays four spreads to open and four to close. In a wide-spread name that can consume the entire credit.

Letting it expire

An option that expires worthless costs nothing to exit: no spread on the way out. That's one real reason sellers hold small remaining credits to expiration rather than buying back a $0.05 option quoted $0.05 by $0.15. Weigh it against the gamma of a short option with two days left. The spread you save is small, and the move you're exposed to isn't.

Where it shows up

Trades that seem to lose a little on average for no reason usually have the spread in them. A log that records the mid at entry and the fill at exit will show it. Over a year of trading a name with 10% round-trip spreads, the cost is on the order of the whole expected return of the strategy. The spread isn't a detail. In illiquid names it's the decision.