Credit Spreads: Why a 70% Win Rate Is Still a Losing Strategy

A credit spread is the trade most people move to after they get tired of buying options that expire worthless. You sell one option, buy a further out-of-the-money option in the same expiration for protection, and collect the difference as a credit. Time decay works for you, the maximum loss is capped, and you profit if the underlying does nothing.

It is also the structure where traders most reliably misjudge risk, because the win rate is high enough to feel safe long before the math says it is.

The Two Constructions

Bull put spread. Sell a put at a strike below current price, buy a further out-of-the-money put. You keep the full credit if the stock stays above your short strike. This is the bullish-to-neutral version.

Bear call spread. Sell a call above current price, buy a further out call. You keep the credit if the stock stays below your short strike. Bearish to neutral.

The long leg is not there to make money. It exists to define the maximum loss and to reduce the margin requirement, which is what makes the position sizeable in a normal account rather than requiring the collateral of a naked short.

The Numbers That Define the Trade

Take a stock at $100. You sell the $95 put for $1.80 and buy the $90 put for $0.60. Net credit is $1.20, or $120 per spread.

Maximum profit is the credit: $120, achieved if the stock closes above $95.

Maximum loss is the width of the strikes minus the credit: ($95 − $90) − $1.20 = $3.80, or $380 per spread, if the stock closes below $90.

Breakeven is the short strike minus the credit: $93.80.

So the trade risks $380 to make $120. That ratio is the part people gloss over. You need to win roughly 76% of the time just to break even before commissions. A 70% win rate — which sounds excellent — is a losing strategy at this risk/reward.

Why the High Win Rate Is a Trap

Credit spreads produce a long string of small wins punctuated by occasional full losses. Nine winners at $120 is $1,080. One loser at $380 knocks it to $700. That still works. The problem is what happens when a trader, encouraged by the win streak, doubles size — or holds ten correlated spreads that all break at once during a market-wide selloff.

Correlation is the specific danger. Ten bull put spreads across ten different tech names is not ten independent bets. It is one bet on the Nasdaq, expressed ten times. The strategy's risk model assumes independence it does not have.

Sizing rule that survives contact with reality: assume every open spread takes its maximum loss simultaneously, and check whether the account survives that. If it does not, the book is too large regardless of how good the individual probabilities look.

Choosing Strikes

Short strike delta is the standard control. A 30-delta short strike is roughly a 70% chance of expiring out of the money and pays a fat credit. A 15-delta short strike is roughly 85% but pays much less, and the risk/reward gets worse as the probability improves — that is the tradeoff the pricing enforces, and there is no strike selection that escapes it.

Width matters separately. A $5-wide spread and a $10-wide spread at the same short strike have the same probability of profit but different absolute risk. Narrower spreads mean smaller max loss per contract and a worse credit-to-width ratio.

Placing the short strike outside one expected move gives a probability-based anchor rather than an arbitrary one. Placing it just outside a significant gamma level gives a structural one. Both are better than picking the strike where the credit looked appealing.

Implied Volatility Is the Entry Condition

Credit spreads are short vega. You want to sell them when implied volatility is elevated relative to the stock's own history, because you collect more premium for identical risk and you benefit if IV contracts while you hold.

Selling spreads into low IV is the most common structural mistake. The credit is thin, the risk/reward is worse than usual, and any volatility expansion hurts the position independently of direction. IV percentile above roughly 50 is a reasonable filter, and it removes a large share of the trades that were never worth taking.

Managing the Position

The common convention is to close at 50% of maximum profit rather than holding to expiration. You give up half the credit and remove the gamma risk of the final week, when a modest adverse move can turn a comfortable winner into a full loser very quickly.

On the loss side, the decision worth making in advance is whether you roll or take it. Rolling out in time for a further credit works when the thesis is intact and the move looks like noise. Rolling repeatedly on a stock in a genuine downtrend is how a defined-risk strategy quietly turns into an undefined one.

The options calculator shows the full payoff profile, breakeven, and probability estimates for any spread before you enter it, which is the cheapest possible way to discover that a trade you liked risks $380 to make $120.