Cash-secured puts, measured

What a cash-secured put is

You sell a put option and set aside enough cash to buy 100 shares at the strike if the buyer exercises. You keep the premium either way. If the stock stays above the strike through expiry, the put expires worthless and the trade is over. If it falls below, you may be assigned: you buy the shares at the strike, which is now above the market price.

So the trade pays a small, fixed amount most of the time and loses an open-ended amount occasionally. The win rate is high by construction. The question that matters is whether the premium pays for the losses, and that can only be answered by counting both.

What happens when a put is assigned

Assignment usually happens at expiry, when the put is in the money, though American-style options can be assigned early. Your cash buys the shares at the strike, and the loss on paper is the strike minus the share price, less the premium you collected.

Three common responses: hold the shares and sell covered calls against them (the wheel), sell the shares and take the loss, or avoid assignment in the first place by buying the put back or rolling it before expiry. Buying back a threatened put costs real money: in our records a guard buyback cost roughly three times the premium collected, so one defensive close can erase several winning trades.

What the history says

We replayed every cash-secured put a set of rules would have sold, with entries priced at the real bid a seller receives rather than the midpoint. Over about 1,100 resolved paper trades, closing at half the premium when possible and otherwise defending the position, the average trade earned about +0.8% on the cash set aside, with an 86% win rate. The average loss was about twice the average win.

That works out to roughly 5% to 7% a year on collateral, and the cash itself earns around 4% sitting in Treasury bills. On the index funds the strategy behaved like a Treasury bill with a tail. On leveraged ETFs it earned a little more and carried single-trade losses of 9% to 26% of the collateral.

On ten years of QQQ and SPY history the plain short put was positive but small: about +0.4% a trade on collateral, with an interval that excludes zero. Real, and modest.

What actually decides the result

Size, more than strike selection. A position that is small relative to the account survives the bad month; a large one turns a normal loss into a drawdown that takes a year to recover. Several puts on tickers that move together are one bet, not several.

The edge itself is the variance risk premium: the gap between the volatility options price in and the volatility that actually shows up. In our data it was steady for months and then fell to about zero for a month. Nothing we could measure at entry predicted which months those would be, which is why sizing, not a clever filter, is the defense.