Lesson

Stop-losses on short puts, replayed: a 2x stop earned less than T-bills

Buying back a short put when it doubles is the most common advice for limiting losses. On real bid and ask quotes since 2022, a 2x stop on SPY and QQQ puts cut the worst trade by more than half and left the strategy earning less than Treasury bills.

The conclusion first

We sold a 25-delta cash-secured put on QQQ or SPY once a week from April 2022 to October 2026, opened 30 to 45 days out, 574 trades in all. Entries filled at the real closing bid. A stop bought the put back at the real closing ask on the first day its price closed at 2x, or 3x, the credit received.

With a 2x stop the trades earned +2.5% a year on the cash tied up, and +3.2% with a 3x stop. A 13-week Treasury bill averaged about 4.15% over the same years. With no stop, holding to expiry earned +5.6%, and closing early at half the credit earned +8.4%. Adding a 3x stop to that half-credit rule cut it to +4.6%, barely above T-bills.

The numbers

QQQ and SPY, 25-delta puts, 574 trades over 55 expiry months, average credit 1.12% of the cash set aside. Held to expiry: won 89%, +0.59% a trade (95% interval +0.26% to +0.88%), 39 days in the trade on average, +5.6% a year, worst trade -13.5%. Closed at half the credit: won 95%, +0.37% a trade, 15.5 days, +8.4% a year, worst trade -13.5%.

Stop at 2x the credit: won 68%, +0.25% a trade (-0.03% to +0.51%), 31 days, +2.5% a year, worst trade -5.1%. Stop at 3x: won 79%, +0.32% a trade (-0.02% to +0.60%), 35 days, +3.2% a year, worst trade -7.7%. Half the credit plus a 3x stop: won 86%, 13 days, +4.6% a year, worst trade -6.1%.

By fund the order is the same. SPY: +2.2% a year with a 2x stop, +2.5% with 3x, against +6.8% closing at half the credit. QQQ: +2.8% and +4.0%, against +10.7%. Across the leveraged ETFs we also tested the stops did no better.

The per-year figures are the profit per dollar per day the cash was tied up, times 365, so trades that close early get credit for freeing cash. They assume the freed cash goes straight into the next trade, and leave out any interest your broker pays on the cash.

What the stop actually did

It did what it promises: the worst single trade went from -13.5% of the cash to -5.1%. The price was paid everywhere else. The 2x stop fired on 188 trades, about one in three, and 127 of those, two in three, would have finished with a profit if held to expiry. Losing trades went from 61 to 182.

A 25-delta put can double in price on an ordinary pullback, long before the stock is anywhere near the strike. The stop buys it back at the ask at exactly that moment, then misses the recovery. The rare trade that would have lost 13% is not common enough to pay for all the small losses the stop books on the way.

It barely helped the worst month

The worst calendar month with a 2x stop was -2.6% of the cash, against -3.6% closing at half the credit with no stop. Bad months come from many positions losing a little at the same time, and a per-trade stop does not change that. What limits a bad month is how much of the account is in the trade, and how many of the positions are really the same bet. See position sizing for option sellers.

Spreads tell the same story

On put credit spreads every stop we tested also earned less than holding, because the large losses arrive as gaps that a daily stop cannot get out in front of. See the lesson on stop-losses for credit spreads.

What it does not show

One window, starting in 2022 with our quote history: one bear market, two sharp dips and a long recovery. Stops trigger and fill on daily closing quotes; a stop order resting intraday fills at a different price, and in a fast market usually a worse one. Commissions are left out, and each stop pays one more. None of these rules beat simply owning SPY over the window, which returned about 13% a year.

If you use a stop on short puts, score your own closed trades with and without it, against the T-bill rate on the same cash. Paper results must use your real exit rule explains why the exit you assume changes the answer.

Historical replays are not a forecast. Options can lose more than the premium collected. This is education, not a recommendation to trade anything.

Test your own rule on history, free