⚠ The edge is NOT consistent — it helped in 5 of 8 years and hurt in 3

This row is the honest one. It asks: after a buy signal, did the fund actually rise more over the next 21 days than it would have on any random day? Positive means the timing added something; negative means you'd have done better buying blind. It swings from -3.02 in its worst year to +1.52 in its best.

What this means in plain terms: most of what this strategy earns comes from the market drifting upward over three weeks (1.84% on average, for buying anything on any day) multiplied by the leverage an option gives you — not from the dip timing being clever. The timing contributes about 0.18 percentage points on average, and that contribution is unreliable. Treat this as a disciplined way to take leveraged long exposure with defined size and defined exits, not as a system that predicts which dips will bounce. In a year when the market falls, expect it to lose money.

1. Stop — watch the STOCK price, not the option price

The Stop ▼ column gives a price on the underlying stock or ETF, not on the option. If the stock closes below that level, the reason for the trade is gone — sell the option at the market next session, whatever it's worth.

Why not a stop order on the option itself? Because it doesn't work, and this was measured rather than assumed. Across 8 years of real bars, the worst single trade lost 100% of the premium — and it lost exactly 100% whether the stop was set at −30%, −65%, on the underlying, or not set at all. Options gap overnight; a stop order can't fill at a price the market skipped past. Meanwhile a tight stop on the option reliably sells you out of good trades over ordinary price noise: the −30% option stop turned a +3.6% average trade into −0.2%. So a stop on the option costs real money and buys no protection.

What actually protects you is the size of the bet — see the Size column, which is set so that losing the entire premium costs a small, fixed slice of your capital. That's the real safety net, and it's the one professionals rely on for long options.

2. Take-Profit — two ways to run it

Simple: SELL TO CLOSE, LIMIT order at the Target price shown. GTC. Fires automatically once reached, locks in the full computed gain.

Let-it-run version: instead of a flat limit, use a GTC trailing stop-limit order. Set the trail % so it never gives back more than you're comfortable with off the peak price, while letting the position keep running if it moves further than the original target. See the full math (which trail % protects which profit floor) in Exit Strategy in the repo — the short version: an 8% trail won't lock in a floor of at least +10% profit until the position first reaches about +20% gain; a 4% trail locks that floor in earlier (around +15% gain) but is more likely to trigger early on normal day-to-day option price noise. Tighter trail = protects sooner but exits more easily on noise; looser trail = lets it run more but the floor guarantee kicks in later.