Sell options on SPY when they are expensive and let time do the work. When implied volatility rank is high, premiums are rich; time decay erodes them every day, and implied volatility tends to fall back. The buyers on the other side are mostly hedgers paying for protection, and the seller is paid to absorb the moves they fear.
volatility risk premium / theta
SPY "iron condor" selling
The idea
How it was tested
Pre-registered on 2026-08-23. The signal: 252-day implied volatility rank above 60, with SPY within ±0.5% of its 20-day average. The structure: one short call at 102% and one short put at 98% of spot, 30-45 days to expiry. Exit at 50% of max profit, 5 days before expiry, or when the loss reaches 50% of the credit.
It was called an iron condor and described as defined risk, but the specified structure has no protective wings. It is a short strangle.
Where it died
It was run once, on 2026-08-23, and reported 142 trades, +10.70% per trade and 22.8% a year. Those prices came from a model: a flat volatility surface, no assignment, fills at mid. The flat surface removes exactly the skew that makes out-of-the-money puts expensive, which is where a short put earns or loses its money.
The alpha gate was written down as "TBD (demo will validate)". In the demo, the executor computed credit and cost with Black-Scholes and never sent an order. On 2026-10-17 the strategy was declared unverifiable, and the verdict became void.
The record
Thesis
Sell out-of-the-money calls and puts on SPY when implied volatility rank is above 60, and let time decay and volatility mean reversion pay the seller. Close at 50% of max profit, 5 days before expiry, or at a stop.
Specification
- signal
- IV rank (252 days) > 60 and SPY within ±0.5% of its 20-day average
- structure
- short 1 call at 102% of spot and short 1 put at 98% of spot, 30-45 days to expiry
- exit
- 50% of max profit, 5 days before expiry, or loss of 50% of the credit
- caveat
- Despite the name and the "defined risk" claim, the specified structure has only the two short legs and no protective wings — it is a short strangle.
Gates
Pre-registrationpass
Was the hypothesis, its parameters and its pass/fail rule written down before any data was looked at?
Raw significance (legacy)not evaluated
Are the strategy's raw returns statistically different from zero?
trades142avg trade pnl pct+10.70annual return pct+22.80
Reported by a backtest that assumed a flat volatility surface, no assignment and fills at mid. Not a valid measurement — see below.
Gate 7 — Jensen alphanot evaluated
After regressing out market exposure, is there alpha left?
Listed as "TBD (demo will validate)" and never measured.
Why it died
The only backtest priced options from a model, not from market data: the lab has no historical option chains, so a flat-volatility simplification stood in for the real skew that a short put lives or dies by. Gates 7 and 8 were left for the demo — where the executor computed credit and cost with Black-Scholes and sent no orders. Nothing about this strategy was ever measured.
What it taught
"Defined risk" has to be checked against the legs actually specified, not the strategy's name. And deferring the alpha gate to a demo that cannot place orders is deferring it forever.
Verdict history
- 2026-08-23Pre-registered and run once on modelled prices: 142 trades, +10.70% per trade, Gates 7/8 left open
- 2026-10-17Declared unverifiable: no historical option chains; demo executor never sent orders Void