Backtest
Backtest — price signal
What happened to price after the watcher's entry signal fired? This page answers that question from real daily closes, and nothing else.
This is a price-signal backtest, not a trade P&L backtest. Option premiums were not recorded historically, so no premium, trade win rate, or return on a put trade is claimed here. The figures below describe price movement after a signal, not the outcome of selling a put.
Results by ticker and horizon
Signal counts differ by horizon because signals near the end of the sample have no complete 45-, 60-, or 75-day forward window yet; incomplete horizons are excluded, not assumed.
| Ticker | Horizon | Signals | Median forward return | % positive | Median worst drawdown | % staying above −10% |
|---|---|---|---|---|---|---|
| SPY | 45 days | 78 | 3.68% | 67.9% | -3.52% | 88.5% |
| SPY | 60 days | 76 | 4.90% | 77.6% | -4.69% | 86.8% |
| SPY | 75 days | 76 | 5.21% | 90.8% | -4.69% | 78.9% |
| QQQ | 45 days | 62 | 6.43% | 71.0% | -3.93% | 90.3% |
| QQQ | 60 days | 62 | 6.81% | 87.1% | -4.63% | 87.1% |
| QQQ | 75 days | 62 | 6.59% | 87.1% | -5.17% | 85.5% |
| IWM | 45 days | 66 | 2.95% | 59.1% | -3.64% | 86.4% |
| IWM | 60 days | 62 | 5.36% | 61.3% | -4.89% | 75.8% |
| IWM | 75 days | 61 | 6.55% | 63.9% | -5.17% | 73.8% |
| DIA | 45 days | 82 | 1.54% | 68.3% | -3.21% | 84.1% |
| DIA | 60 days | 78 | 2.16% | 67.9% | -3.67% | 83.3% |
| DIA | 75 days | 78 | 4.42% | 80.8% | -3.91% | 79.5% |
Sample: 2021-11-15 to 2026-11-13, 1,255 trading days per ticker. Signal totals: SPY 86, QQQ 73, IWM 70, DIA 86. "Median worst drawdown" is the median of each signal's minimum close return over the horizon. "% staying above −10%" is the share of signals whose minimum close return stayed at or above −10%.
How it is computed
- Signal. A day counts as a signal when the day's low touches or crosses below the lower Bollinger Band (20-day mean minus 2 times the population standard deviation) and the day's close is above the 200-day simple moving average. This is the same definition as the live screener.
- Data. Daily closes (and lows, for the touch test) from yfinance, the same source as the bag-hold closes. Closes are not CBOE data; CBOE supplies put/call ratios in this pipeline, not daily closes.
- Forward outcomes. For each signal, the forward return is the close at 45, 60, or 75 trading days later divided by the signal-day close, minus one. The worst drawdown is the lowest close over the horizon divided by the signal-day close, minus one. Closes only — no intraday prices, no option prices.
- The −10% proxy. The watcher's short puts sit about 10% out of the money, so "% staying above −10%" uses the minimum close return staying at or above −10% as a proxy for price staying above the short strike region. It is a proxy, not assignment: assignment depends on the price at expiry, not on the lowest close along the way, and no premium is involved.
- What is not claimed. No premium collected, no put profit or loss, no trade win rate, and no annualised return. A premium backtest needs recorded option prices, which only began accumulating recently (the SPY surface recorder holds days, not years, of data).
Education only. Past price behaviour after a signal does not predict future outcomes.