ROBUSTNESS REVIEW
Slippage, Holdout, And Sharpe Caveats
This page is intentionally skeptical. It highlights which rule families survived cost assumptions and the chronological holdout, while making clear why the Sharpe values should not be treated as production-grade evidence yet.
Cost model1 bp one-way
Holdout splitFinal 30%
Complete snapshots13,104
Rules tested272
SLIPPAGE SENSITIVITY
Net Return Across Cost Assumptions
The stronger reversal families remain positive through 2 bps, and several remain positive at 5 bps. Hybrid rules are more fragile because they turn over more frequently.
| Rule | 0 bp | 0.5 bp | 1 bp | 2 bp | 5 bp |
|---|---|---|---|---|---|
| Zero-gamma reversion, top quartile gap, negative gamma | +1,144 bps | +1,074 bps | +1,004 bps | +865 bps | +459 bps |
| Zero-gamma reversion, top quartile gap, amplifying regime | +1,144 bps | +1,074 bps | +1,004 bps | +865 bps | +459 bps |
| Wall-bias reversal, top quartile strength, amplifying regime | +1,065 bps | +988 bps | +912 bps | +761 bps | +322 bps |
| Wall-bias reversal, top 15 percent strength | +1,029 bps | +953 bps | +878 bps | +729 bps | +294 bps |
| Hybrid: fade extreme GEX, else zero-gamma reversion | +1,122 bps | +996 bps | +871 bps | +626 bps | -77 bps |
| Zero-gamma reversion, top quartile gap | +1,049 bps | +958 bps | +868 bps | +690 bps | +174 bps |
MONTHLY STABILITY
Selected Rule Families By Month
The monthly view is a guardrail against one-event overfitting. Consistency is improving, but several rules still rely heavily on specific regimes.
Zero-gamma reversion, top quartile gap, negative gamma
Zero-gamma reversion, top quartile gap, amplifying regime
Wall-bias reversal, top quartile strength, amplifying regime
Wall-bias reversal, top 15 percent strength
HOLDOUT
Best Robust Rules
- Zero-gamma reversion, top quartile gap, negative gamma: +467 bps holdout net, -41 bps holdout excess.
- Zero-gamma reversion, top 5 percent gap: +438 bps holdout net, +11 bps holdout excess.
- Zero-gamma reversion, top 5 percent gap, negative gamma: +426 bps holdout net, -19 bps holdout excess.
- Zero-gamma reversion, top half gap, negative gamma: +415 bps holdout net, -61 bps holdout excess.
CRITICAL READ
Do Not Oversell The Sharpe
- This is exploratory research, not a production trading model.
- The strategy search creates data-mining risk; best rules should be treated as hypotheses to validate, not final alpha.
- High Sharpe ratios can be inflated by intraday annualization, return clustering, and selective active windows.
- Buy-and-hold comparisons are active-window comparisons, not portfolio-level capital allocation results.
- The strongest evidence so far is that market-structure states contain measurable reversal behavior, not that a finished strategy is ready to sell.