Volatility Surface

IV varies because real markets have features the Black-Scholes model doesn't capture: crash risk (makes puts expensive), leverage effect (falling prices increase volatility), event uncertainty (specific dates have elevated IV), and supply/demand imbalances. The surface shape reflects these real-world factors.

Surface trading typically requires larger accounts due to multi-leg positions and margin requirements. However, you can start by understanding and observing surface behavior before trading. Simple skew trades (risk reversals) or calendars are accessible starting points.

Broker platforms like Interactive Brokers have volatility surface visualization tools. You can also build your own by collecting IV data across strikes and expirations. For XJO, the XVI index is derived from the surface and gives a quick summary of overall vol.

No, skew varies by underlying. Equity indices (XJO) typically have steep skew due to hedging demand. Individual stocks vary - some have less skew. Commodities often have different patterns. Understanding the 'normal' skew for each asset is important.

Term structure inverts (backwardation) when near-term uncertainty spikes - market crashes, imminent events, or sudden fear. Traders buy near-term protection, pushing front-month IV above back-month. This typically reverts as the immediate fear passes.

Measure skew daily (25Δ put IV - 25Δ call IV) for 60+ days. Rank current skew within that history. Percentile = (rank / total observations) × 100. 90th percentile means current skew is higher than 90% of historical readings.

Selling vol (e.g., iron condor) profits if overall IV drops. Selling skew (e.g., risk reversal) profits if the put-call IV differential narrows, regardless of overall vol direction. You can be short skew and long vol, or vice versa.

Risk reversals have directional exposure. Delta-hedge if you want pure skew exposure without directional bet. However, hedging costs money (slippage, carry). Many traders accept some delta exposure or hedge partially.

Highly variable. Skew can stay extreme for weeks during ongoing stress. Mean reversion often takes 10-30 days after the stress catalyst passes. Use time limits (e.g., exit after 21 days) to avoid holding indefinitely.

Calendar spreads at ATM are nearly delta-neutral but have second-order directional exposure (lose if underlying moves far). You can also trade a straddle calendar (calendar on both puts and calls) to reduce directional sensitivity further.

Vanna is highest for OTM options. A risk reversal has vanna exposure (different vanna for put vs call). To isolate vanna, combine positions that are vega-neutral and delta-neutral but have asymmetric OTM exposure. This requires careful Greek calculation.

Local volatility (Dupire model) is a deterministic volatility function that exactly reproduces the observed surface. It answers: 'What volatility process would generate this surface?' Local vol is used for consistent pricing of exotics and for surface interpolation.

Check bid-ask spreads - wide spreads often cause apparent mispricings. Verify multiple data sources. Check if 'arbitrage' exceeds transaction costs. Real violations are usually temporary and small; persistent large violations are usually data issues.

You need historical option data (expensive). Calculate surface metrics historically, generate signals, simulate trades, and track P&L. Account for transaction costs and slippage. Use walk-forward testing to avoid overfitting.

If you have skew trades on correlated underlyings (e.g., XJO and BHP), they'll tend to move together. A market crash will steepen skew on both simultaneously. Consider this correlation when sizing. Diversify across less-correlated names if possible.