February 5, 2018. The VIX closes at 17.31 on a Friday. By Monday it prints 37.32 — a 116% single-session spike. Traders running VIX-based position sizing models had calculated "normal" exposure. By Tuesday, those positions were 2.3R underwater before a single stop-loss trigger fired. The model said safe. The market said otherwise.

This is the volatility-of-volatility problem. VIX measures implied volatility of the S&P 500 over 30 days. But VIX itself is volatile — and that second-order volatility, often measured as VVIX, routinely spikes before, during, and after the events that VIX-based models are supposed to protect against. When VVIX exceeds 120, historical data shows VIX becomes structurally unreliable as a position-sizing input.

CONCEPTVVIX measures volatility of VIX itself — when VVIX spikes, your VIX-based risk model is flying blind.
WARNINGA VIX reading of 15 feels calm — but if VVIX is 130, a move to VIX 40 is statistically plausible within 48 hours.
KEY IDEARegime shifts don't announce themselves in VIX — they arrive in the shape of VVIX divergence first.

The core mechanical failure is mean-reversion assumption. Standard VIX risk models assume volatility reverts to a long-run mean — roughly 19-20 for the VIX index. Position size is then calculated as: Risk % ÷ (ATR × VIX scalar). During low-VIX regimes, the scalar shrinks, position sizes expand, and total portfolio exposure balloons precisely when fragility is highest. AQR Capital's research on volatility targeting identifies this as a structural leverage trap.

Portfolio Drawdown: Standard VIX Model vs VVIX-Adjusted0%-15%-28%-40%VIX-only modelVVIX-adjusted→ Regime shift eventVVIX > 120

A rules-based adjustment traders apply is a VVIX circuit breaker: if VVIX exceeds 110, maximum position size is capped at 50% of the VIX-model output, regardless of the VIX reading itself. If VVIX exceeds 130, all new entries halt. This isn't discretion — it is a hard rule baked into the algorithm before the session opens. The drawdown reduction across historical regime shifts is material. Understanding the mechanics behind VIX construction and its limitations, the statistical basis of volatility risk premium behaviour during stress, and AQR's documented research on volatility targeting strategies gives traders the framework to build these guardrails properly.

The bridge doesn't fall because engineers account for resonance, not just static load. Your risk model needs to account for the volatility of your volatility input — or it only works in the conditions that need it least.

This content is for educational purposes only and does not constitute financial product advice. Past performance is not indicative of future results. Profit Logic Ltd (ACN 688 669 936) accepts no responsibility for errors or omissions in this content or anywhere on this website. Always seek advice from a licensed financial adviser before making investment decisions.