Volatility Arbitrage and Where the Edge Comes From

A volatility arbitrage edge comes from a forecasting advantage, a structural liquidity edge, or the variance risk premium, and each fails differently.

Volatility arbitrage is the practice of buying or selling options because their implied volatility looks mispriced relative to a trader’s own forecast of realised volatility, then delta hedging the position so the resulting profit or loss comes from the gap between the two rather than from the direction of the underlying. The edge, when it exists, comes from a forecasting advantage, a structural pricing advantage, or the collection of a risk premium the market pays reliably for bearing tail risk.

A position built from an option and its delta hedge earns or loses money each period from the difference between the gamma-weighted realised move in the underlying and the implied variance the option was priced with, the same relationship worked through from the hedging side in gamma scalping and delta-neutral hedging. A trader who buys an option because implied volatility looks cheap relative to a forecast is long gamma and profits from rehedging into a move that turns out larger than the option’s implied volatility priced in; a trader who sells because implied volatility looks rich is short gamma and profits when the underlying moves less than that implied level over the life of the…

One source of edge is a genuinely better short-horizon volatility forecast, built from models that capture volatility clustering, the tendency for large moves to follow large moves, and calibrated more tightly than the level the options market happens to be pricing at a given moment. A second, narrower source is informational: a desk that has traded a specific name through many prior earnings cycles may hold a more precise view of the likely realised move around the next release than the implied volatility embedded in the options expiring right after it, even without any non-public information at all.

A different kind of edge does not depend on forecasting anything better than the market. Market-making desks earn a persistent return from providing liquidity itself, capturing the bid-ask spread on the options they trade many times over, closer to a transaction-based edge than a volatility view. Systematically selling volatility on broad index options and collecting the variance risk premium, the tendency for implied volatility to sit above subsequently realised volatility on average, described in implied volatility compared with realised volatility, is a structural edge in the sense that it does not require forecasting any single event correctly, only holding a diversified…

A single delta-hedged position is a bet on one realised outcome, and the edge in either the forecasting or the structural version of the trade only becomes reliable once it is spread across many largely uncorrelated positions, different names, different expiries and different underlyings, so the book’s overall result depends on an average across many independent bets rather than the outcome of any one of them. Vega and gamma limits at the position and book level exist specifically to keep any single name from dominating that average.