Sterling and Burke Ratios

The Sharpe ratio divides return by volatility, but volatility punishes upside swings just as much as downside ones, which feels wrong for judging a trading strategy — nobody minds a good week. The Sterling and Burke ratios both fix this by replacing volatility with a measure built purely from drawdowns, the peak-to-trough declines an equity curve actually experiences.

The Sterling ratio divides annualized return by the average of the largest few drawdowns (often the worst three to five, plus a small adjustment), rather than by the single maximum drawdown. Using an average of several bad drawdowns instead of just the worst one makes the ratio less sensitive to one freak event and more representative of how the strategy behaves during its typically bad periods.

The Burke ratio goes further: instead of averaging the worst drawdowns linearly, it takes the square root of the sum of their squares. Squaring before summing means large drawdowns are penalized disproportionately more than small ones — a single 30% drawdown hurts the Burke ratio far more than three separate 10% drawdowns would, even though the linear total is the same, which better matches how painful large losses actually feel to an investor.

A quick illustration: a strategy with the same annualized return and the same set of historical drawdowns will score differently under each measure — Sterling treats a few decent-sized drawdowns roughly in proportion to their size, while Burke's squaring pushes the ratio down harder if any one of those drawdowns is unusually deep. That makes Burke the stricter of the two whenever drawdown sizes are uneven.