Sortino & Other Ratios

The Sharpe ratio is the most famous risk-adjusted measure, but its biggest flaw, punishing upside and downside equally, has inspired a family of alternatives. Each one measures risk in a different way, and knowing what each captures, and ignores, makes you a sharper judge of performance. This lesson tours the main ones.

The Sortino ratio: only downside counts

The Sortino ratio fixes the most criticized flaw of the Sharpe ratio. The Sharpe ratio divides by total standard deviation, which treats a big gain as just as much risk as a big loss. But no investor complains about a sudden gain. The Sortino ratio replaces total volatility with downside deviation, which measures only the volatility of returns below some target, usually zero or the risk-free rate.

It rewards investments whose volatility is mostly to the upside and penalizes only the harmful, downside variability that investors actually fear.

The rest of the family

• The Treynor ratio divides excess return by beta instead of total volatility, so it adjusts for systematic market risk specifically. It is the right measure when an asset sits inside an already-diversified portfolio, where only its systematic risk matters.
• The Calmar ratio divides return by the maximum drawdown, the worst peak-to-trough loss. It measures return against the deepest pain an investor would have had to sit through, a very tangible form of risk covered in the drawdowns lesson.
• The information ratio divides a portfolio's return above its benchmark by the tracking error, the volatility of that excess return. It measures how consistently a manager beats a benchmark, a natural partner to the alpha discussion from earlier.

Each ratio defines risk differently

The common thread is simple: every one of these ratios uses the same excess-return numerator but a different definition of risk in the denominator. Sharpe uses total volatility, Sortino uses downside volatility, Treynor uses systematic risk, and Calmar uses maximum drawdown. None is universally right, because each captures a genuine but partial slice of risk. Which one matters most depends on the situation, and looking at several together gives a fuller, more honest picture than trusting any single number.

Every risk-adjusted ratio answers a different question, because every one defines risk differently. Read several, and always know what each one leaves out.

The shared blind spot

There is one weakness all of these ratios share, and it is the recurring theme of this curriculum. They are backward-looking, built from historical returns, and most of them struggle to capture tail risk, the rare and extreme losses that a calm return history simply has not shown yet. A strategy can look wonderful by every ratio here while carrying a hidden vulnerability to a catastrophe that has not happened during the sample. So treat these ratios as a panel of imperfect lenses, each lighting up one facet of risk, and never forget that all of them can be flattered by a history that has not met its worst case. That is exactly why stress testing exists later in this unit.

These ratios all measure risk from a return history. The next several lessons turn to the practical craft of managing risk directly, starting with the single biggest decision an investor makes: asset allocation.