Why drawdown beats Sharpe as a risk metric for retail-facing managers

The Sharpe ratio is the industry-standard measure of risk-adjusted return. Excess return divided by standard deviation. Elegant. Comparable across managers. Beloved of consultants for the same reason that a single number is beloved of any committee that has to make a decision.

For a retail-facing manager, Sharpe misses the only risk metric that actually predicts client behaviour: peak-to-trough drawdown.

Two strategies can post identical Sharpe ratios, identical means, and identical standard deviations, and still produce wildly different client outcomes. The difference is the path: the shape of the equity curve as it gets to that final number. A strategy that delivers a steady 1% per month for twelve months is mathematically indistinguishable from a strategy that delivers 30% in January, sits at a 25% drawdown through July, then grinds back to flat. On paper, the same Sharpe. In practice, only one of them is a business.

Why path matters more than averages

Three reasons drawdown is the more honest risk metric for any manager who serves human investors rather than institutional liability matchers.

First, investors experience the path, not the average. Statistics treat each return observation as a draw from a distribution. Investors live the sequence. A 22% drawdown in month four feels nothing like the same dollar loss spread evenly across the year, even though both contribute identically to the standard-deviation term in the denominator of the Sharpe ratio.

Second, redemption behaviour clusters at troughs. The data on this is unambiguous: investors pull capital near drawdown lows. This converts a temporary mark-to-market loss into a permanent realised loss. Any manager who has actually managed retail money has watched this happen. Most have learned that the way to prevent it is to set expectations honestly about drawdowns before they occur, not to claim a smooth ride that will never be delivered.

Third, recovery time matters more than recovery probability. A 30% drawdown that recovers in six months is psychologically very different from a 20% drawdown that takes three years to recover. The shallower drawdown is technically less severe by every standard textbook measure. It is also far more likely to cause the investor to lose patience and redeem before recovery completes.

A drawdown is a number on a screen. A recovery is a relationship.

How we report

Every Bostock strategy report we share with clients and prospective clients includes, on the same page, all of the following: maximum historical drawdown, average drawdown, drawdown duration distribution, time to new equity high, and the worst rolling twelve-month return. The Sharpe ratio is there too. It does not lead. It belongs in the appendix of any honest performance pack.

This is not a criticism of statistical rigour. It is a recognition that we manage money for human beings, and human beings do not invest in standard deviation. They invest in the path.

The next time a prospective manager hands you a performance deck whose lead metric is a Sharpe ratio, ask for the drawdown table. The answer will tell you almost everything.