When someone tells you an investment “made 20%,” they’ve only told you half the story. The half they left out is how much risk they took to get there. A 20% return from a stable strategy and a 20% return from a rollercoaster are not the same thing — even though the final number looks identical.
The Sharpe ratio fixes this. It measures return per unit of risk. In plain terms: how much reward did you actually get for the stress you endured?
How to read it
A higher Sharpe ratio is better — it means more return for less turbulence. As a rough guide: below 1 is mediocre, around 1 to 2 is solid, and above 2 is genuinely good. Anything dramatically higher than that on real, out-of-sample data deserves suspicion, not excitement — it usually means something is being measured wrong or the test was rigged in hindsight.
In our Funding Rate Arbitrage postmortem, the strategy showed a Sharpe of 6.94 on historical data — spectacular on paper. On unseen data it dropped to -0.71. That collapse, not the headline number, is the real story.
The catch
A high Sharpe ratio on past data means very little on its own. Anyone can find a strategy that looks brilliant when they already know what happened. The number only becomes trustworthy when it survives on data the strategy has never seen — which is exactly what the next lesson is about.
The Sharpe ratio answers one question: was the return worth the risk? A great return with terrifying swings can have a worse Sharpe than a modest, steady one.
- 1The Sharpe ratio measures return relative to risk, not return alone.
- 2Higher is better, but suspiciously high on past data is a red flag, not a win.
- 3It only matters when measured on data the strategy hasn’t seen before.
Sharpe — appears in every Strategy Graveyard postmortem.