Sharpe Ratio
Excess return per unit of volatility. The standard way to compare two strategies - rewards smooth equity curves, but doesn’t distinguish good volatility from bad.
The Sharpe ratiomeasures how much excess return a strategy earns per unit of volatility. Two systems with the same annual return aren’t equally good if one of them gets there with calm monthly gains and the other with violent swings. Sharpe is the standard way to tell them apart.
Formula: Sharpe = (Strategy Return − Risk-Free Rate) ÷ Standard Deviation of Returns. The risk-free rate in India is usually the 10-year G-Sec yield (around 7%). Standard deviation is computed on the period-by-period returns - daily, weekly, or monthly - and then annualized.
Worked example - annualized Sharpe
- Annual return
- 30%
- Risk-free rate (10Y G-Sec)
- 7%
- Excess return
- 23%
- Annualized std dev of monthly returns
- 15%
- 23% ÷ 15%
- 1.53
Sharpe ratio
1.53
Rough scale: < 1 is weak, 1–2 is good, 2–3 is excellent, 3+ is rare and usually means short backtest or a strategy that hasn’t seen a real drawdown yet.
What Sharpe rewards (and what it doesn’t)
Sharpe rewards smoothness. A system that makes 1% every month gets a stellar Sharpe. A system that makes 30% in two months and breaks even the rest of the year gets a mediocre one - even if the total return is identical. For a discretionary trader, this matters because smooth equity curves are easier to scale with leverage and easier to hold during losing streaks.
The fair critique
Sharpe treats upside volatility and downside volatility the same. A strategy that suddenly makes a +50% month gets penalizedon Sharpe - even though most traders would happily take it. That’s why some funds report Sortino ratioinstead, which only penalizes downside deviation. Sharpe is also unreliable on small samples (<3 years of data) and on strategies with fat-tailed return distributions (most options strategies).
When it’s most useful
Comparing two strategies on the same instrument, same period, same return profile. Less useful for comparing a quiet pair-trader to a volatility-arbitrage shop - they don’t live on the same return distribution. As a single number for a single retail trader: it’s a fine sanity check, not a verdict.
Related terms
Sharpe is one lens. Use several.
Find My Edge shows expectancy, profit factor, drawdown, and equity curves side-by-side - so no single number can mislead you. Free, no card.
Start journaling free