Sharpe Ratio

The Sharpe ratio, developed by Nobel laureate William F. Sharpe in 1966, is the most widely used measure of risk-adjusted return in finance. It answers a deceptively simple question: how much return are you earning per unit of risk?

Raw returns alone are misleading. A strategy that returns 20% annually sounds great — until you learn that it has 40% annual volatility and experienced a 60% drawdown. The Sharpe ratio normalizes returns by risk (measured as standard deviation), making it possible to compare strategies that have very different return and risk profiles.

The concept is intuitive: if two strategies earn the same return but one has half the volatility, the less volatile strategy has a higher Sharpe ratio — it's generating the same return with less risk. In quantitative finance, the Sharpe ratio is the first thing evaluated when assessing any strategy, and it is closely linked to optimal sizing via the Kelly criterion.

The Sharpe ratio formula is:

Sharpe = (R_p - R_f) / σ_p

Where:

• R_p = the portfolio's return (annualized)
• R_f = the risk-free rate (typically the yield on short-term Treasury bills)
• σ_p = the standard deviation of the portfolio's returns (annualized)

The numerator (R_p - R_f) is called the excess return — the return above what you'd earn with zero risk. The denominator is the volatility — a measure of how much the returns fluctuate.

Annualization: If you're computing the Sharpe from daily returns, multiply the daily Sharpe by √252 (there are approximately 252 trading days per year) to annualize it. This assumes returns are independently distributed across days.

Interpretation guidelines:

• Sharpe < 0: The strategy underperforms the risk-free rate — it's losing money on a risk-adjusted basis.
• Sharpe 0-1: Mediocre. Acceptable for passive investments, not for active quant strategies.
• Sharpe 1-2: Good. Many successful hedge fund strategies fall in this range.
• Sharpe 2-3: Very good. A strong systematic strategy.
• Sharpe > 3: Exceptional. Typical of HFT and market-making strategies.

Let's compare two trading strategies:

Strategy A — Momentum:

• Annual return: 15%
• Risk-free rate: 5%
• Annual volatility: 20%
• Sharpe = (15% - 5%) / 20% = 0.50

Strategy B — Statistical Arbitrage:

• Annual return: 8%
• Risk-free rate: 5%
• Annual volatility: 3%
• Sharpe = (8% - 5%) / 3% = 1.00

Strategy A has higher raw returns (15% vs. 8%), but Strategy B has a higher Sharpe ratio (1.00 vs. 0.50). Strategy B generates more return per unit of risk. With leverage, Strategy B could be scaled up to match Strategy A's returns while maintaining much lower risk.

Scaling with leverage: If you lever Strategy B 5x, it would generate roughly 5 × 3% = 15% excess return with 5 × 3% = 15% volatility, giving a Sharpe of 1.00 — still much better risk-adjusted performance than Strategy A. This illustrates why the Sharpe ratio, not raw returns, is the primary metric: a high-Sharpe strategy can always be leveraged to achieve any desired return level.

At quant trading firms, the Sharpe ratio is used at every stage of the strategy lifecycle:

• Backtesting: The first metric reported for any backtested strategy. A strategy with a backtest Sharpe below 1.5 is typically not worth further investigation at a top quant firm. Note that backtest Sharpes are inflated relative to live performance — a backtest Sharpe of 2 might correspond to a live Sharpe of 1-1.5.
• Capital allocation: Firms allocate capital to strategies proportional to their Sharpe ratios (this is directly related to the Kelly criterion). Higher-Sharpe strategies get more capital.
• Live monitoring: The rolling Sharpe ratio (e.g., trailing 6-month Sharpe) is monitored in real time. A significant decline in Sharpe may trigger position reductions or strategy review.
• Performance attribution: Breaking down the Sharpe ratio by time period, market regime, and signal component helps researchers understand what drives strategy performance.
• Hiring decisions: Understanding the Sharpe ratio and its properties is a baseline expectation in quant interviews. If you can't compute and interpret a Sharpe ratio, you won't get through the first round.

Typical Sharpe ratios by strategy type: market making (3-6), HFT stat arb (3-10), daily stat arb (1.5-3), medium-frequency strategies (1-2), long-only equity (0.3-0.7).