Performance Metrics: Sharpe Ratio, Drawdown, and Beyond
Why Performance Metrics Matter
Evaluating a trading strategy requires more than just looking at total return. A strategy that returns 50% but experiences a 70% drawdown along the way is fundamentally different from one that returns 30% with a maximum drawdown of 15%. Performance metrics provide a standardized framework for comparing strategies, understanding risk-return tradeoffs, and identifying potential weaknesses.
This guide presents the most commonly used performance metrics for educational purposes. Understanding these metrics is essential for any student of quantitative finance or systematic trading.
Return Metrics
CAGR (Compound Annual Growth Rate)
CAGR represents the annualized rate of return, accounting for compounding. It answers: "What constant annual return would produce the same total result?"
CAGR = (Ending Value / Beginning Value)^(1/Years) - 1
CAGR is useful for comparing strategies with different time horizons. However, it tells you nothing about the path taken to achieve that return — two strategies with identical CAGR can have vastly different risk profiles.
Annualized Return vs Total Return
Always report annualized returns rather than total returns when comparing strategies over different time periods. A 100% total return over 10 years (about 7.2% annualized) is very different from 100% over 2 years (about 41% annualized).
Risk-Adjusted Return Metrics
Sharpe Ratio
The Sharpe ratio is the most widely used risk-adjusted performance metric. It measures excess return per unit of total risk (standard deviation):
Sharpe = (Return - Risk-Free Rate) / Standard Deviation of Returns
Interpreting Sharpe Ratios
- Below 0.5: Generally considered poor
- 0.5 - 1.0: Acceptable for many strategies
- 1.0 - 2.0: Good performance
- Above 2.0: Excellent, but should be verified carefully (may indicate overfitting)
- Above 3.0: Exceptional and rare in practice; treat with skepticism in backtests
Limitations of the Sharpe Ratio
- Assumes normally distributed returns (financial returns often have fat tails)
- Penalizes upside volatility equally with downside volatility
- Can be manipulated through return smoothing or leverage
- Annualization assumes independent returns, which may not hold
Sortino Ratio
The Sortino ratio addresses one of the Sharpe ratio's main limitations by only penalizing downside volatility:
Sortino = (Return - Risk-Free Rate) / Downside Deviation
Downside deviation only considers returns below a minimum acceptable return (often 0% or the risk-free rate). This makes the Sortino ratio more appropriate for strategies with asymmetric return distributions.
Drawdown Metrics
Maximum Drawdown (MDD)
Maximum drawdown measures the largest peak-to-trough decline in portfolio value. It answers: "What was the worst loss an investor would have experienced from a peak?"
MDD = (Trough Value - Peak Value) / Peak Value
Maximum drawdown is arguably the most important risk metric because it directly relates to the psychological and financial pain an investor would experience. A 50% drawdown requires a 100% gain to recover.
Drawdown Duration
How long does it take to recover from a drawdown? The maximum drawdown duration measures the longest period between reaching a new equity high. Extended drawdown durations test investor patience and can lead to strategy abandonment at the worst possible time.
Calmar Ratio
The Calmar ratio divides annualized return by maximum drawdown:
Calmar = CAGR / |Maximum Drawdown|
A Calmar ratio above 1.0 means the strategy's annualized return exceeds its worst drawdown, which is generally considered acceptable. Higher values indicate better return per unit of maximum pain.
Trade-Level Metrics
Win Rate
The percentage of trades that are profitable. Many successful systematic strategies have win rates below 50% — they make money by having larger average wins than average losses.
Profit Factor
Profit factor is the ratio of gross profits to gross losses:
Profit Factor = Sum of Winning Trades / |Sum of Losing Trades|
A profit factor above 1.0 indicates overall profitability. Values between 1.5 and 2.0 are typically considered good for systematic strategies.
Average Win / Average Loss Ratio
This ratio, combined with win rate, determines overall profitability: a strategy with a 40% win rate needs an average win/loss ratio of at least 1.5 to break even (before costs).
Advanced Metrics
Information Ratio
Similar to Sharpe ratio but measures excess return relative to a benchmark (tracking error) rather than the risk-free rate. It is the standard metric for evaluating active portfolio managers.
Deflated Sharpe Ratio
Proposed by Bailey and Lopez de Prado, the Deflated Sharpe Ratio adjusts the observed Sharpe ratio for the number of strategy variations tested, skewness, kurtosis, and sample size. This is particularly important when evaluating strategies selected from a large universe of candidates.
Omega Ratio
The Omega ratio considers the entire return distribution rather than just the first two moments. It calculates the ratio of gains above a threshold to losses below it, capturing higher-order distribution properties that Sharpe and Sortino miss.
Disclaimer: This guide is for educational purposes only. Performance metrics calculated from historical backtests do not guarantee future strategy performance. Readers should understand that all metrics can be influenced by the specific time period analyzed and should be interpreted with appropriate caution.
The information contained in this report is for educational purposes only and should not be construed as investment advice. Historical backtest results are provided as statistical data and do not guarantee future performance.
Related Guides
Machine Learning in Backtesting: Opportunities and Challenges
Explore the application of machine learning techniques to trading strategy development and backtesting. Learn about feature engineering, random forests, neural networks, and the critical pitfalls of applying ML to financial data.
Factor-Based & Multi-Factor Models: Building Systematic Portfolios
Study factor investing from the Fama-French three-factor model to modern multi-factor portfolio construction. Learn about value, momentum, quality, size, and low volatility factors and how they are used in educational research.
Volatility-Based Trading Systems: Harnessing Market Uncertainty
Explore volatility-based trading approaches including ATR-based position sizing, volatility targeting, VIX-based market timing, and GARCH modeling. Learn how volatility can be used as both a trading signal and a risk management tool.
Pairs Trading & Statistical Arbitrage: A Quantitative Approach
Learn the quantitative foundations of pairs trading and statistical arbitrage, including cointegration analysis, the Engle-Granger methodology, Johansen test, and spread construction techniques for educational research.