Sharpe Ratio vs Sortino Ratio vs Treynor Ratio: Risk-Adjusted Return Comparison

Risk-adjusted return ratios measure whether a portfolio earned excess return commensurate with its risk. Raw return alone is misleading — a portfolio that earns 15% in a year where the S&P did 25% has UNDERPERFORMED, but a portfolio earning 12% with low volatility may have BEATEN a 20% high-volatility portfolio on a risk-adjusted basis. **The three main ratios**: **Sharpe Ratio** = (Portfolio Return - Risk-Free Rate) / Standard Deviation of Portfolio Returns Measures excess return per unit of TOTAL risk (both upside and downside volatility). The most widely used risk-adjusted return measure. **Sortino Ratio** = (Portfolio Return - Risk-Free Rate) / Downside Deviation of Portfolio Returns Similar to Sharpe, but only considers DOWNSIDE volatility (returns below a target, typically 0 or the risk-free rate). More appropriate when upside volatility isn't considered 'bad' risk. **Treynor Ratio** = (Portfolio Return - Risk-Free Rate) / Beta of Portfolio Measures excess return per unit of SYSTEMATIC (market) risk. Uses beta instead of standard deviation. Appropriate for well-diversified portfolios where idiosyncratic risk is assumed to be diversified away. **Key distinctions**: | Ratio | Risk measure | Best for | |-------|--------------|---------| | Sharpe | Total standard deviation | General comparisons, single asset or portfolio | | Sortino | Downside deviation only | Asymmetric strategies (absolute return, alternatives) | | Treynor | Beta (systematic risk) | Well-diversified portfolios compared to market | **Interpretation benchmarks**: - **Sharpe > 2.0**: excellent - **Sharpe 1.0 - 2.0**: good - **Sharpe 0.5 - 1.0**: acceptable - **Sharpe 0 - 0.5**: subpar - **Sharpe < 0**: underperforming risk-free rate — very poor Similar benchmarks apply to Sortino (generally higher values are expected) and Treynor (interpretation is more contextual based on market comparisons). Snap a photo of any portfolio performance problem and FinanceIQ calculates all three ratios, compares them across portfolios, and explains which ratio is most appropriate for the specific context. This content is for educational purposes only and does not constitute financial advice. Past performance does not guarantee future results. Investing involves risk of loss.

The Sharpe ratio, developed by William F. Sharpe in 1966, is the most widely used risk-adjusted return measure in finance. **Formula**: Sharpe Ratio = (Rp - Rf) / σp Where: - Rp = portfolio return (often annualized) - Rf = risk-free rate (typically the 3-month T-bill yield) - σp = standard deviation of portfolio returns (annualized) **Worked example**: A portfolio returns 12% annually with a standard deviation of 15%. The risk-free rate is 4%. Sharpe = (12% - 4%) / 15% = 8% / 15% = 0.53 This is a modest Sharpe — acceptable but not impressive. A Sharpe of 0.53 means the portfolio earned 0.53 percentage points of excess return for each percentage point of risk (standard deviation). **Annualization**: Most Sharpe ratios are reported on an ANNUAL basis. If you have daily or monthly returns: Annualized Sharpe = Sharpe × √N Where N is the number of periods per year (252 for daily, 12 for monthly, 52 for weekly). Example: a strategy has monthly Sharpe of 0.25. Annualized = 0.25 × √12 = 0.87. **Interpretation**: - **Sharpe > 2.0**: exceptional. Often indicates a strategy with high return and low volatility. Rare for long-term stock strategies; more common for short-term arbitrage or hedged strategies. - **Sharpe 1.0-2.0**: very good. Most professionally-managed portfolios aim for this range. - **Sharpe 0.5-1.0**: acceptable. The S&P 500 has historically been in this range (about 0.4-0.6 long-term). - **Sharpe 0-0.5**: subpar risk-adjusted performance. - **Sharpe < 0**: portfolio underperformed the risk-free rate. Always a red flag. **Historical context**: - US stocks (S&P 500), 1928-2024: long-term Sharpe ~0.40-0.50 - Long-term Treasury bonds: Sharpe ~0.30-0.40 - 60/40 balanced portfolio: Sharpe ~0.50-0.60 - Best hedge funds (Renaissance Medallion, Citadel, Two Sigma): reported Sharpe > 2.0 - Bernie Madoff's claimed Sharpe was ~2.5 — a classic red flag for too-good-to-be-true performance **Comparing portfolios with Sharpe**: When comparing two portfolios: - Portfolio A: 15% return, 20% std dev → Sharpe = (15-4)/20 = 0.55 - Portfolio B: 10% return, 12% std dev → Sharpe = (10-4)/12 = 0.50 Portfolio A has HIGHER raw return but very similar Sharpe. Portfolio A's extra return came at the cost of significantly more volatility. On a risk-adjusted basis, they're roughly comparable. **Portfolio leverage and Sharpe**: Leverage doesn't change Sharpe ratio. A portfolio with Sharpe = 0.8 leveraged 2x has the same Sharpe (0.8). This is one of the Sharpe's useful properties — it's scale-invariant. **Limitations of Sharpe**: 1. **Treats upside and downside volatility equally**: a portfolio that occasionally spikes up is penalized the same as one that occasionally drops. Most investors don't mind upside volatility. 2. **Assumes normal distribution**: real returns have 'fat tails' — extreme events are more common than normal distribution predicts. Sharpe can overstate performance for strategies with left-tail risk. 3. **Sensitive to measurement period**: shorter time periods can produce misleading Sharpe values if the period happens to be a bull or bear market. 4. **Can be gamed**: selling options or taking on tail risk can artificially inflate Sharpe until a rare event occurs. Despite these limitations, Sharpe remains the industry standard because it's simple, universally understood, and directly comparable across strategies.

The Sortino ratio, developed by Frank Sortino in the 1980s, addresses a key weakness of the Sharpe ratio: it penalizes upside volatility the same as downside. **Formula**: Sortino Ratio = (Rp - Rf) / σd Where: - Rp = portfolio return - Rf = risk-free rate (or another 'minimum acceptable return' threshold) - σd = downside deviation (standard deviation of returns BELOW the minimum acceptable return) **Calculating downside deviation**: 1. Identify all returns below the minimum acceptable return (MAR), typically Rf or 0. 2. Calculate the deviations of these returns below MAR. 3. Square each deviation. 4. Average them. 5. Take the square root. Returns above MAR are NOT included in the calculation — only the downside matters. **Worked example**: Monthly returns: 5%, 3%, -2%, 4%, -3%, 2%, -1%, 6%, -4%, 3%, 5%, 1% Assume MAR = 0 (typical simplification). Negative returns: -2%, -3%, -1%, -4% (all below 0) Deviations below MAR: -2%, -3%, -1%, -4% Squared deviations: 4, 9, 1, 16 Sum: 30 Average (divided by TOTAL number of months, 12): 30/12 = 2.5 Downside deviation (monthly): √2.5 = 1.58% Annualized: 1.58% × √12 = 5.48% Average monthly return: (5+3-2+4-3+2-1+6-4+3+5+1)/12 = 19/12 = 1.58% Annualized: 1.58% × 12 = 19% Sortino = (19% - 4%) / 5.48% = 15/5.48 = 2.74 This is an excellent Sortino. Note that the same returns would have a lower Sharpe because Sharpe uses TOTAL standard deviation (including upside volatility), which is higher than downside deviation alone. **When Sortino is more appropriate than Sharpe**: 1. **Absolute-return strategies**: hedge funds, market-neutral, long/short strategies. These aim for positive returns regardless of market direction. Upside volatility is good; downside is bad. 2. **Asymmetric return distributions**: strategies with skewed returns (e.g., covered calls, momentum strategies) where upside and downside are structurally different. 3. **Retirement planning**: investors nearing retirement care more about downside (losing principal) than upside (occasional big gains). 4. **Options strategies**: options have inherently asymmetric payoffs. **Interpretation**: - **Sortino > 3.0**: excellent - **Sortino 2.0-3.0**: very good - **Sortino 1.0-2.0**: acceptable - **Sortino < 1.0**: poor Note that Sortino ratios are typically HIGHER than Sharpe ratios for the same portfolio because the denominator (downside deviation) is smaller than total standard deviation. **Relationship to Sharpe**: For a portfolio with symmetric return distribution, Sortino ≈ Sharpe × √2 (because half the volatility is upside, half is downside). For highly asymmetric distributions, Sortino can be much larger than Sharpe: - Momentum strategies: typical Sortino is 1.5-2x Sharpe - Options selling strategies: Sortino can be much lower than Sharpe (tail events dominate) **Limitations of Sortino**: 1. **Less standardized**: multiple definitions exist for downside deviation and MAR. Less directly comparable across sources. 2. **Requires more data**: needs enough periods with negative returns to accurately estimate downside deviation. Not reliable for short or very stable track records. 3. **Still assumes normality**: downside deviation still underestimates tail risk in non-normal distributions. 4. **MAR choice affects the result**: different MARs (0, Rf, or investor-specific benchmarks) produce different ratios. FinanceIQ calculates Sortino with flexible MAR definitions and compares it to Sharpe to reveal when the two ratios diverge significantly (indicating asymmetric return distributions).

The Treynor ratio, developed by Jack Treynor in 1965, uses a different risk measure entirely: beta, rather than standard deviation. **Formula**: Treynor Ratio = (Rp - Rf) / βp Where: - Rp = portfolio return - Rf = risk-free rate - βp = portfolio beta (from CAPM) **Worked example**: A portfolio returns 12%, has a beta of 1.3. Risk-free rate is 4%. Treynor = (12% - 4%) / 1.3 = 8% / 1.3 = 6.15 This represents excess return per unit of systematic (market) risk. **When Treynor is most appropriate**: **1. Well-diversified portfolios**: if a portfolio is well-diversified (holds many stocks across sectors), its idiosyncratic risk is largely diversified away. Only systematic risk (beta) remains as a meaningful risk measure. Treynor captures this. **2. Comparing portfolios to the market**: Treynor makes sense when all portfolios use the same market benchmark. Comparing a large-cap US stock portfolio to the S&P 500 using Treynor works well. **3. Performance evaluation of diversified funds**: mutual funds, index funds, and ETFs are well-suited for Treynor evaluation. **When Treynor is NOT appropriate**: **1. Single stocks or concentrated portfolios**: idiosyncratic risk is not diversified away. Using only beta understates the true risk. **2. Strategies with beta near zero**: market-neutral hedge funds may have beta near zero but still have significant risk. Treynor becomes meaningless (dividing by zero). **3. Alternative assets**: hedge funds, private equity, real estate, commodities may not fit a single-factor beta model well. **4. Cross-asset comparison**: comparing a stock portfolio to a bond portfolio using Treynor is misleading because they have different betas to the same market. **Interpretation**: Treynor is unit-less (percentage / beta = dimensionless). Absolute values are harder to interpret than Sharpe. Best used as a relative comparison between portfolios. - Portfolio A: Treynor = 6.15 - Portfolio B: Treynor = 8.20 - Portfolio B has better risk-adjusted performance on a systematic-risk basis Compare to market: the market portfolio (β = 1) has Treynor = Market Return - Rf. If the market's Treynor is higher than your portfolio's, your portfolio underperformed on a risk-adjusted basis. **Choosing between the three ratios**: **Decision framework**: 1. **Is the portfolio well-diversified?** - **Yes** → consider Treynor (or Sharpe) - **No** → use Sharpe or Sortino 2. **Are returns symmetric (roughly equal up and down moves)?** - **Yes** → Sharpe is fine - **No** (asymmetric) → Sortino is better 3. **Is the benchmark clearly the market?** - **Yes** → Treynor fits - **No** (absolute return strategy) → Sortino 4. **Are you comparing hedge funds or alternative strategies?** - **Yes** → Sortino is often preferred - **No** → Sharpe or Treynor **Practical convention**: most investor reports include BOTH Sharpe and Sortino. Treynor is less commonly reported but useful for diversified equity portfolios. **Worked comparison**: Portfolio A (a diversified equity mutual fund): - Return: 14% - Std dev: 16% - Downside deviation: 10% - Beta: 1.05 - Rf: 4% Sharpe = (14-4)/16 = 0.625 Sortino = (14-4)/10 = 1.00 Treynor = (14-4)/1.05 = 9.52 Portfolio B (a market-neutral hedge fund): - Return: 8% - Std dev: 5% - Downside deviation: 3% - Beta: 0.05 (nearly zero) - Rf: 4% Sharpe = (8-4)/5 = 0.80 Sortino = (8-4)/3 = 1.33 Treynor = (8-4)/0.05 = 80 (extremely high — not meaningful because beta is near zero) For Portfolio B, Treynor is distorted by the near-zero beta. Sharpe and Sortino give more meaningful comparisons. **Common mistake**: comparing ratios across portfolios with different time periods, frequencies, or risk-free rates. Always ensure consistency in inputs when comparing. FinanceIQ calculates all three ratios from return data, identifies which is most appropriate for the portfolio type, and explains when the different measures diverge (which reveals information about the portfolio's characteristics).