The Sharpe Ratio Trap — Why Same Returns Get Different Scores
Two funds delivered the same +12% annual return. One climbed steadily at +1% per month; the other swung between +50% and −30% and landed at the same +12% by accident. Are these two funds equally skilled? The Sharpe Ratio scores them completely differently — same +12%, one gets 1.5 and the other 0.5. This article walks through the math, and why Sortino / Calmar exist as alternatives.
1. Definition — return penalized by volatility
The Sharpe Ratio (William Sharpe, 1966) is a simple ratio:
Where:
- Rp: annual portfolio return (arithmetic mean)
- Rf: risk-free rate (treasury / base rate)
- σp: standard deviation of portfolio returns (volatility)
Numerator: "how much more than holding risk-free" — the reward for taking risk. Denominator: "how much risk was that." Higher ratio = more excess return per unit of risk.
2. Same 12% — different scores
Hypothetical: risk-free Rf = 3%, both funds return 12% annual.
| Metric | Steady | Volatile |
|---|---|---|
| Annual return (Rp) | +12% | +12% |
| Volatility (σ, annualized) | 6% | 18% |
| Sharpe = (12 − 3) / σ | 1.5 | 0.5 |
Same return, 3× difference in score. The "Steady" fund delivered the same 9pp excess return with 1/3 the volatility — strictly better risk efficiency.
3. Sharpe levels — rules of thumb
| Sharpe | Interpretation |
|---|---|
| < 0 | Lost money even vs. holding cash |
| 0.0 – 0.5 | Mediocre — most active funds |
| 0.5 – 1.0 | Acceptable — S&P 500 long-term ≈ 0.5 |
| 1.0 – 2.0 | Strong |
| > 2.0 | Exceptional / verify |
Historical S&P 500 (1928–2023): Sharpe ~0.4–0.5. Berkshire Hathaway (1976–2023): Sharpe ~0.76 — legendary precisely because Sharpe > 0.7 sustained over decades is rare.
4. The trap — volatility ≠ risk
Sharpe's biggest limit: it treats all volatility as bad. But up-volatility (sudden gains) and down-volatility (sudden losses) are not the same.
Example: a fund that delivers +1%, +1%, +1%, +1%, +20%, +1%, +1% has high volatility (the +20% spike) — Sharpe is penalized. But the volatility was all upside. Compare to −10%, +1%, +1%, +1%, +1%, +1%, +20% — same average, same σ, but the path is very different.
This is why Sortino Ratio exists: only downside σ in the denominator. And Calmar Ratio: return divided by maximum drawdown.
| Ratio | Denominator | Use |
|---|---|---|
| Sharpe | Total σ | General benchmark |
| Sortino | Downside σ only | Strategies that target asymmetric returns |
| Calmar | Max drawdown | Capital-preservation strategies |
5. Other practical pitfalls
① Short measurement window
1-year Sharpe is statistically weak. Academic minimum: 3–5 years. Funds advertise stellar 1-year Sharpe that collapse next year.
② Risk-free rate choice
Different Rf inputs (3-month T-bill vs 10-year treasury) shift Sharpe by 0.1–0.3. Always compare funds with the same Rf assumption.
③ Return distribution
Sharpe assumes returns are roughly normal. Strategies with fat tails (options writing, leveraged ETFs) can show high Sharpe in calm times but blow up — see "picking up pennies in front of a steamroller."
④ Gross vs net
Marketing Sharpe is often gross-of-fees. Net Sharpe (after expense ratio) is what investors actually get.
Summary
- Sharpe = (return − risk-free) / σ. Standard "risk-adjusted return."
- Same return, different σ = different Sharpe — 3× gap easy.
- Sharpe treats all volatility as bad — its biggest blind spot.
- Sortino / Calmar handle asymmetry — use when strategy targets downside protection.
- Watch the 4 pitfalls: window, Rf, distribution, fees.