Sharpe Ratio: Formula, Calculation & How to Use It 2026

What Is the Sharpe Ratio?

The Sharpe ratio is a risk-adjusted performance metric that measures how much excess return an investment generates per unit of total risk. It answers a straightforward question: for every unit of volatility you endure, how much return above the risk-free rate do you earn?

William Sharpe introduced the ratio in 1966, originally calling it the "reward-to-variability ratio." The concept earned him a share of the 1990 Nobel Memorial Prize in Economics and became the single most widely used measure of risk-adjusted performance in finance. Whenever you see a hedge fund factsheet, a mutual fund comparison table, or a quant strategy backtest, the Sharpe ratio is almost certainly there.

The intuition is simple. Two funds might both return 15% in a year, but if one did it with smooth, steady gains while the other swung wildly between +8% and -6% each month, they're not equivalent. The first fund gave you the same return with less stress, less risk of a margin call, and less temptation to panic-sell at the bottom. The Sharpe ratio captures this difference in a single number.

In 2026, the Sharpe ratio remains the default language for comparing risk-adjusted returns across asset classes, strategies, and time periods. It's the first number institutional allocators look at, and often the last thing they check before making a decision.

---

The Sharpe Ratio Formula

The Sharpe ratio equals the portfolio's excess return over the risk-free rate, divided by the standard deviation of those returns.

Sharpe Ratio = (Rp - Rf) / σp

Where:

• Rp = the portfolio's average return over the measurement period
• Rf = the risk-free rate of return over the same period
• σp = the standard deviation of the portfolio's returns

Each component deserves its own explanation.

The Numerator: Excess Return (Rp - Rf)

The numerator strips out the return you could have earned by doing nothing risky. If the risk-free rate is 4% and your portfolio returned 12%, the excess return is 8%. This is the compensation you received specifically for bearing risk.

The risk-free rate is typically proxied by the yield on short-term government bonds - UK gilts for GBP-denominated strategies, US Treasury bills for USD strategies. In 2026, with UK base rates settling around 3.5-4%, the risk-free rate has a meaningful impact on Sharpe calculations. In the near-zero rate environment of 2020-2021, the risk-free rate barely mattered. Now it does.

The Denominator: Total Volatility (σp)

The denominator is the standard deviation of the portfolio's returns. This measures total volatility - both upside and downside moves. A return series that jumps around a lot has a high standard deviation; one that moves in a narrow range has a low standard deviation.

Standard deviation treats a +3% return and a -3% return as equally "risky," which is one of the Sharpe ratio's key limitations. Investors don't actually mind upside surprises. Metrics like the Sortino ratio address this by using only downside deviation, but the Sharpe ratio's simplicity and universality keep it as the industry standard.

Putting It Together

A Sharpe ratio of 1.0 means the portfolio earns one unit of excess return for each unit of risk (standard deviation). A ratio of 2.0 means two units of return per unit of risk. Higher is better - you're being compensated more generously for the volatility you're taking on.

---

Step-by-Step Sharpe Ratio Calculation

Let's work through a concrete example with actual numbers. Suppose you're evaluating a UK equity portfolio with the following monthly returns over six months, and the annualised risk-free rate is 4.0%.

Step 1: Collect the return data.

Step 2: Convert the risk-free rate to a monthly figure.

Annual risk-free rate = 4.0%

Monthly risk-free rate = 4.0% / 12 = 0.333%

Step 3: Calculate the excess return for each month.

Step 4: Calculate the mean excess return.

Mean = (2.467 + (-1.533) + 3.167 + (-0.733) + 1.567 + 2.267) / 6

Mean = 7.202 / 6 = 1.200%

Step 5: Calculate the standard deviation of the excess returns.

First, find the squared deviations from the mean:

• (2.467 - 1.200)^2 = 1.605
• (-1.533 - 1.200)^2 = 7.477
• (3.167 - 1.200)^2 = 3.868
• (-0.733 - 1.200)^2 = 3.741
• (1.567 - 1.200)^2 = 0.135
• (2.267 - 1.200)^2 = 1.138

Sum = 17.964

Variance (using N-1 for sample) = 17.964 / 5 = 3.593

Standard deviation = sqrt(3.593) = 1.896%

Step 6: Compute the monthly Sharpe ratio.

Sharpe Ratio (monthly) = 1.200 / 1.896 = 0.633

Step 7: Annualise.

Annualised Sharpe Ratio = 0.633 x sqrt(12) = 0.633 x 3.464 = 2.19

This portfolio has an annualised Sharpe ratio of approximately 2.19, which - as we'll see in the next section - is very good. The strategy is generating strong returns relative to the risk it's taking.

---

Annualising the Sharpe Ratio

The raw Sharpe ratio depends on the frequency of your return data. A monthly Sharpe will be numerically different from a daily Sharpe computed over the same period. To make figures comparable across different data frequencies, you annualise by multiplying by the square root of the number of periods in a year.

Annualised Sharpe = Sharpe (per period) x sqrt(N)

Where N is the number of return periods per year:

Why Square Root?

The square root scaling comes from the assumption that returns are independently and identically distributed (i.i.d.). Under this assumption, the mean return scales linearly with time while the standard deviation scales with the square root of time. The ratio of the two therefore scales with the square root of time.

This assumption is rarely perfectly true - financial returns exhibit autocorrelation, volatility clustering, and time-varying distributions. But it's a reasonable approximation for most practical purposes, and it's the universal convention. If you deviate from it, make sure your audience understands why.

A Common Mistake

One frequent error is annualising the numerator and denominator separately rather than the ratio as a whole. Some practitioners compute an annualised mean return and an annualised standard deviation, then divide. Under the i.i.d. assumption, this gives the same result. But if the data has serial correlation, the two approaches can diverge. The standard practice is to compute the Sharpe ratio at the native data frequency and then multiply by sqrt(N).

---

What Is a Good Sharpe Ratio?

A Sharpe ratio above 1.0 is generally considered good, above 2.0 is very good, and anything above 3.0 is exceptional but should be examined closely for sustainability. These benchmarks depend on strategy type, asset class, and market conditions.

Context by Strategy Type

These thresholds aren't one-size-fits-all. What counts as a good Sharpe ratio varies significantly by strategy:

• Passive equity index (e.g. FTSE All-Share): Historically around 0.3 to 0.5 over long periods. Equity markets compensate investors for risk, but not enormously efficiently on a Sharpe basis.
• Long/short equity hedge funds: Typically target 0.8 to 1.5. The hedging reduces volatility relative to long-only, which improves the Sharpe ratio even if absolute returns are lower.
• Market-neutral statistical arbitrage: Often targets 1.5 to 3.0. The low net exposure keeps volatility down, and consistent small gains can produce high Sharpe figures.
• High-frequency market making: Can show Sharpe ratios above 5.0 or even 10.0 on a daily basis, but these strategies are capacity-constrained and not accessible to most investors.
• Trend-following / CTA: Typically 0.5 to 1.0. These strategies experience extended drawdown periods between large profitable moves, which drags on the Sharpe ratio.

Beware of Unrealistically High Figures

A Sharpe ratio above 3.0 sustained over multiple years should prompt questions, not celebrations. Possible explanations include: the strategy is operating at very small scale (capacity-constrained), the return data is smoothed or stale-priced, the measurement period doesn't include a stress event, or the strategy is taking hidden tail risk that hasn't yet materialised. Always interrogate outlier Sharpe ratios before allocating capital.

---

Calculating the Sharpe Ratio in Python

Here's a practical Python implementation for computing the Sharpe ratio from a return series, including a rolling Sharpe calculation for monitoring strategy performance over time.

import numpy as np
import pandas as pd


def sharpe_ratio(
    returns: pd.Series,
    risk_free_rate: float = 0.0,
    annualise: bool = True,
    periods_per_year: int = 252,
) -> float:
    """
    Calculate the Sharpe ratio for a return series.

    Parameters
    ----------
    returns : pd.Series
        Period returns (e.g. daily or monthly).
    risk_free_rate : float
        Risk-free rate per period. Default is 0.
    annualise : bool
        If True, scale the ratio by sqrt(periods_per_year).
    periods_per_year : int
        Trading days (252), months (12), or weeks (52).

    Returns
    -------
    float
        The Sharpe ratio.
    """
    excess_returns = returns - risk_free_rate
    mean_excess = excess_returns.mean()
    std_excess = excess_returns.std(ddof=1)

    if std_excess == 0:
        return float("inf") if mean_excess > 0 else 0.0

    ratio = mean_excess / std_excess

    if annualise:
        ratio *= np.sqrt(periods_per_year)

    return ratio


def rolling_sharpe(
    returns: pd.Series,
    window: int = 63,
    risk_free_rate: float = 0.0,
    periods_per_year: int = 252,
) -> pd.Series:
    """
    Calculate a rolling annualised Sharpe ratio.

    Parameters
    ----------
    returns : pd.Series
        Period returns.
    window : int
        Lookback window in periods. Default is 63 (approx. 3 months).
    risk_free_rate : float
        Risk-free rate per period.
    periods_per_year : int
        Trading days (252), months (12), or weeks (52).

    Returns
    -------
    pd.Series
        Rolling annualised Sharpe ratio.
    """
    excess = returns - risk_free_rate
    rolling_mean = excess.rolling(window).mean()
    rolling_std = excess.rolling(window).std(ddof=1)
    return (rolling_mean / rolling_std) * np.sqrt(periods_per_year)


# --- Example usage ---
np.random.seed(42)
dates = pd.bdate_range("2024-01-02", periods=504)  # ~2 years of daily data

# Simulated daily returns for a long/short equity strategy
daily_returns = pd.Series(
    np.random.normal(0.0003, 0.008, len(dates)),
    index=dates,
)

# Convert annual risk-free rate to daily
annual_rf = 0.04
daily_rf = annual_rf / 252

# Full-period Sharpe
full_sharpe = sharpe_ratio(
    daily_returns,
    risk_free_rate=daily_rf,
    periods_per_year=252,
)
print(f"Annualised Sharpe Ratio: {full_sharpe:.2f}")

# Rolling 63-day (3-month) Sharpe
rolling = rolling_sharpe(
    daily_returns,
    window=63,
    risk_free_rate=daily_rf,
    periods_per_year=252,
)
print(f"Rolling Sharpe (latest): {rolling.iloc[-1]:.2f}")
print(f"Rolling Sharpe (min):    {rolling.min():.2f}")
print(f"Rolling Sharpe (max):    {rolling.max():.2f}")

A few things to note about this implementation:

• Sample standard deviation (ddof=1) is used in the denominator rather than population standard deviation. This is the standard convention for ex-post Sharpe ratio estimation and avoids downward bias in small samples.
• The risk-free rate is per-period. If you're working with daily returns and a 4% annualised risk-free rate, convert it first: daily_rf = 0.04 / 252. Forgetting this conversion is a common source of error.
• The rolling Sharpe uses a 63-day window (roughly one quarter of trading days). This lets you monitor how the strategy's risk-adjusted performance evolves. Shrinking the window makes the series noisier; widening it makes it smoother but less responsive to regime changes.
• Annualisation multiplies by sqrt(252) for daily data, following the standard square root of time convention.

For production use, you'd want input validation, handling for edge cases like constant return series, and potentially a confidence interval around the Sharpe estimate (standard error of the Sharpe ratio is approximately sqrt((1 + 0.5 x SR^2) / N), where N is the number of observations).

---

Sharpe Ratio vs Other Risk-Adjusted Metrics

The Sharpe ratio is the most widely used risk-adjusted metric, but it's not the only one. Different ratios capture different aspects of the risk-return relationship.

When to Use the Sharpe Ratio

The Sharpe ratio is the right default choice when you need a single number to compare investments that don't share a common benchmark. It works well for absolute-return strategies, cross-asset comparisons, and any situation where total volatility is a reasonable proxy for risk.

When to Use Something Else

Switch to the Sortino ratio when return distributions are meaningfully skewed - options strategies, trend following, or any approach that deliberately produces non-normal payoffs. The Sharpe ratio penalises large positive returns, which misrepresents these strategies.

Switch to the Calmar ratio when maximum drawdown is the binding constraint. Allocators with hard drawdown limits (common in managed futures) care more about the worst loss than about general volatility.

Switch to the Information ratio when evaluating performance relative to a specific benchmark. A UK equity manager benchmarked to the FTSE 100 should be assessed on how consistently they beat that index, not on their absolute risk-adjusted return.

Switch to the Treynor ratio when the portfolio is one component of a larger diversified allocation and you want to evaluate compensation for systematic (non-diversifiable) risk only.

In practice, institutional investors examine multiple metrics together. No single ratio tells the complete story.

---

Common Pitfalls and How to Avoid Them

The Sharpe ratio is simple to compute but easy to misuse. Here are the mistakes practitioners make most often.

Non-Normal Returns

The Sharpe ratio assumes that standard deviation adequately describes risk. For normally distributed returns, this works well - the mean and standard deviation fully characterise the distribution. But real-world returns often exhibit skewness (asymmetry) and excess kurtosis (fat tails). A strategy that produces frequent small gains and occasional catastrophic losses can have a deceptively high Sharpe ratio because the standard deviation doesn't fully capture the tail risk.

This is the "picking up pennies in front of a steamroller" problem. A short volatility fund might show a Sharpe ratio of 2.0 right up until the day it loses 40%. The standard deviation, calculated from the many small positive returns, understates the true risk.

Sharpe Ratio Manipulation

Because the Sharpe ratio is so closely watched, there are well-documented ways to inflate it artificially:

• Smoothing returns: Using illiquid assets with infrequent mark-to-market reduces measured volatility without reducing actual risk.
• Writing out-of-the-money options: Generates consistent premium income that boosts the mean return while concentrating risk in rare but severe loss events.
• Selective time periods: Choosing a measurement window that excludes a drawdown can dramatically improve the Sharpe ratio.
• Compounding frequency games: Reporting annualised figures from very short track records can produce impressive but meaningless numbers.

The best defence is to look at the full return distribution - not just the first two moments - and to examine performance over complete market cycles.

Time Period Dependency

The Sharpe ratio is highly sensitive to the measurement window. A strategy might show a Sharpe of 3.0 from 2023 to 2025 (a period of generally rising markets) and 0.4 over 2020 to 2025 (which includes the COVID crash). Neither figure is "wrong," but they tell very different stories. Always ask what period a Sharpe ratio covers, and be sceptical of figures from cherry-picked windows.

You need at least three to five years of data, ideally spanning different market regimes, before treating a Sharpe ratio as a reliable indicator of future risk-adjusted performance.

Leverage Effects

Leverage doesn't change the Sharpe ratio in theory - it scales both the numerator and denominator proportionally. In practice, however, leverage introduces non-linearities: margin calls can force liquidation at the worst time, borrowing costs eat into the excess return, and the risk of ruin makes a high-leverage strategy fundamentally different from a low-leverage one with the same Sharpe ratio.

Two strategies with identical Sharpe ratios but different leverage profiles are not equivalent investments. The higher-leverage version carries risks that the Sharpe ratio doesn't capture.

Ignoring Skewness and Kurtosis

Standard deviation treats upside and downside volatility as equally undesirable. An investor who cares about downside risk should supplement the Sharpe ratio with metrics that specifically address the shape of the return distribution - the Sortino ratio for downside deviation, or direct examination of skewness and kurtosis statistics. Understanding risk management frameworks helps you see beyond a single summary statistic.

---

The Sharpe Ratio in Practice

How Hedge Funds Report It

The Sharpe ratio is a headline figure on virtually every hedge fund factsheet and pitch deck. Funds typically report the annualised Sharpe ratio calculated from monthly net-of-fee returns since inception. Some also show a rolling 36-month Sharpe to give prospective investors a sense of how risk-adjusted performance has evolved.

Institutional-quality reporting uses excess returns over the prevailing risk-free rate (usually 3-month T-bills), though some funds - particularly in the near-zero rate era - reported Sharpe ratios with Rf = 0. In 2026, with meaningful risk-free rates, this distinction matters again. Always check the fine print.

How Allocators Use It

Fund-of-fund managers and institutional allocators use the Sharpe ratio as a first-pass filter. A common screen might exclude any strategy with a Sharpe ratio below 0.5 over its track record. Surviving funds then face deeper due diligence that goes well beyond the Sharpe ratio - examining drawdown characteristics, factor exposures, liquidity terms, and operational risk.

Sophisticated allocators also look at the Sharpe ratio's stability over time. A fund with a consistent Sharpe ratio of 1.2 across different market environments is more attractive than one that swings between 3.0 and -0.5, even if the average is similar.

How to Compare Strategies Fairly

Comparing Sharpe ratios across strategies requires care:

• Use the same time period. Comparing a 2024-2025 Sharpe to a 2020-2025 Sharpe is meaningless.
• Use the same risk-free rate. Mixing Rf = 0 and Rf = 4% produces incomparable figures.
• Use the same return frequency. Daily and monthly Sharpe ratios can differ due to autocorrelation effects.
• Account for fees. Compare net-of-fee returns to net-of-fee returns. A gross Sharpe of 2.0 with a 2-and-20 fee structure looks very different net.
• Consider capacity. A small-capacity niche strategy with a Sharpe of 3.0 isn't comparable to a large-scale diversified fund with a Sharpe of 1.0. The former might not be scalable.

---

Limitations of the Sharpe Ratio

The Sharpe ratio's strengths - simplicity, universality, and interpretability - come with genuine weaknesses that every practitioner should understand.

Assumes Normally Distributed Returns

The ratio uses standard deviation as the complete measure of risk. This works perfectly for normal distributions, where two parameters (mean and standard deviation) fully describe the distribution. But financial returns are almost never perfectly normal. They exhibit fat tails (more extreme events than a normal distribution predicts) and skewness (asymmetric upside and downside). For strategies with highly non-normal returns, the Sharpe ratio can be misleading.

Penalises Upside Volatility Equally

A strategy that suddenly gains 15% in one month will see its standard deviation increase, dragging its Sharpe ratio down. But that 15% gain isn't risk from the investor's perspective - it's exactly what they wanted. The Sharpe ratio doesn't distinguish between welcome and unwelcome volatility. The Sortino ratio directly addresses this by using only downside deviation.

Can Be Gamed

As discussed in the pitfalls section, the Sharpe ratio can be artificially inflated through return smoothing, tail-risk selling, or strategic choice of measurement window. Its prominence in the industry creates a perverse incentive to optimise for the metric rather than for genuine risk-adjusted performance.

Ignores Path Dependency

Two strategies can have the same mean return and standard deviation - and therefore the same Sharpe ratio - but very different return paths. One might produce steady gains with a brief 5% drawdown, while the other swings through a 30% drawdown before recovering. The Sharpe ratio treats these as identical, but investor experience differs dramatically. The Calmar ratio captures this distinction by using maximum drawdown rather than standard deviation.

Serial Correlation Distortion

When returns are autocorrelated - common in strategies using illiquid assets, mean-reverting signals, or smoothed NAV reporting - the measured standard deviation understates true volatility. This inflates the Sharpe ratio beyond what the actual risk warrants. Andrew Lo's 2002 paper demonstrated that even modest autocorrelation can substantially distort annualised Sharpe ratios, and proposed adjustments to correct for this. In practice, many allocators now check for autocorrelation in reported returns as a basic due diligence step.

Despite these limitations, the Sharpe ratio remains the most important single metric in performance evaluation. It's flawed, but it's a necessary starting point - the lingua franca of risk-adjusted return measurement. The key is to use it alongside other metrics, not in isolation.

---

Frequently Asked Questions

What is the Sharpe ratio in simple terms?

The Sharpe ratio tells you how much extra return you earn for each unit of risk you take. It compares the return above a risk-free investment (like government bonds) to the volatility of those returns. A higher number means better risk-adjusted performance. If a fund has a Sharpe ratio of 1.5, it earns 1.5 units of excess return for every unit of standard deviation. It's the most common way to compare investment performance on an apples-to-apples basis, because it accounts for the fact that higher returns usually come with higher risk.

What is a good Sharpe ratio?

A Sharpe ratio above 1.0 is generally considered good for most investment strategies. Between 1.0 and 2.0 suggests the strategy is generating solid returns relative to its volatility. Above 2.0 is very strong, and above 3.0 is exceptional but rare over sustained periods. Context matters significantly: a long-only equity fund with a Sharpe of 0.5 might be performing well relative to peers, while a market-neutral hedge fund would be expected to achieve 1.0 or higher. Be cautious with Sharpe ratios above 3.0 sustained over years - they often signal smoothed returns, hidden tail risk, or a measurement period that hasn't included a real stress test.

How do you calculate the Sharpe ratio?

Start with a series of returns (daily, weekly, or monthly). Subtract the risk-free rate from each return to get excess returns. Calculate the mean of these excess returns and the standard deviation. Divide the mean by the standard deviation - that's your Sharpe ratio at the native data frequency. To annualise, multiply by the square root of the number of periods in a year (sqrt(252) for daily, sqrt(12) for monthly). For example, if your monthly excess return averages 0.5% with a standard deviation of 1.2%, the monthly Sharpe is 0.42 and the annualised Sharpe is 0.42 x sqrt(12) = 1.44.

What is the difference between the Sharpe ratio and the Sortino ratio?

Both measure risk-adjusted return, but they define risk differently. The Sharpe ratio uses total standard deviation - all volatility, up and down. The Sortino ratio uses only downside deviation, ignoring returns above a target threshold. This means the Sortino ratio doesn't penalise an investment for having large positive returns. For strategies with roughly symmetric return distributions, both give similar rankings. For strategies with significant skew - such as trend-following (positive skew) or short volatility (negative skew) - the two ratios can diverge substantially, and the Sortino ratio often provides a more accurate picture of genuine risk.

Can the Sharpe ratio be negative?

Yes. A negative Sharpe ratio means the portfolio's return is below the risk-free rate. You would have been better off holding risk-free government bonds. This can happen during bear markets, when a strategy's losses drag its average return below the risk-free threshold, or when a strategy is genuinely poor. A negative Sharpe ratio doesn't necessarily mean the portfolio lost money in absolute terms - it could have gained 2% while the risk-free rate was 4% - but it does mean the risk taken wasn't compensated.

How do you annualise the Sharpe ratio from daily data?

Multiply the daily Sharpe ratio by the square root of 252 (the approximate number of trading days in a year). If your daily excess return averages 0.03% with a daily standard deviation of 0.8%, the daily Sharpe is 0.0375. Multiply by sqrt(252) = 15.87 to get an annualised Sharpe of approximately 0.60. The same principle applies for other frequencies: multiply by sqrt(12) for monthly data, sqrt(52) for weekly data, and sqrt(4) for quarterly data. This scaling assumes returns are independently distributed across periods - an approximation that works reasonably well in practice but breaks down when returns are autocorrelated.