Sharpe vs Sortino Ratio: Risk-Adjusted Returns Explained

Why Total Return Is a Misleading Scorecard

Two strategies can both return 20% a year and be nothing alike. One grinds out steady gains with the occasional shallow dip; the other lurches between +15% and −12% months and happens to land in the same place. If you judge them on total return alone, they look identical. If you have to actually hold one through a drawdown, they are worlds apart. The question a return number cannot answer is: how much risk did you take to earn it?

That is what risk-adjusted return ratios measure. The two most widely used are the Sharpe ratio and the Sortino ratio. Both divide your excess return by a measure of risk, but they define “risk” differently — and that difference is the whole point. This guide explains both formulas, why they diverge, and how to read the output. If you just want the numbers, paste your return series into the Sharpe & Sortino calculator and it computes both at once.

The Sharpe Ratio: Excess Return per Unit of Total Volatility

The Sharpe ratio asks how much return you earned above a risk-free baseline, per unit of total volatility. The formula, per period, is:

• Sharpe (per period) = (mean return − risk-free rate) ÷ standard deviation of returns

Three pieces matter here. The mean return is the average of your periodic returns. The risk-free rate is what you could earn with no risk (a short-dated government bill, say), de-annualized to match your period — the calculator divides the annual rate by the number of periods per year rather than compounding it. Subtracting it gives your excess return: the reward for taking risk at all. The denominator is the sample standard deviation of every return in the series.

Because the denominator is total volatility, the Sharpe ratio penalizes all variance equally. A month you were up 15% adds just as much to the standard deviation as a month you were down 15%. Sharpe treats a violent upside spike as “risk” in exactly the same way it treats a loss — which, for most traders, does not match how risk actually feels.

The Sortino Ratio: Punishing Only the Downside

The Sortino ratio fixes the asymmetry that bothers most people about Sharpe. It keeps the same numerator — excess return — but swaps the denominator for downside deviation, which only counts returns that fell below the risk-free rate:

• Sortino (per period) = (mean return − risk-free rate) ÷ downside deviation
• Downside deviation = √( Σ min(0, return − risk-free)² ÷ N )

Only the periods where you underperformed the risk-free rate contribute to the sum; every up period is treated as a zero, not as risk. Note that the sum is still divided by the total number of periods N, not just the count of losing periods — that is the standard convention and it keeps the ratio comparable across series.

The consequence: a strategy with big upside spikes and small, controlled losses will show a much higher Sortino than Sharpe. The upside that inflated its standard deviation (hurting Sharpe) is invisible to downside deviation (helping Sortino). If the two ratios are close, your returns are roughly symmetric. If Sortino towers over Sharpe, your gains are lumpy and your losses are tame — usually a profile traders like.

Annualizing the Ratios

Raw per-period ratios are hard to compare across daily, weekly, and monthly data, so both are annualized by multiplying by the square root of periods per year — √252 for daily, √12 for monthly, and so on. The same √time rule scales your volatility for reporting. This is why the calculator asks you to pick a frequency: it needs to know the annualization factor.

Worked Example

Take twelve monthly returns from a strategy: +4, −2, +6, +1, −3, +5, +2, −1, +3, 0, −2, +7 (all in %), with a 0% risk-free rate for simplicity and monthly frequency. The average monthly return is 1.667%, the sample standard deviation is 3.37%, and the downside deviation — built only from the four losing months (−2, −3, −1, −2) — is just 1.22%.

Metric	Value	How it is built
Mean monthly return	1.667%	average of the 12 returns
Standard deviation	3.37%	sample std dev of all 12
Downside deviation	1.22%	only the 4 sub-zero months
Annualized return	20.0%	mean × 12
Annualized volatility	11.66%	std dev × √12
Sharpe ratio	1.71	(1.667 ÷ 3.37) × √12
Sortino ratio	4.71	(1.667 ÷ 1.22) × √12

The gap tells the story. A Sharpe of 1.71 is respectable. But the Sortino of 4.71 reveals that this strategy’s volatility is overwhelmingly upside volatility — the +7 and +6 months blow out the standard deviation, dragging Sharpe down, while the actual losses are shallow. A trader who only glanced at Sharpe would underrate how well this strategy controls the downside.

How to Use the Sharpe & Sortino Calculator

The tool needs a return series and two settings:

1. Periodic Returns (%, comma or newline separated) — paste your list of returns, one per period, in percent. Commas or line breaks both work, so you can copy a column straight out of a spreadsheet. Enter at least two; the tool shows how many periods it parsed.
2. Frequency — choose Daily (252/yr), Weekly (52/yr), Monthly (12/yr), Quarterly (4/yr), or Annual (1/yr) to match your data. This sets the √time annualization factor.
3. Risk-Free Rate (annual %) — the annual return you could earn risk-free. The tool de-annualizes it by simple division and subtracts it from each period’s return to get excess return. Enter 0 if you want the raw ratios.

The results panel returns the Sharpe Ratio, the Sortino Ratio, the Annualized Return, and the Annualized Volatility — enough to compare any two strategies on both raw and risk-adjusted terms.

Reading the Numbers Honestly

As a rough guide, a Sharpe or Sortino above 1 is acceptable, above 2 is very good, and above 3 is excellent. But those thresholds vary by asset class and time horizon, and the single biggest trap is sample size. A handful of returns can swing wildly on one outlier, so a short or lucky run can produce a misleadingly high ratio; aim for at least 20–30 periods before you trust a reading, and always compare strategies over the same period length and frequency. A daily Sharpe and a monthly Sharpe are not the same units.

These ratios also describe the smoothness of the ride, not the depth of the worst hole — for that, pair them with a drawdown and recovery analysis, which shows how far equity fell and how long it took to come back. And risk-adjusted return is a diagnostic, not a position-sizing rule: how much to stake on each idea still comes from your position sizing and risk management framework. If your interest is the loss side specifically — how much you could lose at a given confidence level rather than the return-per-risk score — the companion Value at Risk guide covers that measure directly.