Sortino Ratio
Learn how to calculate and interpret the Sortino ratio for crypto portfolios using minimum acceptable return and downside deviation.
Reading progress — saved on this device
The Sortino ratio compares average return above a chosen minimum acceptable return with downside deviation, penalising harmful shortfall rather than all volatility equally.
Learning objectives
- Calculate downside deviation relative to a stated minimum acceptable return.
- Explain the difference between Sortino and Sharpe ratios.
- Recognise how target return, sample size, liquidity and tail events affect interpretation.
What it is
The Sortino ratio modifies the familiar risk-adjusted-return framework by counting only returns below a minimum acceptable return, or MAR, as downside risk. Positive volatility above the target is not penalised in the denominator.
The MAR is part of the definition. It may be zero, a periodic risk-free rate, an investor hurdle or another stated target. Changing it changes both the excess-return numerator and which observations count as shortfalls.
This makes Sortino useful when upside variability is not considered harmful, but it does not make the measure universally superior to Sharpe. The two metrics answer different questions.
How it works
For each periodic return, calculate the shortfall relative to MAR. Positive excess returns contribute zero to downside deviation; negative shortfalls are squared. The mean squared shortfall is then square-rooted.
Conventions can differ on whether the denominator averages across all observations or only downside observations. A report should state its convention so ratios are comparable.
Sortino becomes unstable when there are very few downside periods. An illiquid or smoothed strategy may appear to have tiny downside deviation even though liquidation or gap risk is substantial.
Annualisation needs the same caution as Sharpe. Scaling a periodic ratio by the square root of time assumes sufficiently well-behaved return dynamics; crypto volatility clustering, serial correlation and fat tails can weaken that assumption.
Measurement methodology
| Check | Purpose | What to verify |
|---|---|---|
| MAR | Defines acceptable return | State whether it is zero, risk-free or another hurdle. |
| Downside convention | Defines denominator | State whether averaging uses all periods or downside periods only. |
| Sample quality | Tests stability | Count downside observations and inspect stale prices. |
| Tail analysis | Shows what Sortino misses | Review maximum drawdown, liquidation, skew and worst-period loss. |
Worked example and thought exercise
A portfolio earns an average 1.0% per month. The chosen monthly MAR is 0.2% and measured monthly downside deviation is 3.2%. Monthly Sortino = (1.0% − 0.2%) ÷ 3.2% = 0.25.
The number is meaningful only with the stated MAR and downside-deviation convention. A higher MAR would reduce the numerator and generally create more downside observations.
Thought exercise: how could a strategy with many smooth positive months and one rare liquidation event show an attractive pre-crash Sortino ratio?
Common mistakes and practical workflow
- Comparing Sortino ratios calculated with different MARs or denominator conventions.
- Assuming upside volatility has no relevance simply because Sortino does not penalise it.
- Trusting a ratio based on only a handful of downside observations.
- Ignoring stale prices, leverage, liquidity and tail-loss mechanisms.
Practical workflow
- Define reporting frequency and return series.
- Select and disclose the periodic MAR.
- Calculate downside deviations using one documented convention.
- Calculate the periodic Sortino and annualise only with stated assumptions.
- Interpret it alongside Sharpe, drawdown, skew, liquidity and leverage metrics.
Knowledge checkpoint
- What does the Sortino ratio penalise in its denominator?
- How does MAR affect the calculation?
- Why can a strategy with few downside months have an unstable Sortino?
- What major risk can remain hidden even when Sortino is high?
FAQs
❓ Is Sortino always better than Sharpe?
No. Sortino focuses on downside shortfall; Sharpe measures total volatility. Both can be useful when interpreted correctly.
❓ What should MAR be?
There is no universal choice. Use a clearly stated target appropriate to the portfolio and reporting currency.
❓ Can Sortino be negative?
Yes. If average return is below MAR, the numerator is negative.
❓ Does Sortino capture liquidation risk?
Only if liquidation losses are actually represented in the return sample. Scenario and tail analysis remain necessary.
Summary
The Sortino ratio focuses risk adjustment on returns below a stated target. Its usefulness depends on a transparent MAR, consistent downside-deviation methodology and enough adverse observations to make the denominator credible.
Want this in a personalised order?
Take the crypto assessment and get a custom path of 10 modules matched to what you already know. Free, no card required.
Build my path →