# Sortino Ratio

Return per unit of downside risk, ignoring upside swings.

Source data: AMFI daily NAV (17,900+ schemes) + Nifty benchmark indices. Last updated: 2026-07-02. Interactive tool: https://mfpro.tigzig.com

## What is the Sortino Ratio?



The Sortino Ratio is like Sharpe but only penalizes downside volatility. It ignores
 "good" volatility (days when the fund goes up more than expected) and only counts
 "bad" volatility (days when the fund falls below the risk-free rate). This makes it a
 better measure for investors who care about losses more than total variability.

 


## How We Compute It


 
 ret_minus_rf = fund_daily_return − (6% / 252)

Downside deviation = √(mean of: if ret_minus_rf < 0 then ret_minus_rf², else 0)

Sortino = (Mean daily fund return − rf_daily) × 252 / (Downside dev × √252)
 

Key difference from Sharpe: instead of total std dev, we use downside deviation -
 the RMS of only the negative excess returns, with zeros for positive days. This means
 the denominator includes all days in the count but only penalizes negative ones.

 


## Sortino vs Sharpe



When they diverge


Fund X: Sharpe = 0.8, Sortino = 1.5


The big gap tells you this fund has a lot of upside volatility that Sharpe is
 penalizing but Sortino is not. The fund frequently has large positive days, which
 inflates total std dev but doesn't hurt investors.


Fund Y: Sharpe = 0.8, Sortino = 0.9


Similar Sharpe and Sortino means the volatility is roughly symmetric - the fund has
 bad days about as often and as severely as good days.

 
 


## How to Interpret





- Sortino ≥ 1.0: Good downside-adjusted returns (shown in green)

- Sortino is usually higher than Sharpe (because denominator is smaller)

- Sortino is especially useful for comparing funds with asymmetric return profiles


 


## Validation



We validated all risk metrics by independently computing them in Python (pandas + scipy)
 from raw NAV data, then comparing against the API output. 104 out of 108 precomputed
 checks passed within tolerance. The 4 differences were in the 1Y period boundary
 (1 observation difference due to CURRENT_DATE evaluation timing). Custom date range
 validation showed zero differences across all 18 metrics checked.

## Related metrics

More Risk Metrics methodology from the MFPRO analytics tool:

- [Beta and R-Squared](/mfpro/beta-and-r-squared)

- [Alpha and t-statistic](/mfpro/alpha-and-t-stat)

- [Sharpe Ratio](/mfpro/sharpe-ratio)

- [Tracking Error and Information Ratio](/mfpro/tracking-error-information-ratio)

- [Capture Ratios (Upside and Downside)](/mfpro/capture-ratio)

- [Win Rate](/mfpro/win-rate)

---
Source: https://www.tigzig.com/mfpro/sortino-ratio