Article · Backtesting & Walk-Forward Testing

Sortino vs Sharpe in a Backtest

Sharpe punishes all volatility; Sortino punishes only the downside. Here is how the two differ in a backtest and where each one can mislead you.

Published June 16, 2026 · Primary topic: sortino vs sharpe

← Back to the Backtesting & Walk-Forward Testing hub

Sharpe and Sortino are both reward-per-risk ratios, and both turn a series of returns into a single number you can compare across strategies. They differ in one respect: what they count as risk. That difference is small in arithmetic and large in consequence, and a trader who quotes one without understanding the other can be flattered into trusting a strategy they should question.

What Sharpe measures

Sharpe divides a strategy's excess return by the volatility of all its returns — up moves and down moves alike. Its assumption is that variability itself is the risk. A strategy with sharp upside swings is penalised the same as one with sharp downside swings, because both add to the volatility in the denominator. That is sometimes fair and sometimes harsh.

What Sortino changes

Sortino keeps the same idea but counts only downside volatility — the moves below a target, usually zero. Upside swings no longer count against the strategy. The reasoning is intuitive: a trader does not lose sleep over a violently good day. By ignoring upside variance, Sortino tends to produce a higher number than Sharpe for the same return series.

Where Sortino can flatter

That higher number is precisely the trap. A strategy with rare, enormous gains and a steady drip of small losses can post a handsome Sortino while still being fragile — the upside it is rewarded for may depend on a handful of outlier days that will not repeat. Sortino measures less, so it hides more. It is a useful lens, not a better one.

Read them alongside drawdown

Neither ratio describes the worst moment you would have lived through. Two strategies can share a Sortino and differ enormously in their deepest drawdown. That is why a single ratio is never enough: read reward-per-risk and peak-to-trough pain together, and demand both hold up out-of-sample rather than on one tuned window.

For the broader pattern of metrics that flatter a strategy into looking ready, see backtesting myths that cost money, and learn how to spot overfitting in a strategy before either ratio convinces you. A strong ratio is permission to investigate further, never to go live.

Important

This is not investment advice.

GreatDane Trades is an education, backtesting, and trading automation platform. Nothing on this site is financial advice. Results are simulated. Backtests do not guarantee future results. Markets can diverge from simulations. Trading cryptocurrencies involves substantial risk including the total loss of capital. Paper trading should come before live trading. Users are responsible for their own trades.

Read the full risk disclaimer →

More in Backtesting & Walk-Forward Testing

Closely related guides in this topic cluster.

See every guide in this cluster on the hub: Backtesting & Walk-Forward Testing.

Explore the other pillars

Put the discipline into practice — in paper first.

Start in paper mode, validate with walk-forward backtests, and let the risk engine hold the line. No real capital is at risk until you decide to connect Kraken.

No profit promises. Paper trading by default.