The Sharpe ratio divides excess return (return above the risk-free rate) by the standard deviation of returns. Higher is better. That much is simple. What gets complicated is applying it to real trading decisions where the assumptions behind the formula start to break down.
A Sharpe ratio of 1.0 is generally considered acceptable. Above 2.0 is strong. Above 3.0 over any sustained period is exceptional and should probably make you suspicious. Backtested Sharpe ratios above 3.0 almost always degrade in live trading because they reflect overfitting, favorable market conditions, or both.
The time period matters enormously. A strategy can have a Sharpe of 2.5 over three years and 0.8 over ten years. The shorter period might have captured favorable conditions that flattered the result. Annualized Sharpe ratios calculated from daily data will differ from those calculated from monthly data because of how volatility aggregates across different frequencies. Daily Sharpe ratios tend to be higher because they capture less of the tail risk that shows up in monthly or quarterly data.
One fundamental limitation is that the Sharpe ratio treats upside and downside volatility the same. A strategy that produces occasional large gains and small consistent losses can have the same Sharpe as one that produces consistent small gains and occasional large losses. These are very different risk profiles, and the Sharpe ratio does not distinguish between them. The Sortino ratio addresses this by only penalizing downside deviation, which is often more appropriate for evaluating trading strategies.
Leverage distorts Sharpe comparisons. A strategy running at 2x leverage will have roughly the same Sharpe ratio as the unleveraged version (since both return and volatility scale proportionally), but the risk experience is completely different. Two strategies with identical Sharpe ratios can have vastly different maximum drawdowns depending on their leverage.
Serial correlation in returns can inflate the Sharpe ratio. If a strategy has positively autocorrelated returns (today's gain makes tomorrow's gain more likely), the measured volatility underestimates the true risk, and the Sharpe looks better than it should. This is common in illiquid strategies, trend-following, and some momentum approaches.
For crypto specifically, the risk-free rate choice matters. Using US Treasury rates when evaluating a crypto strategy denominated in a volatile base currency introduces distortions. Some practitioners use stablecoin lending rates as the crypto-native risk-free rate, which changes the calculation meaningfully.
The practical way to use Sharpe ratios is as a relative comparison tool within the same asset class, time period, and frequency. Comparing the Sharpe of a crypto momentum strategy against a bond carry strategy is largely meaningless. Comparing two crypto momentum strategies over the same period with the same rebalancing frequency gives you useful information about which one extracts more return per unit of risk.