sigma = sqrt(Var(X)).
In finance this is volatility. Annualising assumes independent returns and uses the square-root-of-time rule: multiply a daily figure by sqrt(252) ≈ 15.9. A 1% daily standard deviation is roughly 16% annualised, which is a conversion worth being able to do instantly.
For a normal distribution, about 68% of outcomes fall within one standard deviation, 95% within two and 99.7% within three.
The trap. Those percentages assume normality. Financial returns have fat tails, so "a five-sigma event" occurs far more often than the normal distribution implies - which is why the phrase is usually a sign the model is wrong rather than that something remarkable happened.