For n trials with success probability p, the number of successes has mean np and variance np(1−p). The standard deviation, its square root, shows how much the outcome varies.