Probability histogram

A probability histogram looks like a histogram of data, except the bar heights are probabilities instead of counts, so the heights add to 1. Rolling a fair die gives six bars of equal height $1/6 \approx 0.1667$, and $6(1/6) = 1$. The mean of the random variable is the balance point of the picture, which for the die is $\mu_X = 3.5$ (mu-X, the mean of X). You read shape from it the same way you read shape from data: look for symmetry, skew, and gaps.