Adjust the mean (average) value of a dataset by weighing each value based on its z-score. Values closer to the mean are considered "better" and so are given more weight.
Very useful for giving less weight to "troll votes" in rating systems.
Pass an array of values to adjustedMean().
For example let's say they are IMDb ratings:
// All the `1` ratings are considered more illegitimate than the ones closer
// to the mean (≈6.3)
const ratings = [
10, 10,
9, 9, 9, 9, 9, 9,
8, 8, 8, 8, 8, 8, 8, 8, 8, 8,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
6, 6, 6, 6, 6, 6, 6, 6,
5, 5, 5, 5,
4, 4,
3,
2,
1, 1, 1, 1, 1, 1,
]
mean(ratings) // ≈6.3
adjustedMean(ratings) // ≈6.8