Honest question: how do people interpret/use the partly-shaded area in plots like these that can be made easily in Seaborn and ggplot2? What does one take away from a point being in or out of the area? what other information can be gleaned?

Is it something like "given <some assumptions> about these points being <a sample> of <the output of some process>, and the line being <some real thing>, then there's some n% likelihood that the true value of the line lies in this area"?

The question behind the question is basically: what are the assumptions used to compute these areas, how often are these assumptions reasonable for a given chart, and do readers of any given chart know the answers to either of those questions? :)

Another possible interpretation: the shaded area contains n% of the lines that can be drawn by using a random sample of m% of the points?

@hadleywickham @michaelwaskom I'd love to hear your thoughts on this! I get many requests to implement similar features in Plotly Express and I'm trying to get an intuition about how these work/what they're for...

@JessicaHullman @mjskay I'd love your input into the above and/or commentary on the responses below if you felt so inclined :)