No it isn't. If everyone responded that they are 50% skilled (or per this article, that it's randomly distributed), then we 1. See the same graph, and 2. Bad

No it isn't. If everyone responded that they are 50% skilled (or per this article, that it's randomly distributed), then we 1. See the same graph, and 2. Bad people overestimate, and good people underestimate This article merely describes Dunning Kruger. Accidentally proves it mathematically, but thinks that it debunks it.