For me, this is one of those things where the distributions have different skew. I think the median example of less simple writing is indeed worse than the medi

For me, this is one of those things where the distributions have different skew. I think the median example of less simple writing is indeed worse than the median example of simple writing, but that the graphs cross at some point to the right of that, such that better examples of less simple writing are better than better examples of simple writing, and great examples of less simple writing are much better than great examples of simple writing. (I recognize that this is all subjective, but this is how I think about it.) The genre also matters a lot. For sure I think simple writing is the better choice for Graham's topics, but I think it's much more open for fictional writing, and that some kinds of non fiction essays also benefit from less simplicity (again, keeping in mind the distribution effect discussed above). For instance, one of my favorite essays is David Foster Wallace's "This Is Water". It is not simple, but it is much more effective than an essay using simple language to make the point that we often don't realize the things that we are immersed in. But I come across examples of this by good writers who do not usually use simple language who I believe are usually more compelling than writers like Graham.