My god, lol. His example of the 5-person room doesn't even reflect a basic understanding of what money is; it's no wonder why he infers what he does from those graphs. I've addressed a few of the fallacies of these graphs in other posts.
I'm going to ignore the ones that convey no real useful information. The average earnings one tends to be deflated with the standard CPI measure (which often overstates inflation); it also is a false measure. Compensation is the more accurate measure of earnings (click the post link for graph):
The 'shares of income' and 'income by quintile' graphs are tricky, and one needs to be careful not to squeeze too much meaning out of them. As Joseph Schumpeter
once said, "the upper strata of society are like hotels which are indeed always full of people, but people who are forever changing. They consist of persons who are recruited from below to a much greater extent than many of us are willing to admit." My linked post from above cites a Treasury mobility study that looks at this. I also avoid using household income measures due to changing structure, size, etc.