I agree averages can be misleading. As the joke goes when Bill Gates walks into a bar suddenly everybody's a billionaire, on average! At the same time medians can also be misleading. We have 50 people earning $10, 1 earning $100, and 50 earning $200. We swap to a system where we have 50 people earning $90, 1 earning $100, and 50 earning $300. There would have been a major shift upward with nearly everybody seeing major increases in earning, yet the median would not have shifted at all.

This [1] is the real median personal income. The data there only starts at 1974 but you once again see a 32% increase in income. Now factor in the change in hours worked. The average American works more than 100 less hours then back then. [2]. These numbers combined along with arguing that most people only saw a real increase in wages of 12% is simply not possible, nor is it possible to simply attribute all growth to the rich.

Now there is this [3]. The numbers from that paper are really what made me start digging into all of this stuff. To give the long and short of it - the poor are becoming middle class, and the middle class are becoming rich. With the net effect being a major decline in the number of poor, a major increase in the number of rich, and a small decline in the number of middle class. Probably not coincidentally, not entirely dissimilar to the hypothetical I proposed where the median can end up being misleading. These 'nobody except the rich are seeing more money' articles seem to be simply untrue, but they are click magnets.

[1] - https://fred.stlouisfed.org/series/MEPAINUSA672N

[2] - https://fred.stlouisfed.org/series/USAAHWEP

[3] - https://news.ycombinator.com/item?id=16952930