[JackFr]

This is pretty muddled thinking. He talks about categorical variables and then points out that they are not normally distributed. Which doesn’t make sense. (It almost could, but he skips an important step.)

Apart from that it's mostly handwaving and opinion with no actual data. When you don't use data it's unsurprising that your personal experience matches your hypothesis.

[ErikVandeWater]

Yes, he just chose a satisfactory cutoff for whatever counts as a "true" flagship. Some people would say only this generation iPhone. Some would say only this generation iPhone + Pixel + Galaxy. Some would say this generation, and the previous generation of iPhone + Pixel + Galaxy.

And even then, we're ignoring features like internal storage capacity for each phone.

Even if you are making $10 million a year, you split that between your yacht, your car, your mansion, and jewelry for the wifey.

You aren't going to have the biggest mega-yacht and no jewelry for the wifey, so there will be a tapering at the higher end of the demand curve for yachts for people making $10 million a year.

[nkozyra]

Yeah, I felt like there is perpetually a point the author is about to get but then just moves on to the next thought. Kind of frustrating to read, honestly.

[febeling]

True. Still an interesting idea, even if it needs verification (or falsification).

[coldtea]

>He talks about categorical variables and then points out that they are not normally distributed.

Smartphone prices and city populations are not categorical variables...