I guess it would depend on how you sample the music (i.e. from the top 50 hits playlists, randomly, by streaming number, by sold copies, etc.) and where you establish the baseline, and again how you compute forumlaic.

So if you are comparing top hits today with experimental rock of the 70s, and measure it in variance of chords, timbre and vocabulary. Yes it would be easy to show how music is becoming less varied. However if you establish the baseline during the late classical era (and limit your self to western music; as is often done) I’m sure you will find music today to be more varied.

If you sample randomly and make sure to include all of the experimental genres I’m sure you will find music today more varied then ever, and even if you go by top sales (and make sure you include music from around the word) I sure you might find that music is just as varied as it was back in the 70s.

Then there the question of how you measure musical variance. It is easy enough to do it by measuring (among other) the chord progression, or timbre, or proportion of the chorus, etc. but when people do this they often undermine many genres of music (e.g. minimalist music of the 80s and 90s) or hip hop, etc.

Max Martin is also a fantastic musician, originally a singer (this is arguably the most important part of pop), and has a deep understanding of music. From a harmony standpoint, his stuff tends to be a bit more interesting than most pop tunes. The dude is beast.

https://en.wikipedia.org/wiki/Max_Martin_production_discogra...

Yes, less formulaic. Classical music is far more formal than any sort of pop music and can quite literally be algorithmically generated.