They also outline their methodology:Several multi-decadal trends have been uncovered. For example, there is a clear downward trend in ‘happiness’ and ‘brightness’, as well as a slight upward trend in ‘sadness’. Furthermore, songs are becoming less ‘male’. Interestingly, successful songs exhibit their own distinct dynamics. In particular, they tend to be ‘happier’, more ‘party-like’, less ‘relaxed’ and more ‘female’ than most.
Seems legit! Amongst the study's interesting findings, it observes a downward trend in the popularity of rock, and confirms pop and dance as the most popular genres. It also notes that even though 'sadder' songs are being released more frequently, 'happier' songs are still most popular. Speaking to the Associated Press, co-author Natalia L. Komarova said, "So it looks like, while the overall mood is becoming less happy, people seem to want to forget it all and dance."It's an interesting observation, and with the simultaneous rise of sadness as an aesthetic online (as exemplified by Twitter accounts like @sosadtoday, which at the time of writing has 665,000 followers), it's one that seems to make sense. We look for music that reflects the time we're living in and what we're feeling (either for identification or catharsis,) and it is pretty sad around here these days, so it follows that even our pop music is getting more somber. Science, thank you.Follow Lauren on Twitter.We used random forests to predict the success of songs, first based on their acoustic features, and then adding the ‘superstar’ variable (informing us whether the song’s artist had appeared in the top charts in the near past). This allowed quantification of the contribution of purely musical characteristics in the songs’ success, and suggested the time scale of fashion dynamics in popular music.