Data nerds discover that weirder is better.
Looking back at this year’s biggest songs – and aside from the obvious gluteal themes running through hits by Meghan Trainor, Iggy Azalea and Nicki Minaj – it’s hard to hear a strong similarity between the likes of ‘Fancy’, ‘Happy’ and ‘Dark Horse’
But researchers from Spotify’s data lab The Echo Nest reckon they’ve worked out the trick to having a hit single. In short, to make it to the top, a song should sound different from anything else on the charts, but no so different that it falls off the radar entirely.
The team analysed the audio attributes of over 25,000 songs on the Billboard 100 from 1958 to 2013 to create a new metric called Song Conventionality. They found that songs at the top of the charts are more likely to sound unconventional than songs in the middle.
Songs in the top 20 are the least conventional of any section of the Billboard Hot 100 over time. If a song is too weird it’s unlikely to make the charts at all, but within the charts it looks like the weirder the song, the better it performs.
The graph shows that the top songs of the past 60 years tend to be the least conventional of the top 10, and in turn, the top 10 are less conventional than the top 20. That certainly seems to apply to this year’s motley crop of chart-toppers – who could’ve predicted that a bubblegum country and hip-hop mongrel would take over the Billboard 100 for half the year?
“Nevertheless,” the team warns, “predicting hit songs is nearly impossible to do, because performance is largely contingent on a song’s relationship to other songs that are produced and released contemporaneously.”
Find out more about how Echo Nest calculated songs’ conventionality in the ‘Behind The Scenes’ part of Spotify’s report.
