Incubus Billboard History & Spotify Analysis

Billboard History

Background

My favorite band of all time is Incubus. They’ve released many albums over more than two decades. I thought it’d be interesting to plot Incubus’ history on various Billboard charts.

First analysis

To start, I wanted to see which albums charted, what their peak position was, and how many weeks they were on a Billboard chart. Initially, I wasn’t concerned with any chronological information.

I created the plot below to show what Billboard chart each album appeared on (bubble color), as well as how high the album charted (position on the x-axis) and how many weeks the album was on the chart (bubble size).

*The below plot is interactive. You can hover over each data point to see more detailed information.

What stood out?

The Make Yourself Billboard 200 data point immediately stood out to me. Make Yourself spent more time on the Billboard charts than any other Incubus album, 98 weeks, but only peaked at #47.

Possible explanations/additional context?

This is an example of when context is important. Make Yourself was most people’s first introduction to Incubus, so its release wasn’t highly anticipated as was the case for future albums. People likely discovered the album at different times. This data point supports this hypothesis and indicates that, once discovered, the album was well-received.

Further interpretation

As the years and albums continued, the band’s popularity grew. Their next several albums hit peaks at or near #1, perhaps the result of an anticipated album release. More recently, their albums have charted highest on the Hard Rock and Top Rock Album charts, two charts that didn’t exist before 2006/2007. While Incubus’ newer music may not reach as broad of an audience, they still have a fanbase that clearly keeps listening.

Further exploratory analysis

I then decided to play with the chronological dimension of the data. A timeline of Incubus’ Billboard history could shed light on the interrelationships of their albums - did the release of one cause people to listen to others?

The below plot is a static plot that showcases some of the customization that can be done more easily with a static image (i.e. label placement to prevent overlapping text).

Change of emphasis

While this plot contains much of the same information as the plot above, the addition of time and the deletion of # of weeks on the chart changes the emphasis of the plot and shifts the focus of the reader to a chronological view.

Spotify Analysis

Background

As someone who knows Incubus albums well, I was curious if Spotify’s categorization of each studio album’s emotion was in-line with what I remembered. To do this, I used the Spotipy library to access Spotify’s API and relied on the API’s official documentation to understand the different metrics for each track.

Incubus studio albums are ordered by release date with the oldest album at the bottom and the most recent album at the top.

First Analysis

Of the metrics reported by Spotify, I thought energy and valence (positiveness) would give a good sense of the overall emotion of an album. Not surprisingly, Morning View and If Not Now, When? had the lowest overall energy and positiveness. These two albums have a greater number of bleak, slow songs than Incubus’ other albums.

Incubus’ first album, Enjoy Incubus, only contained high energy songs, unlike future albums. This was also expected considering Incubus was younger and more punk-rock when this first album was made.

I wanted to incorporate a third Spotify metric to give a fuller picture of the emotion each album evoked - loudness. I also aggregated each track’s information to give median metric values by album making it easier to draw comparisons.

This plot better shows that valence has, for the most part, decreased with each album. Meanwhile, loudness and energy remained high across most albums, except for Morning View and If Not Now, When?

Second Analysis

Next, I was curious if I could glean which albums were more experimental. I decided to look at how many songs on each album were instrumental, composed in a minor key, or did not have the typical 4 time signature. While these metrics are not truly indicative of experimentation, they were the best ones recorded by Spotify.

Instead of looking at median values, I decided to plot the percentage of album tracks that were categorized as one of these three song types.

I was not surprised to see that Morning View had the highest number of instrumental songs, as that album is gloomier and slower than most. I was surprised to see that the albums deemed to have the lowest valence (positiveness) did not have the highest number of minor songs, which are often associated with a sadder sound.

Overall, I did not think this plot told a compelling story about any of the three metrics. There were no clear trends to indicate that Incubus’ writing style has changed over time.

The above plots were built entirely in Python. I used BeautifulSoup to pull Billboard chart data, the Spotipy library to access Spotify’s API, and Plotly to build the visualizations.