Discover the Billboard Music Charts

 

Purpose

Billboard is one of the most important and famous music charts.  It provides lots of information for both listener and producer. The Billboard music chart used a straightforward cumulative total of yearlong sales and airplay points to calculate each piece’s  ranking position.  I wondered how are the Billboard ranking directly related to the artist's’ production. So I performed some simple analysis to test.

I decided to do a web scraping project to discover more information from this chart from 2006-2016.      

Workflow

I created eight different Python spiders to scrape different kinds of charts from Billboard. I scraped different genres such as country music, electric dancing music, rap music, rock music and r&b music. I also collect information about top hottest artists every year and the top ranking artist all time(since Billboard’s inception). I used the python seaborn package to visualize my result.

Result

I calculate the total quantities of each artist's’’ work that is ranking on the chart for the 2006-2016 period. I picked the top 20 artists from each chart to visualize more detail.

All Music (2006-2016)

Rihanna has a domain advantage for this decade. As we can discover from the graph, she has a total of 30 songs on the Billboard Top 100 songs from this time period.

The second graph shows us information about how many songs from the artists are ranked in a particular year so that we can discover how those artists’ work performed each year.

This word cloud shows the most popular words from all the top ranking song titles. That indicates which topics are popular in this decade.  

The gender chart shows if gender is a factor in appearing on the top 100 chart. It appears about evenly divided between male and female artists. The group category stands for music group or bands.

Top Artists (2006-2016)

This chart is showing how popular are these artists every year in this decade. We can see that Rihanna and Kenny Chesney are on the chart every year.

This swarm plot shows that top 20 popular artist ranking is very stable this decade.

All-Time Top Artist

I made a cross check with top artists (2006-2016) with all time top artist. We can see how many times the artist is ranked in the 06-16 period. The ranking column shows how the artist ranks in this all time chart. The count column shows how many time they were ranking from the 06-16 chart.

Rap Music (2006-2016)

R&B Music (2007-2016)

Rock Music (2008-2016)

Country Music(2006-2016)

Electronic Dance Music (2013-2016)

Total studio albums

Total Singles

Statistical Analysis (Simple Linear Regression Analysis)

It shows that the artist's production has no direct linear relationship with the Billboard chart ranking.

 

Conclusion

  1. Some artists dominate the whole decade, while others are just a flash in the pan.
  2. In some specific music genres, gender imbalance happened.
  3. Quantity is no guarantee of quality;  releasing more songs and albums doesn’t necessarily improve your odds for making the chart.

 

For the fully detailed code and the result, please check with my GitHub: https://github.com/nash13cxw/BillboardAnalysis.

Xiaowei Cheng
Xiaowei Cheng
Xiaowei holds a Bachelor’s degree in Finance and now is pursuing his Master’s in Biostatistics at George Washington University. Xiaowei discovered his passion in data through research projects in graduate school, where he worked with professors on analyzing health care data and drawing results from medical studies. Xiaowei is a passionate and savvy programmer with proficiency in Python, R, and SQL. During his time in bootcamp, he has been working extensively on projects involving data visualization, web scraping techniques, and machine learning.

Leave a Reply

Your email address will not be published. Required fields are marked *