Streaming Wars Victory: Data Driven Decision Making

Jessica Joy
Posted on Feb 7, 2021

GitHub


 

In an age when your only plans are to sit on your couch and watch the newest release on a Friday night, you might end up scrolling through too many streaming platforms before you give up and finally throw your remote across the room.  

With endless possibilities at our fingertips, what we often struggle with these days is actually finding something to watch. Sometimes in the same amount of time it takes to pick a movie, you could have already been halfway through it. Indecisiveness and frustration has led to me shutting off the TV rather than flipping through different streaming services with thousands of movies in each. It's even worse when you know what you want to watch and do a quick Google search just to see that you’ll have to pay extra because you don’t have the service it is offered on. For most, it is not feasible to have a subscription to every streaming service when you cannot possibly take advantage of everything they have to offer.

To limit disappointment, delay, and wasted dollars we should invest in the streaming platform that best caters to our needs.

The Data

To tackle this problem, I used a Kaggle data set containing 16,744 observations of movies offered on popular streaming services including Netflix, Hulu, Prime Video and Disney +.  My initial thought was to look through the different offerings and come to a conclusion of which is the optimal streaming service. However, there’s no way to pick the best one. Everyone will have their own optimal service based on their preferences. Therefore, I analyzed this data using R Shiny to create data visualizations that will allow the user to get a clearer idea of which service is best for them.

Service

# of Obs. in Data

Actual # Offered

Prime Video 12,3504 12,828  (June 2020)
Netflix 3,560 3,781 (July 2020)
Hulu 903 2,700 (Dec 2020)
Disney+ 564 500+

Factors Considered

  • Popular Rating Sites
  • Country of Production
  • Age-Based Ratings
  • Genre Distribution

 

Analysis

To start off, I used R packages dplyr and ggplot2 to create density plots of ratings on Rotten Tomatoes and Imdb across the platforms. According to rottentomatoes.com, they use a scale better known as the “The Tomatometer” which is based on the opinions of hundreds of movie and TV show critics and the website gives a number out of 100 which represents the percentage of positive reviews for the movie. If it is above 60%, the movie awarded a red tomato score or considered fresh status; below 60% is given a green splat and considered rotten status. 

According to Imdb.com, their ratings are based on their millions of registered users who can submit a rating from 1 to 10 and the votes are then aggregated into a single IMDb rating. A weighted calculation may be implemented in order to preserve the authenticity of the rating.

It is important to note the differences in each service's density plot based on whether it was from a movie critic or a user based rating. Disney+ had the lowest ratings in Rotten Tomatoes, critic ratings, but also the highest on IMDb, user ratings. The opposite occurred for Hulu, having the highest Rotten Tomatoes ratings and one of the lower densities in high IMDb scores. In addition, even though Prime Video has the highest count of movies, by both rating systems they come up on the lower end, indicating the quality of the movies might not be the best. This would be valuable to take into consideration when choosing a streaming service. Many people religiously check these sites before seeing a movie so your preference on whether you trust critic ratings or user ratings is necessary.

The next factor observed was the country of production for each movie. As shown below, I used Googlevis to create an interactive map that shows the count of movies each platform offers in each country. After choosing a service the user can hover over each country to reveal a count of the number of movies produced in that country. While these are U.S. based services so the largest concentration is the United States for all, the services differ in international offerings. Prime Video showed the largest number of movies produced internationally while Disney + offered very few. 

 

Moreover, to evaluate the age-based ratings given to movies on each platform, I used ggplot2 to create pie charts of the distribution across each service. Netflix's distribution shown below displays that about 50% of the ratings given to their movies were 18+ and about 25% were 7+ and all. Hulu and Prime Video's chart resembled this as well. On the other hand, the Disney+ pie chart showed that the majority of their movies are for a large audience and definitely the choice if you are looking for family friendly options. 

 

Lastly, to analyze the genres offered by each service, I used ggplot2 to create bar plots of the number of each service's movies in a certain genre as a percentage of that service's total movies offered. Using the selector input, the user can toggle through 27 different genres and see the percent each service offers for that genre. This can be especially useful for niche genres that do not normally have a large presence on these services.

 

Difficulties faced:

  • This dataset is about 7 months old so it is not an exhaustive list of all that these services have to offer. It would be difficult to find a completely extensive list since these services are being updated frequently, sometimes even weekly.
  • There were numerous missing values throughout the data so I was forced to disregard these values within each category

 

Further research:

  • In the future, I would like to update this app in a way that it can be used as a tool for users. I would implement a quiz that a user can take and use all these factors to calculate the most accurate answer of which service is best for them based on their preferences in each category
  • Many of the variables I analyzed were categorical however it would be valuable to evaluate different numerical values such as individual service revenue or high vs low budget movies in order to find more trends

Resources:



About Author

Jessica Joy

Jessica Joy

Recent graduate from Binghamton University with a Bachelor of Science in Financial Economics. Highly motivated problem solver seeking opportunities to leverage data wrangling and analysis skills to provide key insights in real-world business problems.
View all posts by Jessica Joy >

Leave a Comment

No comments found.

View Posts by Categories


Our Recent Popular Posts


View Posts by Tags

#python #trainwithnycdsa 2019 airbnb Alex Baransky alumni Alumni Interview Alumni Reviews Alumni Spotlight alumni story Alumnus API Application artist aws beautiful soup Best Bootcamp Best Data Science 2019 Best Data Science Bootcamp Best Data Science Bootcamp 2020 Best Ranked Big Data Book Launch Book-Signing bootcamp Bootcamp Alumni Bootcamp Prep Bundles California Cancer Research capstone Career Career Day citibike clustering Coding Course Demo Course Report D3.js data Data Analyst data science Data Science Academy Data Science Bootcamp Data science jobs Data Science Reviews Data Scientist Data Scientist Jobs data visualization Deep Learning Demo Day Discount dplyr employer networking feature engineering Finance Financial Data Science Flask gbm Get Hired ggplot2 googleVis Hadoop higgs boson Hiring hiring partner events Hiring Partners Industry Experts Instructor Blog Instructor Interview Job Job Placement Jobs Jon Krohn JP Morgan Chase Kaggle Kickstarter lasso regression Lead Data Scienctist Lead Data Scientist leaflet linear regression Logistic Regression machine learning Maps matplotlib Medical Research Meet the team meetup Networking neural network Neural networks New Courses nlp NYC NYC Data Science nyc data science academy NYC Open Data NYCDSA NYCDSA Alumni Online Online Bootcamp Online Training Open Data painter pandas Part-time Portfolio Development prediction Prework Programming PwC python python machine learning python scrapy python web scraping python webscraping Python Workshop R R language R Programming R Shiny r studio R Visualization R Workshop R-bloggers random forest Ranking recommendation recommendation system regression Remote remote data science bootcamp Scrapy scrapy visualization seaborn Selenium sentiment analysis Shiny Shiny Dashboard Spark Special Special Summer Sports statistics streaming Student Interview Student Showcase SVM Switchup Tableau team TensorFlow Testimonial tf-idf Top Data Science Bootcamp twitter visualization web scraping Weekend Course What to expect word cloud word2vec XGBoost yelp