Evaluating Hollywood (1980 - 2020)

With over 500,000 movies in the world, is the movie industry still growing?

Comment

About Evaluating Hollywood

By Raj Chaphekar (AB), Ethan Hu (AB), Angelina Lum (AC)

Inspiration

Although movies aren’t necessarily world-ending or life-changing by themselves, our team wanted to learn how the movie industry is growing. We know that the pandemic left a devastating blow on the movie industry overall. With theaters closed, there were not as many new releases and many movies were delayed. So we were wondering if movie production companies/distributors are still making a profit, and if so, what factors weighed into their high-grossing movie(s) or how has those differences changed over time.

What it Does

  • Creates visualizations of the individual factors that affect how well a movie performs
  • Creates catplots of the average finances (budget, gross, and profits) of the top 10 profitting productions (company, director, and writer)
  • Creates a relational plot that shows the relationship between the year movies were distributed and their profits
  • Looks at how accurate IMDb scores were of different models based on the combination of different factors that affect a movie

How We Built It

We did all of our coding in a github repository using python and a single excel file for testing. Along with the repo, we have a detailed writeup of all our findings here. As for the code, first we compared the best fit lines for numerical variables (change in score / change in variable) and found the range for categorical variables, such as the difference between high and low scoring movies. We then found the profit of each movie by subtracting the budget from the gross. Then we compared the average budgets, gross, and profits of individual companies, directors, and writers in a catplot and created a relational plot to see the year vs. profit trend. Finally, we used an additional dataset of movies from 2021 and built a machine learning model to compare the mean squared error of movie IMDb scores.

Challenges We Ran Into

  • Using ed workspace was not viable due to memory issues
  • Having to rewrite the majority of our code halfway through the process

Accomplishments That We're Proud Of

  • To collaborate on Github with no major conflicts or bugs
  • To grow our confidence and learn more about new libraries

What We Learned

That the perfect Hollywood movie production would be produced by the Marvel Studios company, directed by Anthony Russo, and written by Irene Mecchi.

What's next for Evaluating Hollywood (1980 - 2020)

  • Analyze the effects on profit from other variables such as movie star
  • Expand usage of the plotly library to online graphs for easier access
  • Add new data to the 2021 dataset as movies are made & compare model again
  • Run model for profit instead of score

Built With

Share this project:

Updates