Can data give you an edge in your office’s Oscar pool?

Feb 18, 2019 | Blog

Spoiler alert – probably not, but we can still have fun looking at some insights.

From a high level, this year’s Best Picture nominees are a very interesting selection of films:

• Five films are bio pics (six, if you count Roma)

• Two could be considered musicals

• Minorities are the main characters in more than half of the films

Black Panther is the first comic book movie to be nominated for best picture

So, with such a unique selection of films in the running, how do you know which one to pick? You go to the data. There are a lot of factors that go into what film is chosen as Best Picture: campaigning, politics, and the even the peculiar voting system itself. For simplicity I’m only looking at reviews and box office revenue. I did this by collecting review data from Rotten Tomatoes and IMDb and box office revenues from Box Office Mojo. From there, I inserted the datasets into Tableau and created a Best Picture nominee dashboard that shows how each film preformed according to critics, audience members, and at the box office both domestically and abroad. ​

Now we can go in and compare these films across multiple dimensions. But before we do, it might be helpful to take a step back and think about the Best Picture winners over the last several years (The Shape of Water, Moonlight, Spotlight, and Birdman). How did these films perform in terms of audience and critic reviews and box office sales? Well, they were all well received amongst audiences and critics, but they weren’t blockbusters.

Applying this past track record to our current list of nominees rules out Black Panther, Bohemian Rhapsody, and A Star is Born. I think we can likewise eliminateVicebased on the unfavorable reviews from both critics and audience members. I’m also taking The Favouriteoff my short list since it has the worst domestic box office performance. This leaves BlackkKlansman Green Book, and Roma. Based on the data, I’m going with Green Book—a movie I haven’t even seen myself. Here’s why.

IMDb breaks down ratings by age. Almost all films follow the pattern of getting worse ratings as the age of the reviewer increases. However, Green Book bucks this trend with its best ratings coming from the age group of 45 and up. Given that the Academy members’ median age is around 62, this could bode very well for Green Book. Green Book also performed well at the box office, doing just as well as BlackkKlansmanand better than Viceand The Favourite.

Based on the age and box office trends in the data, I’m going with Green Book. However, there is a gaping hole of missing data that might affect this prediction. Because Roma is a Netflix film, we don’t have good numbers on box office performance or viewership. How do we handle missing data? Well, in real life you may try to impute the values, drop that variable, or create a separate category altogether for when data is missing. In this case, because I’m not making a model, and there is just a single entry with missing data, I’m leaving it as is, making Roma the dark horse of the nominees.

Final bet, Green Book for best picture, with Roma being a possible upset. But feel free to play around with the dashboard and see what strikes you as insightful.