Projects submitted to the Data Journalism Awards competition

Right here you will find a list of all the projects submitted to the Data Journalism Awards competition.  

Two theaters or 1,000? How to release an Oscar-winning film
Country: United States
Organisation: MarketWatch
Data visualisation of the year
InteractiveVisualisationBreaking newsArts
Applicant
William
Davis
Team Members
Evie Liu, Sarah Squire
Project Description
This project sought to introduce the concept of film release strategies to a broader audience and determine which formula is used most by Best Picture winners.The two main release strategies, platform and wide, are concepts known by some film fans, but to a larger audience it is an abstract idea. Previously, without a macro view of these release strategies, it was difficult to visualize the consistent patterns used within the industry across years.We believe we met the goal of simply showing these patterns and creating a quick takeaway for readers – 7 out of 8 Best Picture Winners used a platform release (this was later updated to 8 out of 9 after "The Shape of Water" won). This story was created (in part) as an attempt to broaden the audience of MarketWatch through its growing Entertainment section, as well as, create examples of interactive stories that can inspire the newsroom and for marketing/sales pitches.
What makes this project innovative?
This project distilled an abstract idea using clear data visualizations and created simple takeaways for readers. We had to be innovative in how we made a largely market-focused audience interested in an entertainment story. The numerous charts and data-driven elements of this story made it easier to bridge that gap. We also believe it created many \'ah-ha\' moments because it is a concept that is relatable to millions of people, but is rarely visualized.This project used modern data journalism and design tools (Scrollama.js, d3.js, and R) to create a story that let the data tell the story. The simplicity of a column chart and two main color choices meant that the patterns could easily be seen without having to hold the hand of the reader.We chose to present the data as "small multiples" so that we can achieve two goals at the same time: to easily highlight the patterns and insights we\'ve identified in the data; and to give readers the opportunity to explore the full data on their own. Having all the data in one view is also a powerful visual message that might intrigue readers\' interest on social media.While we\'ve definitely seen this type of data storytelling in other publications, it was an innovative step for us and helped expose the publication to fans of numbers, film, data visualizations, and data journalism.
What was the impact of your project? How did you measure it?
The project was successful in drawing a new audience to MarketWatch. Reposts of the article on FiveThirtyEight, FlowingData, Bloomberg, Reddit and a couple of statistics blogs meant we were exposed to a new group of readers like we intended on attracting.We used Parse.ly to measure our audience and the relative metrics. We found long engagement times and continued viewership overtime. This article was also able to recur throughout the 2018 film awards season. Having dedicated more resources to create the interactive story, it was important to have something with a longer lifetime.
Source and methodology
While the story was anchored to a breaking news event (announcement of the 2018 Academy Award film nominees), we were able to prepare the majority of the data and visualizations two weeks in advance. We used two main sources:comScore – provided weekly theater counts for all Best Picture nominees from 2009-2017The Numbers – used for estimated film budgetsWe used the statistical software R to merge the individual CSV files (each film\'s theater counts) into one. From there we performed quick data visualizations and created descriptive statistics of the theater count data. It was initially difficult to find a consistent definition for wide and platform release strategies, but after speaking with experts and reading a variety of source material we were able to create our own definitions to use throughout the story.We gathered the significant awards dates for each year so that we could measure each "award bump" (the rise in theater counts after a nomination or win) and show it in the visuals. Wikipedia was used as a starting point and each date was verified using a variety of news sources.We also joined additional fields like estimated budget to the data and discovered several interesting patterns , such as the correlation between a movie\'s budget and its release strategy type, as well as between a nominee\'s release date and its chance of winning the Best Picture. These insights were introduced in the story through a few additional graphics.
Technologies Used
For this project we mainly used open source software, applications, and tools:Google slides – idealization, wireframing, layout Sublime Text – editing, writing, and building codeR – data joining, writing, interpretation, and visualizationExcel – adding additional data fieldsScrollama.js – manipulating our charts based on scroll position in the storyd3.js – small multiple column charts and transitions between scrolling statesAdobe Illustrator - styling and formatting static chartsR and Excel allowed us to create quick mockups during the initial design process. R in particular made it easy to display the 70+ charts for comparison and exploration.We were able to manipulate the small multiple column charts as the user scrolled by combining Scrollama.js and d3.js.