Boxoffice Data Analysis

This article provides a comprehensive analysis for Boxoffice data

1. Boxoffice:

Box office provide reports on financial performance of movies, typically measured by the total amount of revenue generated from ticket sales during its theatrical run.

1.1. Aim of the Boxoffice:

The aim of box office analysis is to understand the factors that contribute to a movie’s financial success at the box office. This involves exploring the relationship between a movie’s budget, release date, genre, runtime, ratings, and other variables, and its box office performance.

2. Methodology:

A public Box office dataset is used in this study which is available in Kaggle and some rows of the dataset are shown below:

Show code

# ensure the results are repeatable
set.seed(7)
# load the data
df <- read.csv('blockbusters.csv')
# head of the data
df %>% arrange(desc(year)) %>% head() %>%
  kbl(caption = "Head of Data sorted by Worldwide Gross in descending order") %>%
  kable_classic(full_width = F, html_font = "Cambria")

(#tab:read the data)Head of Data sorted by Worldwide Gross in descending order

Main_Genre	Genre_2	Genre_3	imdb_rating	length	rank_in_year	rating	studio	title	worldwide_gross	year
Action	Adventure	Drama	7.4	135	1	PG-13	Walt Disney Pictures	Black Panther	$700,059,566	2018
Action	Adventure	Sci-Fi	8.5	156	2	PG-13	Walt Disney Pictures	Avengers: Infinity War	$678,815,482	2018
Animation	Action	Adventure	7.8	118	3	PG	Pixar	Incredibles 2	$608,581,744	2018
Action	Adventure	Drama	6.2	129	4	PG-13	Universal Pictures	Jurassic World: Fallen Kingdom	$416,769,345	2018
Action	Comedy		7.8	119	5	R	20th Century Fox	Deadpool 2	$318,491,426	2018
Action	Adventure	Drama	7.9	147	6	PG-13	Paramount Pictures	Mission: Impossible - Fallout	$220,159,104	2018

3. Results:

3.1. Relationship between worldwide sales and othe features

A correlation matrix is plotted for the numeric values of the dataset to find a pattern between these values.

• The following figure shows that the worldwide sales is correlated with year by 68% which means that recent movies had a better sales in general which could be due to the inflation and loss in the value of the money as well. But, overall, this trend is observed.

• Moreover, worldwide sales is correlated with length of the movie and IMDB rating with the correlation rates of 34% and 21%, respectively.

Show code

df_num <- select(df, c("year", "imdb_rating", "length", "worldwide_sales"))
df_cormatrix <- data.frame(cor(df_num))

ggcorrplot(df_cormatrix, type="lower", hc.order = TRUE, lab = TRUE, insig ="blank")

3.2. Relationship between the worlwide sales and other features?

3.2.1. Worldwide sales and IMDB rating:

• The results show that movies with higher IMD ratings tend to have a higher worldwide sales. The regression line depicts this pattern as well.

Show code

# load the plotly and RColorBrewer packages
library(plotly)
library(RColorBrewer)

# create a color palette with Set1 colors
n_colors <- length(unique(df$Main_Genre))
colors <- brewer.pal(n_colors, "Set1")

# create a scatter plot with plotly
p <- plot_ly(df, x = ~imdb_rating, y = ~worldwide_sales, color = ~Main_Genre,
             colors = colors, type = 'scatter', mode = 'markers') %>%
  layout(title = 'Scatter plot of worldwide sales and IMDB rating',
         xaxis = list(title = 'IMDB rating'),
         yaxis = list(title = 'Worldwide sales [$]')) %>%
  add_lines(x = ~imdb_rating, y = ~fitted(lm(worldwide_sales ~ imdb_rating, data = df)),
            line = list(dash = 'dash'), color= "blue", name = "regression line") 

# display the plot
p

3.2.2. Worldwide sales and year of release:

• The following plot shows the worldwide sales versus year of the movie release. We see a general pattern that the sales increases as the the year of the movie is more recent. However, a specific pattern is observed that after 2015 there is a sudden drop in the worldwide sales.

• The highest sales occured approximately in the range of 2005 to 2015.

Show code

# load the plotly and RColorBrewer packages
library(plotly)
library(RColorBrewer)

# create a color palette with Set1 colors
n_colors <- length(unique(df$Main_Genre))
colors <- brewer.pal(n_colors, "Set1")

# create a scatter plot with plotly
p <- plot_ly(df, x = ~year, y = ~worldwide_sales, color = ~Main_Genre,
             colors = colors, type = 'scatter', mode = 'markers') %>%
  layout(title = 'Scatter plot of worldwide sales and year of the movie release',
         xaxis = list(title = 'year'),
         yaxis = list(title = 'Worldwide sales [$]')) %>%
  add_lines(x = ~year, y = ~fitted(lm(worldwide_sales ~ year, data = df)),
            line = list(dash = 'dash'), color= "blue", name = "regression line") 

# display the plot
p

3.2.3. Worldwide sales and length of the movie:

• Length of the movie has a direct relation with the worldwide sales of the movie. This is illustrated in the following plot where the sales enhances as the length of the movie is more.

• A regression analysis shows the direct relation between these two variabels. Lenghty movies tend to have higher sales.

  Show code

  # load the plotly and RColorBrewer packages
  library(plotly)
  library(RColorBrewer)
  
  # create a color palette with Set1 colors
  n_colors <- length(unique(df$Main_Genre))
  colors <- brewer.pal(n_colors, "Set1")
  
  # create a scatter plot with plotly
  p <- plot_ly(df, x = ~length, y = ~worldwide_sales, color = ~Main_Genre,
               colors = colors, type = 'scatter', mode = 'markers') %>%
    layout(title = 'Scatter plot of worldwide sales and length of the movie',
           xaxis = list(title = 'length [min]'),
           yaxis = list(title = 'Worldwide sales [$]')) %>%
    add_lines(x = ~length, y = ~fitted(lm(worldwide_sales ~ length, data = df)),
              line = list(dash = 'dash'), color= "blue", name = "regression line") 
  
  # display the plot
  p