Dataset Overview and Source

Video Game Sales Dataset

This analysis examines data of video game sales from 1980 to 2016.

Data Source: Kaggle Scraped from VGChartz

Key Variables:

  • Name / Platform / Year / Genre — game metadata
  • Publisher — who released the game
  • NA_Sales, EU_Sales, JP_Sales, Other_Sales — regional sales in millions
  • Global_Sales — total worldwide sales in millions

R Code for Data Preparation

Loading and Preparing data:

library(ggplot2)
library(plotly)
library(dplyr)

vg = read.csv("vgsales.csv")

vg = vg %>%
  filter(!is.na(Year), Year != "N/A") %>%
  mutate(Year = as.integer(Year))

vg$Publisher_Group = ifelse(vg$Publisher == "Nintendo",
                            "Nintendo","Other Publishers")

Global Sales: Nintendo vs. Everyone Else (plotly)

Top 10 Publishers by Global Sales (ggplot)

Nintendo Sales by Genre (ggplot)

3D View: Year, Genre & Global Sales (plotly)

3D Plot Analysis

What the chart shows:

  • Nintendo’s Sports and Platform games have peaks in the mid 2000s driven by the Wii’s popularity
  • Misc and Racing also have peaks during the 2000s
  • There’s a peak for Role-Playing in the mid 90s likely due to Pokemon’s popularity
  • Post 2012 sales drop off across almost every genre, which lines up with the Wii U’s underwhelming performance in the market

Descriptive Statistics

vg %>%
  group_by(Publisher_Group) %>%
  summarise(
    Games      = n(),
    Total_Sales = round(sum(Global_Sales, na.rm = TRUE), 1),
    Mean_Sales  = round(mean(Global_Sales, na.rm = TRUE), 2),
    Median_Sales = round(median(Global_Sales, na.rm = TRUE), 2),
    SD_Sales    = round(sd(Global_Sales, na.rm = TRUE), 2),
    Max_Sales   = round(max(Global_Sales, na.rm = TRUE), 2)
  )
## # A tibble: 2 × 7
##   Publisher_Group  Games Total_Sales Mean_Sales Median_Sales SD_Sales Max_Sales
##   <chr>            <int>       <dbl>      <dbl>        <dbl>    <dbl>     <dbl>
## 1 Nintendo           696       1784.       2.56         0.92     5.68      82.7
## 2 Other Publishers 15627       7036.       0.45         0.16     0.97      21.8

Summary Statistics Analysis

Interpreting the summary table:

  • Nintendo obviously published fewer games in total but pulled in a large share of total global sales
  • Nintendo’s mean sales per game is higher than other publishers. Their games usually perform much better than the industry norm.
  • Wii Sports is the best-selling game in the whole dataset, nearly double the next best (which is a Nintendo title)
  • The median for other publishers is very low which shows how hard it is to have a successful title

Sales Over Time: Nintendo vs. Other Publishers

Conclusions

What this analysis found:

  • Nintendo dominates the video game industry
  • The Wii era in the mid 2000s is Nintendo’s greatest peak from the provided data
  • Platform and Sports titles are Nintendo’s best sellers, while other publishers succeed more in Action/Shooters
  • Nintendo accounts for a significant share of the total global sales each year
  • Nintendo’s strategy focuses on slowly releasing high-quality games rather than releasing a high number of low quality games.

References