← Back to Projects

Video Game Sales Forecasting

Forecasting 2017 sales trends for strategic advertising campaigns using historical video game sales data. Analyzing platform lifecycles, genre performance, and regional market preferences to identify winning patterns for the online store Ice.

Python Pandas NumPy Matplotlib Seaborn SciPy

Project Overview

Business Context: Understanding the gaming industry landscape for investment or publishing decisions.

This project analyzes historical video game sales data from 2013-2016 to identify patterns that determine a game's success. Working with data from the online store Ice, I analyzed 16,715 video games across multiple platforms, genres, and regional markets to forecast trends for 2017 advertising campaigns.

The analysis combines exploratory data analysis, trend forecasting, and statistical hypothesis testing to provide actionable recommendations for marketing strategy.

What This Demonstrates

Learning Challenge

  • Gaming industry dynamics and platform ecosystems
  • Time-series analysis across multiple regions
  • Handling categorical data (genres, platforms, publishers)

Problem-Solving Process

  1. Market Research: Investigated how the gaming industry works (platforms, regions, genres)
  2. Multi-Dimensional Analysis: Examined sales across time, geography, and product categories
  3. Trend Identification: Discovered platform lifecycles and regional preferences
  4. Insight Synthesis: Connected multiple findings into a coherent market narrative

Professional Outcome

  • Delivered insights that could inform publishing strategy, platform prioritization, or market entry decisions
  • Demonstrated ability to quickly become conversant in a new industry
  • Created visualizations that make 35+ years of data immediately accessible

Tools Utilized

  • VS Code with GitHub Copilot for development
  • Jupyter Notebook for interactive analysis
  • Git/GitHub for version control

Key Objectives

Platform Analysis

Identify gaming platforms with growth potential for 2017 based on lifecycle analysis

Genre Performance

Analyze genre performance and profitability across different markets

Regional Markets

Understand regional preferences in North America, Europe, and Japan

Hypothesis Testing

Test hypotheses about user ratings across platforms and genres statistically

Dataset

The analysis uses comprehensive video game sales data spanning 1980-2016:

  • Total Records: 16,715 video games
  • Time Period: 1980-2016 (focused on 2013-2016 for forecasting)
  • Platforms: Multiple gaming platforms including PS4, XOne, PC, 3DS, and more
  • Regions: North America, Europe, Japan, and other markets
  • Metrics: Sales data, critic scores, user scores, ESRB ratings

Analysis Workflow

  1. Data Preparation - Load, clean, and preprocess sales data
  2. Exploratory Analysis - Examine platform lifecycles and genre distributions
  3. Platform Analysis - Identify growing and declining platforms
  4. Profitability Analysis - Determine most profitable genres and platforms
  5. Regional Analysis - Compare market preferences across regions
  6. Hypothesis Testing - Statistical validation of user rating patterns
  7. Recommendations - Data-driven marketing strategy for 2017

Key Findings

PS4 Dominance

49% growth and 314M in total sales (2013-2016), leading platform for 2017 campaigns

Platform Lifecycle

Gaming platforms typically last 10-12 years before rapid decline; generational transition observed

Action Games Lead

Action games dominate with 29.5% global market share; Shooters have highest average sales

Regional Differences

PS4 leads in NA/EU; 3DS dominates Japan (48% market share); RPGs preferred in Japan (36%)

Review Correlation

Moderate positive correlation (0.40) between critic scores and sales

Rating Impact

Mature-rated games generate highest total sales but Teen/Everyone have broader appeal

Statistical Testing

The project employs hypothesis testing to validate key assumptions:

  • Hypothesis 1: Average user ratings of Xbox One and PC platforms are the same
  • Hypothesis 2: Average user ratings for Action and Sports genres are different
  • Methodology: Independent samples t-tests with α = 0.05
  • Results: Both hypotheses tested with statistical rigor using SciPy

Strategic Recommendations

Data-driven recommendations for 2017 advertising campaigns:

  1. Focus on PS4: Prioritize PS4 platform showing strong growth trajectory
  2. Action & Shooter Genres: Emphasize high-performing genres in marketing
  3. Regional Customization: Tailor campaigns to regional preferences (RPGs for Japan)
  4. Quality Signaling: Feature games with high critic scores prominently
  5. Rating Segmentation: Balance Mature-rated blockbusters with broader-appeal titles
  6. Multi-platform Strategy: Consider XOne for North American market penetration
  7. New Releases: Prioritize recent releases (2015-2016) for maximum impact
  8. Lifecycle Awareness: Avoid marketing older-generation platforms (PS3, X360)

Skills Demonstrated

  • Time series analysis and trend forecasting
  • Multi-dimensional data analysis (platform, genre, region)
  • Statistical hypothesis testing with SciPy
  • Data cleaning and preprocessing of large datasets
  • Exploratory data analysis (EDA)
  • Data visualization with Matplotlib and Seaborn
  • Correlation analysis and feature relationships
  • Business strategy formulation from data insights
  • Regional market analysis and segmentation

Technologies Used

Python 3.x Pandas NumPy Matplotlib Seaborn SciPy Jupyter Notebook

Business Impact

This analysis provides strategic value through:

  • Campaign Optimization: Data-driven platform and genre selection for 2017 advertising
  • Budget Allocation: Focus resources on high-growth platforms and genres
  • Regional Strategy: Customize marketing approaches for NA, EU, and Japan markets
  • Product Selection: Identify which games to feature prominently in campaigns
  • Risk Mitigation: Avoid declining platforms and low-performing genres
  • Revenue Forecasting: Predict potential returns from different market segments