WWE Events & Match Ratings Database (2015–2025)

---

World Wrestling Entertainment (WWE) gathering datasets to make a database - Final Project for EDS 213

Author: Marie Tolteca, Student- Masters in Environmental Data Science

Date: May 13, 2026

---

A structured SQL database built from real WWE event and match rating data sourced from Cagematch.net and the Wrestling Observer Newsletter (WON). This project demonstrates end-to-end data engineering and analytics skills from schema design and data ingestion to SQL querying and Python visualization.

---

Table of Contents

• Project Overview
• Dataset
• Schema Design
• Data Cleaning
• Analysis Questions
• Visualizations
• Tech Stack
• How to Run
• Limitations & Future Work

---

Project Overview

This project was built over 7 weeks as part of a database design course. The goal was to:

1. Choose a real-world dataset
2. Design a relational schema
3. Clean and ingest data into a SQL database
4. Write analytical SQL queries
5. Visualize results in Python
6. Document and present findings

The dataset covers 10 years of WWE programming (2015–2025), including televised shows, non-televised live events, and premium live events (PLEs/PPVs).

---

Dataset

Source: Kaggle — WWE Events & Match Ratings (2015–2025)

File	Rows	Description
wwe_events.csv	5,970	Event-level data (show name, type, location)
wwe_match_rating.csv	8,101	Match-level data with fan and critic ratings

Rating Sources:

• CageMatch Rating — crowdsourced fan ratings from cagematch.net (6,235 matches rated)
• WON Star Rating — critic ratings from Dave Meltzer's Wrestling Observer Newsletter (2,171 matches rated)

Event Types:

Type	Count
Televised	3,261
Non-Televised	2,421
PLE/PPV	288

Table Definitions

wwe_events — Event-level fact table

Column	Type	Key
Date	DATE	PK
Promotion	TEXT	PK
Event	TEXT
EventName	TEXT
EventType	TEXT
CityTown	TEXT
StateProvince	TEXT
Country	TEXT

wwe_match_rating — Match-level fact table

Column	Type	Key
Index	INTEGER	PK
Date	DATE	FK → wwe_events
Promotion	TEXT	FK → wwe_events
Match	TEXT
CageMatchRating	DECIMAL
CageMatchRatingVotes	INTEGER
WONStarRating	DECIMAL
Opponent1	TEXT
Opponent2	TEXT

tag_teams — Unpivoted tag team opponents

Column	Type	Key
OpponentID	INTEGER	PK
MatchIndex	INTEGER	FK → wwe_match_rating
Team	TEXT

superstars — Unique superstar lookup table

Column	Type	Key
SuperstarID	INTEGER	PK
SuperstarName	TEXT

---

Schema

---

Data Cleaning

Cleaning was performed in R using the tidyverse package.

Steps taken:

• Dates were already in YYYY-MM-DD format and loaded as proper date types, no reformatting needed
• Numeric columns (CageMatchRating, WONStarRating, CageMatchRatingVotes) contained no text or mixed values
• Missing ratings represented as NA (expected: not every match is rated by both sources)
• Trailing commas removed from opponent strings (e.g. "Raquel Rodriguez, Stephanie Vaquer,")
• Tag teams extracted by detecting & in opponent columns and pivoting to long format
• 515 duplicates looking rows in wwe_events are legitimate. WWE runs multiple house shows in different cities under the same event name on the same date

Known limitation: 81 match dates in wwe_match_rating have no corresponding row in wwe_events. These matches are excluded from any query that joins both tables.

Analysis Questions

Q1 — How has fan engagement varied across WWE event types from 2015 to 2025?

• Line chart showing Cagematch rating votes by event type per year

📓 wwe_viewship.ipynb

---

Q2 — What are the top 5 most voted events for each event type?

• Visualizations: Horizontal bar chart of top 5 most voted events per event type

📓 wwe_event_type.ipynb

---

Q3 — Who are the top 20 highest rated WrestleMania superstars (2015–2025)?

• Visualizations: Horizontal bar chart of top 20 highest rated WrestleMania superstars

📓 wwe_wrestlemania_sups.ipynb

---

Q4 — How did John Cena's match ratings trend throughout his career (2015–2025)?

• Visualizations: Line chart of John Cena's average match rating by year

📓 wwe_johncena.ipynb

---

Q5 — How did John Cena's performance vary across career eras?

• Visualizations: Violin/box plot of John Cena's ratings by career era

📓 johncena_timeline.ipynb

---

Tech Stack

Tool	Purpose
R / tidyverse	Data cleaning and ingestion
SQLite	Relational database
Python / pandas	Data manipulation
Python / matplotlib	Visualization
Kaggle	Data source

---

How to Run

1. Download the CSVs from Kaggle
2. Run the R cleaning script to generate cleaned tables
3. Ingest CSVs into an SQL file using wwe_database script to create tables
4. Run SQL queries against the database - wwe_queries was created to test queries and experiment
5. Run the Python visualization scripts to generate plots - These are in the notebook folder, each script creates one visualization using an Cursor Execute (cur.execute) method.

---

Limitations & Future Work

• No EventID — the join between events and matches relies on Date + Promotion, which is imprecise. A proper EventID foreign key would make the schema more robust.
• 81 unmatched match dates — some matches cannot be linked to an event and are excluded from joined queries.
• Superstar name inconsistencies — name variations (e.g. "Rey Mysterio" vs "Rey Mysterio Jr.") could cause duplicate entries in the superstars table. A fuzzy matching or manual deduplication step would improve accuracy.
• WON ratings are sparse — only 2,171 of 8,101 matches have a WON rating, limiting cross-rating comparisons.