Data Analysis

Statistical Analysis of NBA Team Performance and Winning Percentage

Project Overview

This project analyzed relationships between key basketball performance metrics and overall team success in the NBA. Using historical team statistics, the goal was to determine which measurable factors most strongly influence a team’s winning percentage.

Rather than relying on intuition or reputation, the project applied data-driven analysis to evaluate how offensive output, defensive performance, and bench production contribute to wins. The final deliverable included statistical findings supported by clear visualizations.

Objective

The main objectives of this project were to:

Analyze how different team statistics impact win percentage
Identify which metrics are most strongly correlated with success
Visualize relationships between performance indicators and outcomes
Draw data-backed conclusions about what drives NBA team performance

Data and Methodology

The analysis used season-level NBA team statistics including:

Average points scored
Average points allowed
Bench scoring
Overall team win percentage

The project followed a structured analytical process:

Data Preparation
- Imported and cleaned NBA statistical data
- Created several metrics for individual player and team wide analysis
- Ensured consistency and accuracy
Exploratory Data Analysis
- Examined distributions of key variables
- Created visualizations to explore trends
- Compared offensive and defensive metrics
Statistical Analysis
- Evaluated correlations between performance indicators
- Used visual analytics to interpret relationships
- Compared high-performing vs low-performing teams

Key Conclusions

The project led to several data-driven conclusions:

Defense is the most important predictor of NBA team success.
Offensive strength must be paired with strong defense to translate into wins.
Bench production contributes to success but is secondary to core team performance.
Visual and statistical analysis can clearly separate winning and losing team profiles.

Deliverables

Cleaned NBA statistical dataset
Python/Jupyter Notebook with full analysis
Multiple visualizations illustrating trends
Final report summarizing findings

Project Links

Jupyter Notebook & Code: [GitHub link if available]
Full Report: Available upon request

Federal Owned Buildings and Lease Analysis

Project Overview

This project analyzed publicly available data on federally owned buildings and government leases to uncover trends in property usage, costs, and geographic distribution. The goal was to use data visualization techniques to better understand how federal real estate assets are managed and how leasing and ownership patterns vary across agencies and locations.

Using Tableau, I transformed raw government data into interactive dashboards that allow users to explore key metrics related to federal property portfolios in a clear and intuitive way.

Results and Insights

The final analysis revealed several important findings:

Significant variation in federal property distribution by state and region
Clear differences between agencies in their reliance on owned buildings versus leased space
Patterns showing where leasing costs are highest
Geographic hotspots of federal real estate activity

These insights demonstrate how data visualization can help government stakeholders and the public better understand federal spending and resource allocation.

Project Links

Interactive Tableau Dashboard: https://public.tableau.com/app/profile/pablo.gonzalez4493/vizzes
Project Files and Report: Available upon request

Data and Methodology

The project followed a structured analytics workflow:

Data Preparation
- Cleaned and standardized data fields
- Created calculated metrics such as total square footage, lease costs, and property counts
Exploratory Data Analysis
- Examined distributions of owned vs. leased properties
- Analyzed property data by state, agency, and building type
- Identified trends in federal real estate spending
Visualization Development
- Designed charts, maps, and graphs to communicate insights
- Built filters and parameters for user interaction
- Developed a comprehensive Tableau dashboard

Fitness Center Database Design and Analytics System

Project Overview

This project involved designing and implementing a fully functional relational database system to manage operations for a fitness and training center. The goal was to create a structured database capable of handling customers, class scheduling, trainers, payments, attendance, and facility management while maintaining data integrity and scalability.

The project demonstrates my ability to translate real-world business requirements into a well-designed database architecture and to use SQL to manage, query, and analyze structured data.

Skills Demonstrated

This project highlights my abilities in:

Relational database design
Entity-relationship modeling
Data normalization
SQL development
Query writing and optimization
Translating business logic into technical systems
Analytical thinking and problem solving

System Design and Methodology

The project followed a structured database development process:

Requirements Analysis
- Identified core entities needed for a class-based business
- Determined relationships between customers, schedules, payments, and attendance
Entity Relationship (ER) Modeling
- Designed a conceptual ER diagram to represent system logic
- Defined relationships such as:
  - Customers enroll in classes
  - Trainers teach scheduled sessions
  - Payments are linked to customers
  - Facilities host scheduled classes
Relational Schema Development
- Converted the ER diagram into a normalized relational database
- Created tables including:
  - customer
  - class
  - schedule
  - trainer
  - facility
  - payment
  - enrollment
  - attendance
SQL Implementation
- Wrote SQL scripts to create the full database schema
- Implemented primary keys, foreign keys, and constraints
- Inserted sample data for testing
- Developed queries to simulate real-world analytics

Project Artifacts

ER Diagram and Schema Design
SQL Database Creation Scripts
Sample Queries
Documentation and Design Report

(Available upon request or via GitHub repository)

Analyzing Which Colleges Produce the Best NBA Prospects

Project Overview

This project analyzed over 20 years of NBA player performance data to determine which colleges and universities most consistently produce successful NBA talent. The goal was to use historical data to identify patterns linking collegiate programs to professional basketball success, and to visualize those findings in a clear, interactive format.

Rather than simply evaluating individual players, the project focused on answering a broader analytical question: Which schools develop the strongest NBA prospects, and how can that success be measured objectively through data?

Deliverables

Cleaned and structured NBA player dataset
Excel analysis and calculated performance metrics
Interactive Tableau dashboard comparing colleges
Final presentation summarizing methodology and findings

Skills Demonstrated

Data cleaning and integration
Exploratory data analysis
Performance metric development
Data visualization and dashboard design
Trend analysis and comparative analytics
Storytelling with data

Results and Insights

Through this analysis, several meaningful conclusions were drawn:

A small group of colleges consistently produce a disproportionate share of successful NBA players
Certain programs excel at developing specific types of players (scorers, defenders, playmakers)
Success can be measured not just by number of players produced, but by career longevity and statistical performance
Data-driven rankings of colleges often differ from common public perception

The interactive Tableau dashboard allows users to explore these insights by filtering players by college, position, and performance metrics.

Project Links

Interactive Tableau Dashboard: https://public.tableau.com/app/profile/pablo.gonzalez4493/vizzes
Final Presentation: Available upon request

Page updated

Google Sites

Report abuse