🏀 NBA Team Intelligence Dashboard

A data-driven exploration of what wins games in the NBA (2012–2024)

---

📌 Project Overview

The NBA Team Intelligence Dashboard is an interactive analytics application built using Streamlit that transforms raw NBA game data into actionable insights at the league and team level.

This project goes beyond basic visualizations by answering deeper analytical questions such as:

• What truly drives winning in the NBA?
• How do teams evolve across seasons?
• Can team-level performance metrics predict winning outcomes?
• How can complex analytics be summarized clearly for non-technical users?

The dashboard is designed as a portfolio-grade analytics product — clean, minimalistic, modern, and insight-focused.

---

🎯 Objectives

• Analyze NBA team performance trends from 2012 to 2024
• Identify key performance metrics that correlate with winning
• Classify teams based on overall strength
• Build a machine learning model to predict win probability
• Generate auto-written summaries that explain insights in simple language

---

📂 Dataset

Source: Kaggle – NBA Data (2012–2024)
Granularity: Game-level data

Coverage:

• Team statistics
• Shooting efficiency
• Advanced metrics (EFG%, PIE, Net Rating)
• Game results

The dataset is intentionally large and realistic, making it suitable for real-world analytics workflows.

---

🧠 Dashboard Pages & Features

1️⃣ League Overview

• Big-picture view of the NBA
• League-wide averages and distributions
• Overall competitiveness analysis
• Auto-generated league summary

📌 Purpose: Understand how balanced or dominant the league is across seasons.

---

2️⃣ Team Performance Deep-Dive

• Detailed analysis of a selected team
• Season-level KPIs (Win %, Net Rating, Points/Game)
• Multi-season performance trends
• Team strength classification
• Metric Definitions Panel for clarity
• Dynamic team summary

📌 Purpose: Tell a clear performance story for any NBA team.

---

3️⃣ What Wins Games? ⭐

• Core insight section
• Correlation analysis between metrics and win percentage
• Identification of strongest positive & negative drivers
• Clear interpretation of results via auto summaries

📌 Purpose: Answer the most important question — what actually matters for winning?

---

4️⃣ Team Strength Classification

• Team categorization
• Teams classified as:
  • Elite
  • Strong
  • Average
  • Weak
• Distribution analysis across the league
• Insight-driven summary

📌 Purpose: Quickly understand competitive tiers within the NBA.

---

5️⃣ Win Prediction

• Machine Learning application
• Logistic Regression model
• Uses team-season metrics to predict win probability
• Model accuracy reporting
• Feature importance visualization
• Explainable predictions (no black box)

📌 Purpose: Demonstrate applied machine learning with explainability.

---

🤖 Machine Learning Details

• Model: Logistic Regression
• Target: Win Flag (Win % ≥ 50%)

Features Used:

• FG%
• 3PT%
• FT%
• Assists/Game
• Rebounds/Game
• Turnovers/Game
• Net Rating
• EFG%
• PIE

Preprocessing:

• Feature scaling using StandardScaler

Evaluation Metric: Accuracy

The model is intentionally simple and interpretable — clarity over complexity.

---

🧠 Auto Summary Engine (Key Highlight)

Each dashboard page includes an auto-generated narrative summary that:

• Translates charts into plain English
• Highlights the most important insights
• Makes the dashboard accessible to non-technical users

📌 This feature simulates how analytics insights are presented to stakeholders in real-world environments.

---

🗂️ Project Structure

NBA_Team_Intelligence_Dashboard/
│
├── app.py                      # Main Streamlit app
│
├── pages/                      # Dashboard pages
│   ├── 1_League_Overview.py
│   ├── 2_Team_Performance_DeepDive.py
│   ├── 3_What_Wins_Games.py
│   ├── 4_Team_Strength_Classification.py
│   └── 5_Win_Prediction.py
│
├── src/                        # Core analytics logic
│   ├── data_loader.py
│   ├── preprocessing.py
│   ├── metrics.py
│   ├── insights.py
│   ├── classification.py
│   ├── model.py
│   ├── summaries.py
│   └── metric_definitions.py
│
├── assets/                     # Custom CSS / styling
│
├── requirements.txt
├── .gitignore
└── README.md

📌 Clean separation between logic, presentation, and analytics.

---

⚙️ How to Run the Project

1️⃣ Install dependencies

pip install -r requirements.txt

2️⃣ Run the dashboard

streamlit run app.py

📌 No need to run src files manually — they are imported by the app.

---

🛠️ Tech Stack

• Python
• Streamlit
• Pandas
• Plotly
• Scikit-learn
• Kaggle Dataset

---

📈 Why This Project Matters (For Recruiters)

This project demonstrates:

✅ End-to-end data analytics workflow
✅ Clean modular code structure
✅ Business-oriented insight generation
✅ Explainable machine learning
✅ Strong data storytelling
✅ Production-style dashboard design

📌 It reflects how real analytics products are built — not just notebooks.

---

🚀 Future Improvements (Optional)

• Team-to-team comparison mode
• Playoff-specific analysis
• Advanced ML models
• Player-level integration

(Project intentionally considered complete for portfolio use.)

---

🙌 Acknowledgements

• NBA & Kaggle community for the dataset
• Streamlit & open-source ecosystem

⭐ If you liked this project, feel free to star the repository!