SportTwist 🏈

A Sports Intelligence & Insight System for NFL Game Analysis

Built for HackGT 12 - Crypt of Data Track - PrizePicks Challenge

Project Vision

SportTwist revolutionizes NFL analysis by allowing users to "Change a play. See the alternate outcome."

Our platform combines historical NFL data spanning 9 years (2017-2025) with advanced predictive modeling to deliver unprecedented insights into game dynamics and potential outcomes. Using Monte Carlo simulation methods and statistical analysis of over 107,609 play records, SportTwist transforms raw game data into actionable intelligence.

Core Philosophy

Event Modification Engine: Change any single gameplay element and witness how that alteration cascades through the entire game timeline
Monte Carlo Analysis: Process thousands of game variations using probability-based modeling to show realistic alternate outcomes
Temporal Ripple Tracking: Follow how one modified moment creates waves of change throughout subsequent plays, drives, and final results
Historical Context Intelligence: Leverage 9 years of NFL data to ground alternate scenarios in realistic statistical patterns
Signal Extraction: Transform overwhelming game data into clear cause-and-effect relationships between decisions and outcomes

Event Modification Capabilities

SportTwist goes beyond prediction - it lets you actively reshape game history and analyze the consequences:

🏈 Play-Level Modifications

Formation Changes: Switch from shotgun to I-formation on a crucial 3rd down and see how defensive alignment changes affect success probability
Route Adjustments: Modify receiver routes mid-play and track how different patterns would have exploited defensive coverage
Audible Decisions: Change quarterback audibles at the line and observe how snap count variations affect defensive timing

👥 Personnel Decisions

Player Substitutions: Insert backup players during key moments and analyze how their unique skill sets alter game flow
Position Swaps: Move players to different positions and evaluate performance changes based on their actual statistical profiles
Fatigue Modeling: Adjust player rest patterns and see how stamina affects late-game execution

🎯 Strategic Modifications

Coaching Calls: Change timeout usage, challenge flags, or 4th down decisions and trace their impact on final outcomes
Game Plan Shifts: Modify offensive/defensive schemes mid-game and track how opposing teams would have adapted
Clock Management: Alter pace of play decisions and see how time remaining affects strategic options

Outcome Analysis Framework

Each modification generates detailed impact analysis across multiple dimensions:

📊 Immediate Effects

Play Success Metrics: Yards gained/lost, first downs achieved, turnover probability shifts
Field Position Changes: How modified plays affect starting position for subsequent drives
Down and Distance Impact: Track how single-play changes create different situational contexts

🎯 Game-Wide Consequences

Momentum Shifts: Quantify psychological momentum changes and their statistical impact on subsequent performance
Score Progression: Chart exact point differentials created by individual modifications throughout the game
Win Probability Evolution: Track moment-by-moment changes in victory likelihood across the entire game timeline

🏆 Season Implications

Playoff Seeding: See how single-game changes could have altered final standings and postseason matchups
Statistical Records: Track how modifications would have affected individual player stats and team achievements
Championship Pathways: Analyze how one changed play could have created entirely different playoff scenarios

✨ Key Features

🔍 Advanced Search & Filtering

Smart Team Search: Search by full team names (e.g., "dolphins") or abbreviations ("MIA")
Multi-Dimensional Filters: Filter by season (2017-2025), team, player, game type (Regular/Playoffs)
Chronological Sorting: Games displayed by actual date (most recent first)
Real-time Results: Instant search with live API integration

📊 Data Intelligence

107,609+ Play Records: Comprehensive NFL play-by-play data
842+ Scheduled Games: Complete game schedules with scores
Historical Coverage: 9 years of NFL data (2017-2025)
Team Performance Analytics: High-profile team tracking and analysis

🎮 Interactive UI/UX

Search Overlay: Advanced search interface with filter buttons
Trending Searches: Dynamic latest game and current week shortcuts
Popular Games: Curated high-profile matchups
Responsive Design: Clean, modern interface with real-time updates

🤖 Intelligent Systems (Current + Planned)

Pattern Recognition: Identify trends in team performance and game outcomes
Anomaly Detection: Flag unusual game patterns or statistical outliers
Predictive Simulation: Game outcome prediction based on historical data
Real-time Insights: Live game analysis and trend identification

🏗️ Architecture

Frontend (React + TypeScript + Vite)

frontend/
├── src/
│   ├── components/
│   │   ├── MinimizedGameCard.tsx     # Home page game cards
│   │   └── SearchResultGameCard.tsx  # Search result display
│   ├── services/
│   │   └── nflApi.ts                # API communication layer
│   ├── App.tsx                      # Main application component
│   ├── App.css                      # Styling and responsive design
│   └── main.tsx                     # Application entry point
├── package.json                     # Dependencies and scripts
└── vite.config.ts                   # Build configuration

Backend (Python + Flask + nflreadpy)

backend/
├── nfl_api.py                       # Main API server
├── game.py                          # Game simulation logic
├── interactive_game_sim.py          # Interactive simulation features
├── moremontecarlo.py               # Monte Carlo simulation methods
└── team_logos_dict.py              # Team branding and assets

API Endpoints

GET /api/games - Retrieve all games with scores
GET /api/search?q={query} - Advanced search functionality
GET /api/popular - Get high-profile/trending games
GET /api/game/{game_id} - Detailed game information

🚀 Getting Started

Prerequisites

Node.js 18+
Python 3.11+
pip (Python package manager)
npm (Node package manager)

Installation

Clone the Repository

git clone https://github.com/sandro21/SportTwist.git
cd SportTwist

Backend Setup

# Create virtual environment
python -m venv .venv

# Activate virtual environment
# Windows:
.venv\Scripts\activate
# macOS/Linux:
source .venv/bin/activate

# Install dependencies
pip install flask flask-cors nflreadpy pandas

Frontend Setup
```
cd frontend
npm install
```

Running the Application

Start Backend Server (Terminal 1)
```
cd backend
python nfl_api.py
```
Server runs on: http://localhost:5001
Start Frontend Server (Terminal 2)
```
cd frontend
npm run dev
```
Application runs on: http://localhost:5173 (or next available port)

📡 Data & APIs

Data Sources

nflreadpy: Official NFL play-by-play and schedule data
Real-time Integration: Live NFL statistics and game information
Historical Coverage: 2017-2025 seasons (107,609+ plays)

Data Processing Pipeline

Data Ingestion: Load NFL data via nflreadpy
Data Transformation: Convert to standardized JSON format
Caching Layer: In-memory caching for performance
API Serving: RESTful endpoints for frontend consumption

Search Intelligence

# Smart team name mapping
team_name_map = {
    'eagles': 'PHI', 'philadelphia': 'PHI',
    'cowboys': 'DAL', 'dallas': 'DAL',
    'chiefs': 'KC', 'dolphins': 'MIA',
    # ... 32 NFL teams with multiple search terms
}

# Player-team associations for advanced filtering
player_team_map = {
    'mahomes': ['KC'], 'allen': ['BUF'], 
    'tua': ['MIA'], 'jackson': ['BAL']
    # ... key players with team mappings
}

🎨 UI/UX Design

Design System

Color Scheme: Purple gradient (#8000FF) with white/glass morphism
Typography: GT Standard font family
Layout: Responsive grid system with flexbox/CSS Grid
Animations: Smooth transitions and hover effects

Component Architecture

// Game Card Interface
interface NFLGame {
  game_id: string;
  home_team: string;
  away_team: string;
  home_score: number;
  away_score: number;
  week: number;
  season: number;
  gameday: string;
  gametime: string;
}

Search Experience

Keyboard Shortcuts: Press '/' to open search
Filter System: Season, Team, Player, Game Type filters
Auto-complete: Smart suggestions and team name matching
Real-time Updates: Instant search results as you type

🧠 Intelligence Features

Current Implementations

Pattern Recognition
- Chronological game sorting by actual dates
- High-profile team identification and trending
- Popular game curation based on team significance
Search Intelligence
- Multi-format team name recognition
- Player-based game filtering
- Season and game type categorization
Data Visualization
- Clean game card layouts with team logos
- Score alignment and formatting consistency
- Responsive search result containers

Planned Enhancements

Game Simulation Engine
- Monte Carlo simulation methods
- Historical performance analysis
- Outcome prediction algorithms
Anomaly Detection
- Statistical outlier identification
- Performance trend analysis
- Real-time alert systems
AI Insights Generator
- Natural language game explanations
- Performance trend summaries
- Predictive analysis reports

🔧 Technical Stack

Frontend Technologies

React 19.1.1: Modern component-based UI
TypeScript: Type-safe development
Vite 7.1.7: Fast build tool and dev server
CSS Grid/Flexbox: Advanced layout systems
Fetch API: HTTP client for backend communication

Backend Technologies

Python 3.11: Core programming language
Flask: Lightweight web framework
nflreadpy: NFL data access library
Pandas: Data manipulation and analysis
CORS: Cross-origin resource sharing

Development Tools

ESLint: Code linting and quality
Git: Version control
VS Code: Development environment
Chrome DevTools: Frontend debugging

📈 Performance & Optimization

Backend Optimization

Data Caching: In-memory caching of NFL data (107K+ records)
Efficient Sorting: Optimized chronological sorting algorithms
API Response Limits: Configurable result limits (20-50 games)
Error Handling: Comprehensive error handling and fallbacks

Frontend Optimization

Component Memoization: React performance optimizations
Debounced Search: 300ms search debouncing
Lazy Loading: Efficient data loading strategies
CSS Grid: Hardware-accelerated layouts

Load Times

Backend Startup: ~2-3 seconds (data loading)
Frontend Build: ~465ms (Vite)
API Response: <100ms (cached data)
Search Results: Real-time (<300ms)

🎯 HackGT 12 Alignment

Crypt of Data Track Requirements ✅

Signal from Noise: Advanced filtering and search cuts through 107K+ data points
Pattern Recognition: Chronological sorting and trend identification
Real-time Systems: Live search and filtering capabilities
User Experience: Intuitive interface for analysts and fans

PrizePicks Challenge Elements ✅

Performance Trends: Historical game and player tracking
Filtering Systems: Multi-dimensional data filtering
Analytics Foundation: Structured for future AI/ML integration
Real-time Insights: Live game data and search capabilities

Innovation Points

Smart Search: Natural language team name recognition
Data Intelligence: Historical coverage with predictive foundations
User Experience: Keyboard shortcuts and responsive design
Scalable Architecture: Modular frontend/backend separation

🚀 Future Roadmap

Phase 1: Enhanced Intelligence (Next Sprint)

Monte Carlo game simulation integration
AI-powered game outcome predictions
Statistical anomaly detection algorithms
Performance confidence scoring

Phase 2: Advanced Analytics (Future Development)

Real-time game commentary AI
Trend visualization dashboards
Player performance anomaly alerts
Broadcast-style game explanations

Phase 3: Platform Expansion (Long-term)

Multi-sport support (NBA, MLB, etc.)
Mobile application development
Advanced machine learning models
Professional analytics tools

🤝 Contributing

We welcome contributions to SportTwist! Here's how to get involved:

Development Workflow

Fork the repository
Create a feature branch (git checkout -b feature/amazing-feature)
Commit your changes (git commit -m 'Add amazing feature')
Push to the branch (git push origin feature/amazing-feature)
Open a Pull Request

Code Standards

TypeScript: Strict type checking enabled
ESLint: Follow configured linting rules
Python: PEP 8 style guidelines
Testing: Write tests for new features

📋 Project Status

Current Version: v1.0 (HackGT 12 Submission)

Completed Features ✅

Full-stack application architecture
NFL data integration (2017-2025)
Advanced search and filtering system
Responsive UI with modern design
Real-time API communication
Chronological data sorting
Team and player intelligence
Popular games curation

In Development 🚧

Game simulation engine
Predictive analytics
AI insight generation
Performance optimization

📞 Contact & Team

Built for HackGT 12 by Team SportTwist

Repository: github.com/sandro21/SportTwist
Track: Crypt of Data
Challenge: PrizePicks - Sports Intelligence & Insight Systems
Demo: Available at hackathon presentation

📄 License

This project is developed for HackGT 12 and educational purposes.

🙏 Acknowledgments

HackGT 12 for providing the platform and challenge
PrizePicks for the sports intelligence challenge framework
nflreadpy for comprehensive NFL data access
React & Vite teams for excellent development tools
Flask community for lightweight backend framework

"Change a play. See the alternate outcome." - SportTwist transforms how we analyze and understand NFL games through intelligent data processing and predictive insights. 🏈✨

Name		Name	Last commit message	Last commit date
backend		backend
frontend		frontend
.gitignore		.gitignore
README.md		README.md

Latest commit

History

Folders and files

Repository files navigation