Deadlock Analytics

A comprehensive web-based analytics platform for Deadlock game statistics. View detailed player performance metrics, match history, hero statistics, and interactive visualisations.

Live Application

> Update 2026-06-16: Underlying API's have changed and breaking some elements on the website, not actively maintaining website at the moment.

🚀 Try it now: https://deadlock-analytics-164941517977.europe-west2.run.app

The application is deployed on Google Cloud Run in the europe-west2 region, providing:

• Auto-scaling from 0 to 10 instances based on traffic
• 2Gi memory and 2 CPU cores per instance
• 120-second timeout for comprehensive data analysis
• Managed with Terraform for infrastructure as code

Features

• Player Dashboard: View Steam profile, rank badge, and summary statistics
• Performance Analytics: Track KDA trends with rolling averages over time
• Hero Statistics: See your top 5 most-played heroes with win rates
• Hero Filtering: Click any hero to filter all stats and charts by that hero (with audio feedback!)
• MMR History: Monitor rank progression with smoothed trend lines
• Community Comparison: Compare your performance against community percentiles
• Interactive Charts: Dynamic visualizations with Plotly for kills, deaths, and performance curves
• Match Location Analysis: Visualize kill/death locations on the game map
• Interactive Audio: Hero voicelines, UI sounds, and sequential footstep sounds on interactions

Quick Start

Prerequisites

• Python 3.12 or higher
• uv package manager (or pip)

Installation

# Clone the repository
git clone https://github.com/Filpill/deadlock_analytics.git
cd deadlock-analytics

# Install uv (if not already installed)
pip install uv

# Create virtual environment and install dependencies
uv venv
source .venv/bin/activate  # On Windows: .venv\Scripts\activate
uv sync

Running the Application

# Activate virtual environment
source .venv/bin/activate

# Run Flask app
python app.py

# Open browser to http://localhost:5000

Usage

1. Enter your Player ID (SteamID3 format, e.g., 199540209)
2. Click Analyze Player Stats
3. View your personalized analytics dashboard

Technology Stack

• Backend: Flask 3.1+ with Gunicorn WSGI server
• Data Processing: pandas, numpy, scipy
• Visualizations: Plotly 6.5+
• API Integration: Deadlock API (api.deadlock-api.com, assets.deadlock-api.com)
• Package Management: uv
• Deployment: Google Cloud Run, Docker, Terraform
• Audio: Browser Web Audio API with Deadlock game assets

Project Structure

deadlock-analytics/
├── app.py                  # Main Flask application
├── Dockerfile              # Cloud Run optimized Docker image
├── .dockerignore           # Docker build exclusions
├── templates/              # HTML templates
│   ├── index.html         # Landing page
│   └── results.html       # Analytics dashboard
├── static/                 # Static assets (images, fonts)
│   ├── img/               # Minimap, logos, banners
│   └── fonts/             # Custom fonts
├── terraform/              # Infrastructure as Code
│   ├── main.tf            # Cloud Run service configuration
│   ├── variables.tf       # Terraform variables
│   ├── outputs.tf         # Service outputs
│   └── README.md          # Terraform deployment guide
├── scripts/                # Analysis scripts and notebooks
├── pyproject.toml         # Project dependencies
├── uv.lock                # Locked dependency versions
├── CLAUDE.md              # Comprehensive development guide
└── README.md              # This file

API Documentation

This project uses the official Deadlock API:

• Data API: https://api.deadlock-api.com/docs
• Assets API: https://assets.deadlock-api.com/docs

Development

For detailed development guidelines, API schema reference, and contribution instructions, see CLAUDE.md.

Cloud Deployment

The application is production-ready and deployed to Google Cloud Run using a containerized setup:

Infrastructure

• Platform: Google Cloud Run (fully managed serverless)
• Region: europe-west2 (London)
• Container: Docker image built with Python 3.12-slim
• WSGI Server: Gunicorn with 2 workers and 4 threads
• Package Manager: uv for fast dependency installation
• Infrastructure as Code: Terraform for repeatable deployments

See terraform/README.md for complete Terraform configuration details.

Service Configuration

The Cloud Run service is configured with:

• Memory: 2Gi (sufficient for API responses and data processing)
• CPU: 2 cores (handles concurrent requests efficiently)
• Timeout: 120 seconds (allows time for Deadlock API calls and chart generation)
• Scaling: 0 to 10 instances (scales to zero when idle for cost savings)
• Authentication: Public access (no authentication required)

Acknowledgments

• Powered by Deadlock API
• Built with Python, Plotly and Flask