Scout Report Generator

Inspiration

In esports, teams win by knowing how their opponents play. But watching many matches and writing notes takes a lot of time.

We wanted to build a tool that automatically studies past matches and gives a clear scouting report, just like a coach would do.

What it does

Scout Report Generator:

Reads match data from GRID

Understands what happens in each round

Tracks what each player does

Finds common strategies and habits

Creates an easy-to-read scouting summary

It helps teams prepare faster and smarter for upcoming matches.

How we built it

We built the project step by step:

Collected raw match events from GRID

Cleaned and normalized the data

Grouped events by rounds

Calculated player stats like kills, abilities, and bomb actions

Detected strategies using simple rules

Designed the system so AI can be added later

Everything is written in Python and works from the command line.

Challenges we ran into

GRID data is very large and complex

Events are not always in a simple order

Some rounds have missing or repeated data

Understanding game logic from raw events was tricky

We solved this by adding debug logs, testing small parts first, and building a clean structure.

Accomplishments that we're proud of

Built a working end-to-end pipeline

Converted raw events into meaningful insights

Made the system support VAL and LoL

Designed it to work even without AI

Created a clean project structure like a real production app

What we learned

How esports match data works

How to design clean and reusable code

How to turn raw data into useful insights

How to think like an analyst and a coach

How to build systems that can grow in the future

What's next for Scout Report Generator

Next, we plan to:

Add AI-generated scouting reports

Create visual dashboards

Compare teams automatically

Support more games

Make it usable by real esports teams

Our goal is to make scouting faster, smarter, and automatic.

Built With

• ai
• grid
• pycharm
• python