Stratbase is a project I built to be the tool I always wanted when I was grinding ESEA leagues in my teens. It started as a simple document editor and has since grown to include a cutting edge automated grenade lineup database and a 2D demo viewer that sits right alongside industry leaders.

The Editor

How it all started. A custom built strat editor, based on TipTap. Enhanced with features like mentions, linking to players, nade and demo viewer embeds. Everything you need to put together strats for your team. That data can then be used to customise the view for players, extracting their specific roles out of the documents to give them an inline refresher on their positions.

Nade Database

The first major innovation to land on Stratbase. I built a pipeline to extract nade lineups using the demoinfocs-golang library, then processing them to determine the most commonly thrown. Once these are identified and grouped, my automatic demo renderer will fire up an instance of CS2, automate it to record a clip of the nade being thrown, then upload it and display it on a map directly in Stratbase. I later enhanced this by using ray-tracing to discover safe camera angles to show where the nade lands.

2D Demo Viewer

Building on top of the existing demo pipeline, I expanded the data collection to build a JSON payload that could be consumed by Vue on the frontend, orchestrating an svg layer to replay demos in the browser. This allows players, coaches and analysts to gain key insights in a fraction of the time it takes to fire up the game and analyse the demos that way. I then integrated a multi-view, allowing users to stack multiple replays on top of each other to discover patterns. They can then use filters to cut out the noise and get actionable insight into their opponents patterns.

I then integrated the Pro Nade Database. See that video icon on the screenshot? You click that and it opens the nade lineup, visible right there in the browser.

What's Next?

Next up for Stratbase involves integration with more demo providers, like FACEIT/ESEA, as well as dumping demo events into an analytics database like Clickhouse to allow massive statistics access for users.