The “Manual” Problem: Why We Built a Scout for the Chaos of Esports
Esports is a data goldmine, but for most teams, that gold is buried under hours of manual VOD reviews and messy spreadsheets. I’ve spent…
The “Manual” Problem: Why We Built a Scout for the Chaos of Esports
Esports is a data goldmine, but for most teams, that gold is buried under hours of manual VOD reviews and messy spreadsheets. I’ve spent way too much time watching pro matches of League and VALORANT only to realize that even at the highest levels, a huge chunk of strategy is still based on “vibes” and gut feeling.
That’s why we started building Tactical Scout.
The goal was simple: Could we turn raw match data into a strategic “cheat sheet” that actually makes sense to a coach in the heat of a draft?
The “Oh No” Moment in Development
When we entered the Cloud9 × JetBrains hackathon, we weren’t just looking for a trophy. We wanted to solve the specific friction of scouting.
If you’re a coach, you need to know three things instantly:
- What does this team do in the first 10 minutes?
- Which player is the “weak link” under pressure?
- If they pick X champion, what is our % chance of crumbling?
Building this meant wrestling with raw data from GRID. If you’ve ever looked at raw match logs, you know they are a nightmare of timestamps and coordinate points. Translating that into a “readable story” was our biggest hurdle.
Building in the Flow State
One thing that kept us from losing our minds during the sprint was the tooling. I’ve used plenty of IDEs, but leaning into the JetBrains ecosystem for this project actually changed the pace.
Specifically, using Junie (the AI assistant) didn’t feel like “cheating” or just generating code; it felt like having a junior dev who was really good at the boring stuff. While I focused on the logic of Draft Risk Evaluation, Junie was handling the boilerplate and API structures.
[IMAGE: A screenshot of your code or the JetBrains IDE interface showing a complex function] Caption: Digging into the analytics logic — this is where the “gut instinct” gets turned into data.
Why “Tactical Scout” is Different
We didn’t want another dashboard full of graphs that no one looks at. We focused on two specific “survival” tools:
- Pattern Recognition: Instead of saying “this player has a 2.0 KDA,” the tool says “this player plays aggressive near the Dragon pit at the 6-minute mark.” That is a narrative, not just a stat.
- The Draft Warning System: It analyzes team synergy. If your comp is too “magic damage heavy,” the tool flags it before you lock in.
Lessons from the Trenches
If I learned anything from this hackathon, it’s that data science in gaming is actually storytelling.
A coach doesn’t need a math degree; they need to know how to win the next 30 minutes. By removing the friction of manual data entry using tools like JetBrains, we were able to spend our time on the meaning of the data rather than the plumbing of the code.
What’s the Next Play?
We’re looking at real-time draft simulations next. The dream is to have Tactical Scout running live during a match, predicting the enemy’s next move before they even click it.
Innovation isn’t always about the flashiest tech — sometimes it’s just about making a hard job a little bit easier for the people who love the game.
메타데이터
- post_id
- b00634bc37be
- slug
- the-manual-problem-why-we-built-a-scout-for-the-chaos-of-esports-b00634bc37be
- url
- https://medium.com/@suimanga01/the-manual-problem-why-we-built-a-scout-for-the-chaos-of-esports-b00634bc37be
- canonical_url
- https://medium.com/@suimanga01/the-manual-problem-why-we-built-a-scout-for-the-chaos-of-esports-b00634bc37be
- author_url
- https://medium.com/@suimanga01
- status
- ok
- fetched_at
- 2026-07-13 06:23:13
