Mistake Gravity Index

Inspiration

Coaches review thousands of post-match events, but only a small number of deaths actually decide games. Identifying those moments is manual, subjective, and time-consuming.

What it does

Mistake Gravity Index analyzes official GRID match data to: Filter out traded deaths Detect deaths under objective pressure Rank mistakes by strategic impact using an explainable scoring model The result is an automated game review agenda, not raw statistics.

How I built it

Python CLI built in JetBrains PyCharm Official GRID APIs (Central Data, Series State, Events) Deterministic scoring logic (no black-box predictions) Snapshot tests to guarantee analytical stability Rich-based CLI output for judge-readable demos Junie assisted with refactoring, test design, and compliance checks.

Challenges I ran into

Making sense of dense event streams Avoiding misleading “near objective” noise Designing a scoring model coaches can trust and explain

Accomplishments I'm proud of

Fully explainable “assistant coach” logic Automated macro review generation Production-grade CLI with real match data Snapshot testing for analytical correctness

What I learned

Raw data isn’t insight. Coaches need prioritization, context, and reasoning—not dashboards.

What’s next

Player-specific trend reports Multi-match aggregation Optional “what-if” modeling layered on top of stable insights

Built With

• centraldata
• events
• events)
• grid-esports-apis
• grid-esports-apis-(central-data
• jetbrains-junie
• jetbrains-pycharm
• pytest
• python
• rich
• series-state
• seriesstate