I loaded 399,906 ranked boards into Snowflake to see what separates top‑4 finishes.
Read the analysisSource: 399,906 player boards from 49,977 Set 3 ranked matches (Korean server, Platinum to Challenger), cleaned and queried in Snowflake. Players who are already ahead level faster, so this is an association, not a cause.
A PostgreSQL star schema and Power BI dashboard over 63,635 order items found 7.62% average discounting and a 7.26% return rate in Electronics.
PostgreSQL, SQL, Power BI, DAX
| Name | Return rate |
|---|---|
| Electronics7.62% average discount | 7.26% |
| Accessories44.21% gross margin | 5.55% |
Source: 63,635 order items, $5.94M net revenue, modeled in PostgreSQL.
I scored players on Value Over Replacement so every position shares one scale, then used K-Means to split them into four draft tiers: 22 elite, 36 high-end starters, 73 average, 32 below replacement.
Python, Scikit-Learn, K-Means, Pandas
| Name | Players |
|---|---|
| Elite | 22 |
| High-end starters | 36 |
| Average | 73 |
| Below replacement | 32 |
Source: 163 players in the repo's 2024 sample dataset, K-Means with k=4 on Value Over Replacement.
Ran sprints in Jira, handled client communication, and led a usability audit of the service pages for Woodland Country Manor.
Email is the quickest way to reach me.