PYTHON · STATISTICAL MODELING · REAL-TIME DATA

Sports Pricing & Quality Control Engine

Monitored live performance metrics across 5+ sports simultaneously, using statistical models to catch pipeline anomalies before they impacted pricing decisions — in real time.

RoleQC Analyst · DraftKings / Scouting Heroes
TimelineJuly 2024 – May 2025
StackPython · Statistical Modeling · Excel

📌 The Challenge

In a live sports data environment, a single bad data point can cascade into incorrect pricing across thousands of lines. Speed and accuracy are in direct conflict — and both matter completely.

My role was to be the last line of defense: catching errors in live pipelines before they reached the pricing layer, while simultaneously making cross-functional decisions under time pressure.

🏗️ What I Did

  • Monitored live ingestion pipelines across NFL, NBA, MLB, NHL, and golf, tracking data freshness, completeness, and consistency in real time
  • Developed and applied statistical validation rules to flag outliers and anomalous inputs before they could influence live pricing models
  • Collaborated cross-functionally with traders and operations teams, communicating complex data issues clearly and quickly in high-stakes situations
  • Executed quality assurance testing cycles to identify systematic errors in upstream data sources

📊 Outcomes

5+
Sports monitored simultaneously in real time
Fast
Sub-minute anomaly detection and escalation
Zero
Target: pricing errors from undetected data issues

🛠️ Technologies Used

PythonStatistical Modeling Real-time Data PipelinesQA Testing Frameworks Cross-functional ReportingExcel

💡 Key Learnings

  • High-pressure data environments demand communication skills as much as technical ones — a silent analyst with the right answer helps nobody
  • Statistical thinking isn't just for models — it's how you distinguish signal from noise when everything is moving fast
  • Working across sports taught me to adapt analytical frameworks quickly to new domains without starting from scratch