Chapter 28: Case Study 1 - Building an NBA Team Analytics Department from Scratch

Introduction

This case study examines how a fictional NBA team, the "Expansion City Pioneers," builds their analytics department from the ground up. The team has been awarded an expansion franchise and has two years before their first season to build all basketball operations functions, including analytics. This scenario illustrates the practical decisions, tradeoffs, and strategies involved in creating effective analytics infrastructure.

Part 1: Context and Objectives

The Situation

The Pioneers ownership group has committed to building a "modern, analytically-driven organization." The General Manager, hired from a progressive existing franchise, has been given a substantial budget and mandate to create best-in-class basketball operations. The Head of Analytics will report directly to the GM and serve on the basketball operations leadership team.

Strategic Objectives

1. Support the Expansion Draft: Analyze available players for the expansion draft
2. Prepare for the NBA Draft: Build draft evaluation infrastructure for the first two drafts
3. Create Decision-Support Systems: Develop tools for player acquisition, contract evaluation, and game strategy
4. Establish Data Infrastructure: Build sustainable data pipelines and storage systems
5. Integrate with Basketball Operations: Create effective working relationships with coaching and scouting

Budget and Timeline

• Year 1 Budget: $2.5 million (salaries, technology, data acquisition)
• Year 2 Budget: $3.0 million
• Ongoing Budget: $2.0 million annually
• Staff target: 10-12 full-time employees by start of first season

Part 2: Department Design

Organizational Structure

After researching successful analytics departments across the league, the leadership designs the following structure:

VP of Basketball Analytics
├── Director of Player Evaluation
│   ├── Senior Analyst - College Scouting
│   ├── Analyst - Pro Personnel
│   └── Research Analyst
├── Director of Basketball Strategy
│   ├── Senior Analyst - Game Strategy
│   └── Analyst - Player Development
├── Director of Data Engineering
│   ├── Data Engineer
│   └── Software Developer
└── Cap Analyst

Role Definitions

VP of Basketball Analytics - Strategic leadership of department - Integration with GM, coaching staff, and ownership - Budget management and resource allocation - Final approval on major methodological decisions - External representation (media, league meetings)

Director of Player Evaluation - Owns all player evaluation models and processes - Manages draft board creation - Coordinates with scouting department - Leads trade and free agency analysis

Director of Basketball Strategy - Owns game preparation analytics - Manages relationship with coaching staff - Leads player development measurement - Oversees in-season analysis workflow

Director of Data Engineering - Manages all technical infrastructure - Ensures data quality and availability - Leads tool development for analysts - Maintains security and access controls

Hiring Philosophy

The department establishes core hiring principles:

1. Diverse backgrounds: Mix of sports industry veterans and those from other analytical fields
2. Technical depth: All analysts should be strong programmers, with some specialists
3. Basketball fluency: Everyone must develop strong basketball understanding
4. Communication skills: Equal weight to technical and communication abilities
5. Cultural fit: Collaborative, humble, curious personalities

Part 3: Hiring Process

Year 1 Hiring Plan

Phase 1 (Months 1-3): Leadership - Hire VP of Basketball Analytics - Hire Director of Data Engineering

Phase 2 (Months 4-6): Core Team - Hire Director of Player Evaluation - Hire Director of Basketball Strategy - Hire 2 Data Engineers/Developers

Phase 3 (Months 7-12): Build Out - Hire Senior Analysts (2) - Hire Analysts (2) - Hire Cap Analyst

Recruitment Strategy

Sourcing Channels: - TeamWork Online postings - Direct outreach to known analysts in other organizations - MIT Sloan Sports Analytics Conference networking - University recruiting (top graduate programs) - Public portfolio review (Twitter, GitHub, blogs)

Interview Process: 1. Resume and portfolio review 2. Initial phone screen (30 minutes) 3. Technical assessment (take-home project, 4-6 hours) 4. On-site interviews: - Technical deep-dive (90 minutes) - Basketball knowledge assessment (45 minutes) - Case study presentation (60 minutes) - Culture/fit conversations (multiple 30-minute sessions) 5. Reference checks 6. Executive meeting with GM

Compensation Philosophy

The organization decides to pay above-market rates to attract top talent:

Benefits include relocation assistance, professional development budget ($5,000/person annually), conference attendance, and flexible work arrangements where possible.

Part 4: Technology and Infrastructure

Data Acquisition

League-Provided Data: - Second Spectrum tracking data (included in league revenue sharing) - NBA Stats API access - League-wide injury data

Third-Party Data: - Synergy Sports subscription ($75,000/year) - Sports Info Solutions college data ($25,000/year) - International scouting data feeds ($15,000/year) - Medical/biometric data systems ($50,000/year)

Internal Data Collection: - Practice tracking system installation ($200,000) - Video analysis infrastructure ($100,000) - Player wearables program ($75,000)

Technical Stack

Data Infrastructure: - Cloud: AWS (primary), GCP (secondary) - Data Warehouse: Snowflake - Data Pipeline: Apache Airflow - Version Control: GitHub Enterprise

Analytics Tools: - Languages: Python (primary), R (secondary), SQL - Visualization: Tableau, Custom D3.js dashboards - ML Platform: SageMaker for model deployment - Collaboration: Jupyter Hub, Confluence

Applications: - Custom internal web applications (Flask/React) - Mobile apps for coaching staff - Integration with existing basketball ops systems

Security Considerations

• Strict access controls with role-based permissions
• Two-factor authentication required
• No external sharing of proprietary data or methods
• Regular security audits
• Clear data retention and destruction policies

Part 5: Process Development

Draft Evaluation Process

The team develops a comprehensive draft evaluation workflow:

Data Collection (October - March): - Aggregate all available college/international statistics - Collect physical measurables and athletic testing - Catalog video observations from scouts - Track medical evaluations as available

Model Development (November - February): - Build statistical translation models for college-to-NBA projection - Create physical/athletic comparison models - Develop character/makeup scoring frameworks - Combine into overall player projection model

Board Building (March - May): - Generate initial rankings from statistical models - Integrate scouting observations - Conduct internal debates and adjustments - Produce tiered draft board with confidence intervals

Draft Preparation (May - June): - Scenario modeling for different draft positions - Trade value calculations - Real-time draft room decision support

Game Strategy Process

Pre-Season: - Establish baseline metrics and dashboards - Train coaching staff on analytics tools - Define key questions strategy analytics will address

Game Preparation (Weekly): - Opponent analysis delivered 3 days before games - Video cuts linked to statistical findings - In-person strategy meetings with coaches - Pregame refresher materials

In-Game Support: - Real-time lineup analysis - Timeout insights delivered to coaches - Halftime adjustments recommendations

Post-Game Review: - Statistical game summary within 2 hours - Detailed analysis available next morning - Weekly trend analysis

Integration with Scouting

The analytics department establishes formal integration processes:

• Joint meetings with scouting every 2 weeks during season
• Analytics representative at all major scouting events
• Shared evaluation framework combining stats and observation
• Regular feedback loops on prediction accuracy
• Respect for complementary expertise

Part 6: First Year Results

Expansion Draft Outcome

Using comprehensive analysis of available players, the Pioneers: - Identified undervalued rotation players other teams exposed - Avoided players with concerning underlying metrics despite surface stats - Built tradeable asset collection for future flexibility

Key Success: Acquired a player projected as fringe starter that models identified as likely improvement candidate. Player became an All-Star within three years.

NBA Draft Outcomes

First draft yielded: - Top-10 pick who met statistical expectations - Second-round find identified through international data analysis - One miss on a late first-rounder (character concerns not fully weighted)

Lessons Learned: Integrate character/makeup evaluation more systematically; develop better international projection models.

Organizational Learning

What Worked: - Strong relationship built with coaching staff through consistent value delivery - Data infrastructure enabled rapid iteration and analysis - Diverse team backgrounds brought varied perspectives - Above-market compensation attracted strong candidates

What Needed Improvement: - Initial over-emphasis on technical sophistication vs. practical application - Some tools built that weren't actually used by stakeholders - Communication style needed adjustment for different audiences - Work-life balance during season was challenging

Part 7: Ongoing Evolution

Year 2 Adjustments

Based on first-year experience, the department made adjustments:

Structural Changes: - Added analyst dedicated to player development - Created formal coaching staff liaison role - Established rotation for game travel

Process Changes: - Simplified game preparation outputs - Increased in-person time with scouts - Developed better feedback mechanisms - Created tiered analysis (quick takes vs. deep dives)

Technology Changes: - Rebuilt main dashboard based on user feedback - Improved mobile access for coaches - Created self-service tools for common queries

Long-Term Vision

By year five, the department aims to: - Be recognized as top-five analytics department in the league - Have contributed to at least one championship-contending roster construction - Have developed multiple analysts who advance to other organizations - Have created reusable tools and methods that become industry standard - Maintain strong relationships across basketball operations

Lessons for Aspiring Analysts

This case study illustrates several key lessons:

1. Technical excellence is necessary but not sufficient: The best analysts combine technical skills with communication ability and basketball understanding.

2. Relationships matter enormously: Success depends on trust and collaboration with coaches, scouts, and executives.

3. Practical application trumps sophistication: Simple, actionable insights often more valuable than complex analyses.

4. Iteration is essential: First versions are rarely right; building feedback loops enables improvement.

5. Culture determines outcomes: Hiring for fit and maintaining healthy dynamics matters as much as individual talent.

6. Resources enable but don't guarantee success: Budget and tools provide foundation but execution determines results.

Discussion Questions

1. How would you prioritize hiring if budget constraints limited the team to six analysts instead of twelve?

2. What metrics would you use to evaluate the analytics department's success?

3. How should the department handle disagreements between statistical projections and scout observations?

4. What are the ethical considerations in player evaluation that the department should address?

5. How would the approach differ for a rebuilding team versus a contending team?