Key Takeaways: Introduction to College Football Analytics


One-Page Summary

What Is Sports Analytics?

Analytics = Systematic data analysis to find patterns that inform decisions

Analytics differs from traditional statistics by adding: - Context (a 5-yard gain means different things on 3rd-and-3 vs. 3rd-and-10) - Expectations (was that performance better or worse than expected?) - Prediction (what will happen next?) - Attribution (who deserves credit or blame?)


The Evolution

Era Approach Limitation
Pre-2000s Box score statistics No context
2000s Rate statistics, efficiency Limited prediction
2010s+ EPA, Win Probability, ML Computational power enabled sophistication

Key milestone: Moneyball (2003) showed analytics could find undervalued players.


Core Concepts

Expected Points (EP) - Average points scored from a field position - Opponent's 5-yard line ≈ +6 EP - Own 20-yard line ≈ +0.5 EP

Expected Points Added (EPA) - Change in EP from one play - EPA = EP(after) - EP(before) - Common currency for comparing all plays

Win Probability (WP) - Chance of winning from current game state - Depends on: score, time, field position, down/distance

Win Probability Added (WPA) - Change in WP from one play - Measures clutch performance


How Programs Use Analytics

Application Example
Opponent Analysis Identify tendencies by down/distance
Self-Evaluation Compare performance to expectations
In-Game Decisions Fourth down, two-point conversions
Recruiting Predict prospect development
Player Development Track improvement over time

The Analytics Workflow

┌──────────────────┐
│ 1. QUESTION      │ → What do we want to know?
├──────────────────┤
│ 2. DATA          │ → What information do we need?
├──────────────────┤
│ 3. PROCESSING    │ → Clean, transform, integrate
├──────────────────┤
│ 4. ANALYSIS      │ → Apply appropriate methods
├──────────────────┤
│ 5. COMMUNICATION │ → Make findings actionable
└──────────────────┘

Analytics vs. Traditional Scouting

Traditional Scouting Analytics
Expert observation Systematic data analysis
Qualitative insights Quantitative evidence
Deep on individuals Broad across many
Captures intangibles Captures measurables

Best practice: Integrate both approaches


Ethical Considerations

  • Privacy: Respect player data rights
  • Honesty: Present findings accurately, with uncertainty
  • Credit: Acknowledge sources and contributions
  • Humanity: Remember data represents real people

Key Terms Quick Reference

Term Definition
Analytics Systematic data analysis for decision support
EPA Expected Points Added—value of a play
WP Win Probability—chance of winning from game state
WPA Win Probability Added—clutch impact of a play
Play-by-play Detailed record of every play
Context-aware Accounting for game situation

Decision Checklist

Before starting an analytics project, ask:

  • [ ] Is my question specific and answerable?
  • [ ] Does relevant data exist?
  • [ ] Will the answer inform a decision?
  • [ ] Have I considered ethical implications?
  • [ ] Do I understand the limitations of my data?

What to Remember

  1. Analytics complements, doesn't replace, football knowledge
  2. Context matters—same statistic can mean different things
  3. Expected value thinking enables better decisions
  4. The analytics revolution is changing football at all levels
  5. Ethical responsibility accompanies analytical capability

Looking Ahead

Chapter 2 introduces the data landscape of college football: - Where data comes from - How to access it - What its limitations are - How to start building your data toolkit