Exercises: Introduction to College Football Analytics

These exercises progress from foundational concept checks to challenging applications. Estimated completion time: 2-3 hours.

Scoring Guide: - ⭐ Foundational (5-10 min each) - ⭐⭐ Intermediate (10-20 min each) - ⭐⭐⭐ Challenging (20-40 min each) - ⭐⭐⭐⭐ Advanced/Research (40+ min each)

Part A: Conceptual Understanding ⭐

Test your understanding of core concepts. No calculations required.

A.1. In your own words, explain the difference between "statistics" and "analytics" in the context of college football. Give an example of each.

A.2. A coach says, "I don't need analytics—I've been coaching for 30 years and I know football." How would you respond to this perspective? What value can analytics add even for experienced coaches?

A.3. Explain why a 5-yard gain can be considered either a success or a failure depending on context. What information do you need beyond the yardage to evaluate the play?

A.4. True or False: Analytics should completely replace traditional scouting in college football programs. Justify your answer with at least two reasons.

A.5. Describe the five stages of the analytics workflow. For each stage, provide a brief example related to evaluating a quarterback's performance.

A.6. List three specific decisions that a college football coach makes where analytics could provide useful input. For each, explain what data might be relevant.

A.7. What does "expected points" measure? Why is it useful for evaluating plays of different types?

A.8. A journalist writes: "The team's star running back averaged 5.2 yards per carry, proving he had an excellent season." What contextual information is missing that analytics would consider before making this judgment?

Part B: Application ⭐⭐

Apply concepts to realistic scenarios.

B.1. The following table shows traditional statistics for two quarterbacks:

Based on traditional statistics alone, which quarterback appears better? Now, list at least four pieces of contextual information that analytics would want to know before making a fair comparison.

B.2. Consider the following game situations. For each, indicate whether going for a fourth down conversion or punting would likely be recommended by analytics (assume typical conversion rates):

a) 4th and 2 at your own 32-yard line, tied game, 8 minutes left in 2nd quarter b) 4th and 1 at opponent's 35-yard line, down by 3, 2 minutes left in 4th quarter c) 4th and 6 at opponent's 40-yard line, up by 14, 5 minutes left in 3rd quarter d) 4th and 3 at your own 45-yard line, down by 7, 10 minutes left in 4th quarter

Explain your reasoning for each.

B.3. A college program's analytics department provides the following report to coaches:

"Based on analysis of 500 rushing plays, Zone Left runs have averaged 5.8 yards per carry compared to 4.1 yards for Zone Right. We recommend calling more Zone Left plays."

What questions would you want answered before accepting this recommendation? List at least four considerations.

B.4. Explain how Expected Points Added (EPA) might evaluate the following three plays differently than traditional statistics:

a) A 20-yard pass on 3rd and 25 (team still punts) b) A 4-yard run on 3rd and 3 (first down achieved) c) A 50-yard pass that results in a touchdown in garbage time (winning team up by 35)

B.5. A recruiting analyst has built a model that predicts which high school players will become NFL draft picks. The model shows 75% accuracy on historical data. List at least three potential problems with this model that should be considered before using it for recruiting decisions.

Part C: Critical Thinking ⭐⭐-⭐⭐⭐

Analyze scenarios and justify your reasoning.

C.1. Read the following argument and identify at least two flaws in the reasoning:

"Our team went 11-2 last season while ignoring analytics, so clearly analytics isn't necessary for winning. Other teams who use analytics don't win any more than we do."

C.2. A newspaper headline reads: "Analytics Says Teams Should Never Punt." Based on what you've learned in this chapter, evaluate this claim. Is it accurate? What nuance is the headline missing?

C.3. Consider the ethical implications of the following scenarios. For each, explain whether you think the action is appropriate and why:

a) An analytics intern downloads publicly available play-by-play data and shares his analysis on social media, including insights about opponent tendencies before a big game.

b) An analytics staff member uses injury data obtained informally from a trainer to adjust win probability projections.

c) A program publishes player tracking data showing how fast each player runs, without asking players' permission.

C.4. Compare and contrast how analytics might be used differently by: - A top-5 program with a large budget and many scholarships filled - A mid-tier program trying to compete with fewer resources - A rebuilding program focused on long-term development

C.5. A coach disagrees with an analytics recommendation to attempt a two-point conversion. He says, "The model doesn't know my kicker—he's automatic on extra points." How should the analyst respond? When is it appropriate to override model recommendations?

Part D: Research and Extension ⭐⭐⭐-⭐⭐⭐⭐

Open-ended problems requiring additional research.

D.1. ⭐⭐⭐ Research the history of one of the following and write a 300-500 word summary of how it developed and its current influence on football analytics:

a) Football Outsiders and the development of DVOA b) Expected Points models from Burke to nflfastR c) The New York Times' 4th Down Bot d) PFF grades and their methodology

D.2. ⭐⭐⭐⭐ Interview someone who works in sports analytics, sports journalism, or coaching (even at a high school or amateur level). Ask them: - How do they use data in their work? - What do they see as the benefits and limitations of analytics? - How has the role of data changed during their career?

Write a 500-word summary of what you learned.

D.3. ⭐⭐⭐ Find a recent article from a sports analytics publication (The Athletic, Football Outsiders, ESPN Analytics, etc.) that uses advanced metrics. Summarize: - What question is the article addressing? - What data and methods are used? - What conclusions are drawn? - What limitations or alternative interpretations exist?

D.4. ⭐⭐⭐⭐ Design an analytics project for a hypothetical college football program. Your proposal should include: - A specific question to answer - What data would be needed - How the data would be analyzed - How results would be communicated to coaches - What decisions the analysis would inform - Potential ethical considerations

Write this as a 1-2 page proposal.

D.5. ⭐⭐⭐ Compare the analytics adoption between two college football conferences (e.g., SEC vs. Big Ten, or Power 5 vs. Group of 5). Research: - Which teams have known analytics departments? - How do fourth-down attempt rates compare? - What evidence exists of analytics influencing strategic decisions?

Summarize your findings in 400-600 words.

Part E: Discussion Questions

Questions for classroom discussion or self-reflection.

E.1. As analytics becomes more widespread, will it reduce competitive advantage for teams that use it well, or will there always be opportunities to gain edges? Explain your reasoning.

E.2. Some argue that analytics makes football less entertaining because it makes decisions more "obvious." Do you agree? Is there value in maintaining traditions even when analytics suggests different approaches?

E.3. How might analytics change the fan experience of watching college football? Consider both positive and negative potential impacts.

E.4. If you were a college football head coach with limited resources, would you hire an analytics person or spend that salary on another position coach? Justify your choice.

E.5. Where do you see college football analytics in 10 years? What new data sources or methods might transform the field?

Solutions

Selected solutions are available in: - appendices/g-answers-to-selected-exercises.md (odd-numbered problems)

Full solutions available to instructors upon request.