Chapter 28 — Discussion Guide

Opening question

"Your team goes for it on fourth-and-1 and gets stopped. You lose by four. Was going for it a mistake?"

Take a show of hands first, then answers. Most of the room will say "it depends," and the follow-up ("depends on what?") is where the session starts. Expect three camps: those who judge by the result, those who judge by the odds, and a few who want to know what play was called.

Why it works: the third camp is the chapter's answer. Austin's Week 10 fourth-and-1 at the Portland 18 was stopped for no gain, and on Monday it was graded decision right, call reasonable, execution lost the rep, in that order. Students who wanted to know the call before judging have already found Wade Harmon's question: "What's the call if we go?" Write the three camps on the board and come back to them after §28.7.


Discussion questions

Level 1 — Recall

1. In plain English, what are expected points, and what is EPA? Listen for: EP is what a down, distance and spot are worth on average to the team with the ball. EPA is the change in that value on one snap. Third-down results, turnovers and explosives swing it far more than ordinary early-down gains. Probe: "Why is an interception at the opponent's 35 one of the most expensive plays in football?" (Exercise B8b: roughly −4.)

2. What are Marisol's five honesty rules, and what are the four lines of her slide? Listen for: counts first; sample and caveat; "This doesn't know…"; separate the decision from the result; "I don't know." The slide reads THE NUMBER / THE SAMPLE / THIS DOESN'T KNOW… / WHAT I'D DO. Probe: "Which line do coaches skip most often, and what does skipping it cost?"

Level 2 — Analysis

3. Play-action averaged about 8.6 yards per dropback against 6.2 for straight dropbacks. Why isn't that proof that play-action is better? Listen for: the selection trap. Coordinators call play-action on early downs and in good situations; dropbacks carry third-and-long. Comparing early downs only narrows the gap to about 8.6 vs. 7.0. Compare like with like: same down, distance band, box count and score. Probe: "Why does Marisol code RPOs as their own family?" (Quiz item 10.)

4. Lyle won the "grade the family" argument and lost the reset-run argument. What was different about the evidence? Listen for: the Inside family ledger (plain Inside about −0.02 per play, Slant/Glance about +0.19, Inside PA about +0.41, family about +0.12) showed value moving between plays that share a picture, and Teo's Holloway card showed which defender moved. The reset run had no such partner: 7 of 25 successful, leaving an average third-and-7. Marisol's own caveat on the win: "twenty-five snaps of one safety." Probe: "She told Lyle, 'You win the call sheet. You don't win the argument.' What did she mean?"

5. Why does Wade say "Four-down" before the third-down call instead of after it? Listen for: the fourth-down answer belongs in the third-down call. Once fourth down is available, third-and-2 is one of two downs, so a play-action shot becomes a near-free down (exercise C1). Russ's spring line shows the defense knows it too: "When you're four-down, I'm not defending third down." Probe: "Which third-down call becomes worse once you know you're going on fourth?" (Anything with sack risk and no payoff.)

Level 3 — Judgment and ethics

6. A head coach says, "Analytics doesn't know my defense is playing great today. When we're playing good defense, I punt." Does that pass "Name it, then price it"? Listen for: it names something, which is a start, but it rarely prices it. How much does a good defensive day change the opponent's chance of scoring from its own 17 versus your 43? Usually the change is too small to flip a fourth-and-1 decision. Strong students separate this from fear of blame, the other kind of gut. Probe: "Wade kicked or punted 11 times when the chart said go, and went twice when it said kick. Is a head coach who overrides that often a problem?"

7. Vance's designed-carry budget is about three a game. It's the fourth quarter, he has three, and the sneak is clearly the best fourth-and-1 call. Do you sneak? Listen for: the chapter's 🩺 note says to count the sneak. A good answer weighs the playoff picture, the conversion chance with and without the sneak, and the hit he takes in a pile, and then writes a rule before the game (exercise D3). Probe: "Who should own that rule: the play caller, the quarterbacks coach, or the head coach?"

8. The opponent reads the data too. Is it honest to break a tendency on purpose just to make the opponent's numbers wrong? Listen for: yes; it's strategy, not deception of anyone owed the truth. The sharper point is the reverse: the Sentinels' 29-straight two-high streak may have been built to be trusted. Marisol puts a question mark on perfect streaks for that reason. Probe: "What would you have written beside that row before Game 3?" (Exercise M4.)


Structured role-play: the bye-week self-scout

Setup: the Week 9 bye. Marisol puts up her slide titled "2nd-and-8+: what happens next." Designed runs on second-and-long, Weeks 1–8: 7 of 25 successful, average next third down about 3rd-and-7. Everything else: 17 of 38, about 3rd-and-5. Her line: "The best third down we have is worse than the worst second down we have."

Roles: - Marisol (analytics): presents the four-line slide and must fill in "This doesn't know…" honestly, including what the sample can't tell. - Lyle (play caller): defends the reset run as his comfort call. He must name, and price, one thing the number doesn't see. - Gus (line coach): argues the run "keeps the edge rushers honest" for third down and must propose a way to test his own claim (exercise D2). - Wade (head coach): listens, asks one question of each, and decides.

Rules: every claim needs a count or it doesn't count. Nobody may argue from the result of a single game. Lyle may lose, but he must say what he'd change on the call sheet if he does. Eight minutes, then rotate roles once.

Debrief: compare the class's outcome with the chapter's. In the book, the reset run became half of a Kill call and the second-and-long box rule tightened to six or fewer. Ask: "Did anyone win the argument, or did the call sheet win?"


Connecting to the case studies

Case Study 1, "The Fourth Down That Failed and Was Still Right," is Week 10 at Portland: up 17–10 late in the third quarter, the Trap and Drive drive reached the Portland 18, and Duo on 4th-and-1 was stopped for no gain when Dion Pratt slanted into the right A gap. Portland drove for a field goal; Austin won 27–13 anyway. Have students write the Monday review in three lines before they read the case's (exercise M1). Teo's later note, that Pratt slanted on 3 of 9 short-yardage snaps that season, is a good test of whether a student treats a small count as a fact or a lean.

Case Study 2, "Super Bowl LII: Two Fourth Downs and a Reputation," looks at two widely documented Philadelphia decisions using the book's illustrative values. Set the two cases side by side: one decision failed and was right; the other two worked and changed how a league talked about fourth down. Ask what the reputation was built on, the decisions or the results. Keep the discussion to the documented game facts; the case attributes no quotes to real people.


Closing synthesis prompt

"Pick one fourth-down box on your chart that says GO. Write the call that makes it true and the way that call most often fails. If you can't write the call, what should the box say?"

Students who leave the call blank have found the model's first assumption. Close with the three camps from the opening question and Marisol's sentence: "I'm not here to make the decision. I'm here to make sure you know what you're deciding."