Case Study 1: The Audit — Grading Two Halftimes on a Monday
Austin's view, after a loss. All teams, players, coaches, and numbers are fictional: the Austin Outlaws and Sacramento Sentinels live inside real NFL rules, but every name, snap, and statistic here is invented.
Setup
When and where: The Monday after Week 15, about one in the afternoon, in the offensive staff room in Austin. We'd lost at Sacramento the day before, 23–20. Gus Ferrante had already run the third-quarter sack clip for his linemen nineteen times at 6:40 that morning. I'd stood in his doorway with a coffee and watched him do it.
Who was in the room: Me; Teo Arriaga, our quality control coach; Marisol Vega, our director of analytics; Gus; Rome Pettaway, our quarterbacks coach; and Wade Harmon, our head coach, who came in late with his own coffee and sat on the edge of the table.
What we were doing: A halftime audit. We do one after every game, win or lose. It takes about twenty minutes. We grade our halftime changes, and, as best we can from the film, the other side's. We don't grade them by the score. We grade them by the decision.
What we knew going in:
- Halftime score: Sentinels 13, Outlaws 10.
- My halftime sheet said: "They're giving us the quick game, and they'll take it away." "94 is beating our left tackle." "Twelve isn't working."
- My three changes: Trips with Sutton alone into the boundary and the tight end as #3 on a drag; chip Rollins with Ruffin; less 12 personnel.
- On Sunday night's film I'd seen three things change in the Sentinels after halftime: their corners pressed far more on early downs, their safeties rotated out of two-high far more on early downs, and on our first third-and-7 of the second half, at 6:51 of the third quarter, they ran the double-A mug with the corner fire. (I didn't know until the spring that those were exactly Russ's three halftime changes. On Monday, I knew only what the film showed.)
What happens
Marisol started, as she always does, with one slide. It had one line on it.
SECOND HALF: OUTLAWS 10, SENTINELS 10.
WEEK 15, SECOND HALF -- the scoring and the snaps that decided it
Q3 ~9:40 AUS FG ............................ 13-13
6:51 3rd-and-7, AUS 38: mug + corner fire, sack -9, punt
~2:00+ SAC TD ............................ 20-13
Q4 early 4th-and-1, AUS 36: punt
SAC FG ............................ 23-13
~11:00 1st-and-10, AUS 41: Z Sluggo vs. press, +44
9:24 Jumbo Power Left, Ruffin from the 2 ... 23-20
(no more scoring)
📊 Diagram (described): A timeline of the second half. Austin kicks a field goal at about 9:40 of the third quarter to tie it 13–13. At 6:51 the Sentinels' mug pressure produces a sack and an Austin punt. Sacramento scores a touchdown with a little over two minutes left in the quarter, 20–13. Early in the fourth, Austin punts on fourth-and-1 at its own 36, and Sacramento kicks a field goal, 23–13. With a little under eleven minutes left, the slant-and-go against a pressed corner gains 44 yards, and Ruffin scores from the 2 at 9:24, 23–20. Nobody scores after that.
Marisol: "Ten and ten. If you only looked at the scoreboard, you'd say both halftimes worked about the same. I don't think that's true, and I don't think the scoreboard is the right way to grade it."
Wade: "How would you grade it?"
Marisol: "Four questions per change. Was the diagnosis right? Did the change get tested? What did it cost? Did it survive what they did?"
Teo had the drive log up on the wall. We went change by change.
Change 1: Trips, Sutton alone, TE #3 on a drag.
Teo: "First time it showed up was the tenth snap of the half. Third-and-5 at their 21, the stick to Brooks. Holloway was at nine, drove on it, broke it up. Field goal. Second time, early fourth, the flood on third-and-4 at our 33. Brooks in the flat for three, then the punt on fourth-and-1. Third time, the Sluggo. Forty-four."
Rome: "So one explosive, two misses."
Marisol: "The diagnosis was right. They were letting us have the short game, and that was going to change. It got tested, three times. It cost us a run look with Brooks off the line. And it survived their press, but only for the double move. The short throws out of it didn't get easier."
Lyle: "It was built for off corners. By the time we used it, they were pressing half the time."
Change 2: The chip.
Gus: "First half, Rollins rushed seventeen times. Five pressures and the sack. Second half, sixteen rushes, two pressures, no sacks. First chip was the second-and-6 at their 45. He never touched Darius."
Marisol: "Diagnosis right. Tested on the first drive. Result: it worked."
Gus: "Cost: Jalen's late into everything. And Ben's holding in the fourth came on a snap without the chip."
Teo: "There's another cost, and I didn't see it until last night. Every time we gave Ben help, our help went to the same side, toward 94. On the sack, the slide went left, toward him, same as it had all game. They knew where our protection was looking."
That one stopped the room for a second. Our best halftime change had told the other coordinator where our protection would be.
Change 3: Less Twelve.
Marisol: "For the game, eight Twelve runs for twenty-nine yards against big nickel, three of eight successful by our standard. Almost all of that was before halftime. Diagnosis: right, with a known cause, the third safety. Tested: we stopped calling it, so there's nothing to test. Cost: we gave up a group the defense had to respect. Survived their change: they didn't need one."
Wade: "So that one's a surrender. A reasonable one."
Lyle: "It was a sideline decision dressed up as a halftime change. I could have made it at the end of the first quarter."
Then we did theirs, from the film.
Teo: "Press on early downs: about half after the break, far less before. Rotation on early downs: about half after the break. In the first half they rotated now and then. And third-and-7-plus: they'd been quiet all half, and the first time they got one after halftime, they ran the mug."
Rome: "That last one wasn't something they saw in the first half."
Teo: "No. It was something they'd seen all season."
Wade: "Then here's my grade. Their halftime answered our first half and our season. Ours answered our first half. That's the difference, and it's not ten-and-ten."
THE AUDIT GRID (Monday after Week 15)
DIAGNOSIS TESTED? COST SURVIVED
RIGHT? THEIR MOVE?
OURS
1 Trips/X alone/ yes 3 times run look only for the
Y drag double move
2 Chip 94 yes series 1 back late; yes, but it
help has a side told them
our side
3 Less Twelve yes n/a a group they n/a (they
respected won it)
THEIRS (from film)
1 Press ~half yes Q3 Sluggo +44 until Q4
2 Rotate ~half yes Q3 -- yes
3 Mug on 3rd & 7+ (aimed at 6:51 -- yes: sack,
our card) punt
📊 Diagram (described): A grid grading six halftime changes on four questions: was the diagnosis right, was it tested, what did it cost, and did it survive the other side's move. Austin's trips change was right and tested three times, cost a run look, and survived only as a double move. The chip was right and tested on the first drive, cost the back's timing, and survived, but revealed which side Austin's help would go to. Less Twelve was right but untestable, cost a respected grouping, and needed no answer from the defense. The Sentinels' press change worked until the slant-and-go in the fourth quarter. Their rotation change worked. Their mug change was aimed at Austin's season-long protection rules and produced the sack.
At the end, I did what I always do after an audit. I wrote what changes on the sheet, not what changes in the game plan. From Week 16 on, my halftime sheet had a third section: "Their next move." It has one question at the top: What has worked for us long enough that they've built something for it? And Teo added a column to the drive log, starting the next week: where our hot throws and our protection help had gone during the game. Our own tendencies, charted live, so that at halftime we'd see what the other booth was seeing.
Analysis
1. Grade the decision, not the scoreboard. The second half was 10–10. A scoreboard grade says both halftimes were equal. The audit says they weren't. The Sentinels' third change took the one snap our plan treated as safe and turned it into a punt in a tie game. Our best play of the half came from a change that, by then, was mostly working as a double-move setup. A halftime audit that only asks "did we score more?" learns the wrong lessons from close games.
2. A change can be right and still leak information. The chip did exactly what it was supposed to do to Rollins. It also gave the Sentinels a reliable side for our help. Every adjustment is a new tendency the moment you use it twice. The audit needs a column for "what did this tell them?", and the drive log needs to track your own adjustments as carefully as it tracks the defense.
3. Two of three changes didn't need halftime. "Less Twelve" was a personnel decision for the sideline. The chip was a two-player tag that could have been made on the bench after the second-quarter sack. Halftime is the only time the whole offense is in one room. Spending it on changes that one or two people could make leaves no room for the question that needs everyone, or for the question about the other coordinator.
4. The missing question was about the other locker room. Every one of our changes answered something the first half had shown. None of them answered the question "what is Russ changing right now?" His biggest change wasn't a reaction to our first half at all. It was a plan for our season's habits that he'd chosen to use only once he was sure.
5. Audits change the process, not just the plan. The output of the meeting wasn't a new play. It was a new section on the halftime sheet and a new column in the drive log. Those changes outlast any one opponent.
Discussion questions
- Marisol graded each change on four questions: diagnosis, tested, cost, survived. Which of the four is hardest to answer honestly from film, and why?
- Teo noticed that the chip "told them our side." Can a defense really use that inside a single game? What would you need to see on the film to believe it?
- Wade said the Sentinels' halftime "answered our first half and our season." What's the difference between those two kinds of adjustment, and which is harder for an offensive staff to see coming?
- Lyle called "Less Twelve" "a sideline decision dressed up as a halftime change." Do you agree? Is there any argument for saying it at halftime anyway?
- The audit produced a new section on the halftime sheet, not a new play. Why might a process change be worth more than a scheme change after a close loss?
Your turn
Pick a game you can watch in full, at any level. Using only the film, identify up to three changes each team made at halftime. (Look at first-down corner depth, the shell on early downs, third-and-long pressure, personnel groups, and formations.) Build the audit grid from this case study for all of them: diagnosis right, tested, cost, survived the other side's move. Then write one sentence for each team answering this: Did its halftime answer only the first half, or did it answer the other team's habits?