66 min read

Third quarter, 6:51 on the game clock, Sacramento, Week 15. Tied 13–13. Third-and-7 at our own 38.

Chapter 36 — Halftime Adjustments: Reading the Game, Identifying Problems, and Fixing Your Offense in 12 Minutes

The Hook: Two Locker Rooms, One Snap

Third quarter, 6:51 on the game clock, Sacramento, Week 15. Tied 13–13. Third-and-7 at our own 38.

The call went into Darius Vance's helmet with 27 seconds on the play clock. Eleven personnel, two receivers on each side, Brooks Harlan flexed about four yards off our left tackle, Jalen Ruffin on Darius's left. "60 Left": slide the line to the boundary, toward DeShawn Rollins, and put the back on the man side. Then the look we'd prepared for all week: both inside linebackers walked up into the A gaps, one on each side of the center. A double-A mug. Our card had an answer for that, and it had won us a football game nine weeks earlier. Six-man protection, back on the Mike.

"Fifty-four's the Mike!" Darius pointed at Marcus Bell. Ruffin set his eyes on Bell.

At fifteen seconds my radio went dead, the way it does on every snap in the league. At about ten seconds, Grant Weller, the Sentinels' strong-side end, walked out wide on the boundary. Our rule for a late walk-up said the quarterback re-declares and moves the back to the new threat. Darius did exactly that. He sent Ruffin out to Weller. I stood there with a call sheet in my hand and a dead microphone, and I wanted to say "keep Jalen on the Mike." Nobody could hear me.

The ball was snapped with about three seconds left on the play clock. Weller dropped. Rashad Coleman, their 3-technique, dropped too, straight back into the short window in front of Mekhi Powers in the field slot, which is where every one of our A-gap hot throws had gone for a season. Bell came through the A gap untouched. Darius was on his hot in under a second, saw a 300-pound lineman sitting in the window, pulled the ball down, and went to the ground at our 29. Nine yards lost. Fourth-and-16. Punt.

The Sentinels drove for a touchdown later in the quarter, with a little over two minutes left in it. We lost the game 23–20.

Here's why I start a chapter about halftime with a snap from the middle of the third quarter. That sack was decided about forty minutes earlier, in two locker rooms, by two coordinators with one sheet of paper each.

In our locker room, I wrote three lines on my sheet and made three changes. Every one of them was about something I'd seen in the first half. None of them said "third-and-long."

In the other locker room, Russ Kimbrough made three changes too. I didn't learn what they were until the spring, at the league meetings. One of them was that the mug with the corner fire, which his staff calls Mug Cat Fire, was live on any third-and-7 or longer. He hadn't called it in the first half. He'd been saving it for the snap when he was sure what we'd do.

So here's the uncomfortable truth this chapter is built on: I fixed what I'd seen. Russ fixed what he expected me to keep doing. Both of us were doing halftime adjustments. Only one of us was adjusting to the other coordinator's halftime.

Halftime is roughly twelve or thirteen minutes in an NFL game. In that window you have to turn thirty- odd snaps of information into two or three changes that sixty tired, adrenalized men can carry back onto the field. You have to decide what not to change. And you have to guess, with some evidence and some nerve, what the coach across the building is writing on his sheet at the same moment. This chapter is how we do that in Austin. It's also an honest account of a halftime where the other coordinator did it better.

🏃 Fast Track: If you've coached on game day, you know the clock and the chaos. Skim 36.1–36.2 and slow down at 36.4 (triage: why three, and how to pick them), 36.7 (adjusting to their adjustments), and the Project Checkpoint, where you'll build an In-Game Adjustment Protocol you can carry into a locker room.

🔬 Deep Dive: The Week 15 sack is broken down from the line's view in Chapter 5 and as a rule sheet in Chapter 22. The quick-game half that set up the halftime is in Chapter 21. Appendix F has a blank halftime template: Appendix F.

📺 Film Watcher: This Sunday, write down three things in the first quarter of any game: how deep the cornerbacks are on first down, how often the defense shows two deep safeties on early downs, and what it does on third-and-long. Check the same three things in the third quarter. If any of them changed, you just watched a halftime adjustment. Most broadcasts never mention them.


36.1 Twelve Minutes, Roughly

Halftime in a normal NFL game is roughly thirteen minutes on the stadium clock. (Special events, the Super Bowl above all, run longer.) It sounds like a lot of time. It isn't, because most of it doesn't belong to you.

Here is what our halftime looks like, minute by minute. Every staff does this differently, and the times stretch and shrink with the walk to the locker room, the medical staff, and how the half ended. What matters is the shape.

AUSTIN'S HALFTIME CLOCK (roughly 13:00; times approximate)

 0:00 ----- 1:30 ----------- 5:30 ------------ 9:30 ---- 11:00 -- 13:00
   |  WALK    |  COACHES ONLY  |  POSITION ROOMS |  WADE  | WARM-UP |
   |  OFF     |  the 4-minute  |  Gus: OL board  |  whole | throws, |
   |  Teo     |  decision:     |  Rome: QBs,     |  team  | kicks,  |
   |  dials   |  ONE SHEET     |    tablet stills|        | out for |
   |  in      |  Teo on phone  |  Cal: WRs       |        | the 3rd |
   |          |                |  Lyle: 90 s     |        | quarter |
   |          |  players: tape,|    with the     |        |         |
   |          |  fluids, docs  |    whole offense|        |         |
   '----------'----------------'-----------------'--------'---------'
            |<------- the offensive block, about 8 minutes ------->|

📊 Diagram (described): A timeline of about thirteen minutes. The first minute and a half is the walk to the locker room, during which Teo calls down from the booth. From about 1:30 to 5:30 the coaches meet alone and Lyle builds his one-page sheet while players get taped, drink, and see the medical staff. From about 5:30 to 9:30 the position coaches meet their rooms, and Lyle spends about ninety seconds with the whole offense. Head coach Wade Harmon addresses the team from about 9:30 to 11:00, and the last two minutes are warm-up on the field. The offense's whole block, from the coaches' room through the position rooms, is about eight minutes.

Look at where the coordinator's real decision lives: in about four minutes, from roughly 1:30 to 5:30. The players aren't even in the room yet. They're getting retaped, drinking, getting an ankle looked at. That's not wasted time for them, and it's the only quiet time you get.

Everything after 5:30 is transmission, not decision. The position coaches teach the changes. I walk the hallway and say the same three things to the whole offense in about ninety seconds. Wade talks to the team. Then we're back on the field. If I'm still deciding at 6:00, the changes reach the players half-formed, or they don't reach them at all.

That's why my halftime is one sheet of paper. At the top I write what the first half told us, in plain words. Under that I write what we're changing. Never more than three changes. A team can't absorb four. Since Week 15, the sheet has had a third section too, which this chapter will earn: what the other coordinator is changing right now.

🩺 Player safety: The medical staff owns the first minutes of halftime, not the coordinator. A player being evaluated may miss the whole meeting. Never pull a player out of an evaluation to hear an adjustment. Instead, the position coach catches him on the sideline before the first series and gives him the change in one sentence. In heat, the fluids come first, and a meeting that eats the hydration time is a meeting that will cost you in the fourth quarter.

Why halftime matters more than people think, and less

Fans talk about "halftime adjustments" as if the coaches walk into the locker room with a broken offense and walk out with a new one. That almost never happens, and when it does, it usually goes badly. You'll see a coordinator try it in Case Study 2.

What halftime really does is narrower. It's the only time in the game when the whole offense is in one building, sitting down, with nothing else to do but listen. Between series you can talk to one position group for a minute or the quarterback for thirty seconds. At halftime you can talk to everyone once. That makes it the one moment when you can change something that needs more than one man to change. A new protection rule needs five linemen and a back. A new formation needs eleven people. A new coverage read needs the quarterback and four receivers to see the same thing.

So my first question at halftime isn't "what's broken?" It's this: "What do we need to change that can only be changed with everyone in the room?" Anything smaller can be fixed between series, and should have been.

💡 Coach's point: The first half's best plan is the second half's scouting report — for both sidelines. Whatever worked for you before halftime is now on the other coordinator's sheet. Plan for that before you plan anything else.

🔄 Check your understanding: In Austin's halftime, roughly how long does the coordinator have to decide, and why is the rest of halftime not decision time?

Answer

About four minutes, from roughly 1:30 to 5:30, while the players are getting treatment and fluids. After that, halftime is transmission: position coaches teach the changes, the coordinator gives the whole offense about ninety seconds, the head coach talks, and the team warms up. A decision made after about 6:00 reaches the players half-formed. That's why the coordinator works from one sheet with at most three changes.


36.2 The Charting That Makes Halftime Possible

You can't decide anything in four minutes unless the first half has already been sorted for you. A halftime adjustment is only as good as the charting behind it. By the time I reach the locker room, the first half must already be organized into answers. It can't still be a pile of memories.

Teo's drive log

Teo Arriaga, our quality control coach, sits in the coaches' booth upstairs on game day. His job during the game has two halves. The first is the one you know from earlier chapters: in the quiet window between the whistle and my call, he's the only voice in my headset, with three items in a fixed order: shell, box, one thing. "Two high, six, Will at five." "Single high, eight, corner pressed on Sutton." (The shell is the safety picture before the snap, two deep or one deep. The box is the number of defenders close to the line inside the tight end or tackle. The one thing is whatever Teo thinks I need most for the next call.)

The second half of his job is quieter and matters more for halftime. On every one of our snaps he fills in a row of what we call the drive log. It's his own sheet, and it's built to be added up.

Column What Teo writes Example (Week 15, first quarter)
# / clock Snap number, quarter, game clock Q1 / 6:48
Down & spot Down, distance, yard line 3-11, AUS 44
Our call Short form, personnel and concept Eleven, Dagger
Their personnel Base, nickel, dime, big nickel Dime
Shell pre → post 2H, 1H, or 0, with an arrow for rotation 2H → 2H
Box Light / neutral / loaded Light
Rush Number of rushers, plus a note on who 4, end over-rushed
Coverage His best guess; "?" if unclear Two-deep man
Result Yards, and a success mark by our standard +7 scramble, no
One thing What he'd tell me if he could only say one sentence "Short receiver sat; Gaines ran away from the scramble"

Teo has a "?" rule: if he can't tell what the coverage was from the booth on the live play, he writes a question mark and fixes it on the photos or on Monday's film. A guess written as a fact is worse than a blank, because by halftime everybody's forgotten it was a guess.

At the end of every series, he draws a line under the rows and writes a one-line series note: what the defense showed us that series that we hadn't seen yet. That's the line I read first at halftime.

Marisol's numbers and Gus's notes

Marisol Vega, our director of analytics, keeps a few running numbers during the game. They're not exotic: success rate by personnel grouping, third-down conversions, explosive plays, and how often each of our concepts has been called. She hands Wade a fourth-down chart before every series. At halftime her numbers come down with Teo's.

Gus Ferrante, our offensive line coach, keeps his own notes on the sideline. Every pass-protection snap gets a mark: who won, who lost, and how. When the linemen come to the bench, he has the sideline tablet with still photos of the pre-snap alignment and the moment of the rush. He shows them the picture, says one sentence, and lets them sit down.

Rules on what you can look at

The technology rules move, so here is the principle rather than a rulebook. For most of the tablet era, NFL sidelines have used league-supplied tablets that showed still photographs of each play's pre-snap and post-snap pictures, not video. The league has been loosening and adjusting those limits in recent seasons, so check the current rule before you build a process around one version of it. College rules have also changed recently, and high school rules vary by state association; some allow very little. Coach-to-coach communication between the booth and the sideline is allowed. The coach-to-quarterback radio cuts off at fifteen seconds on the play clock.

What matters for you is that every staff, at every level, works with less information during the game than it has on Monday. The drive log is how you make the little you have add up to something.

HOW THE FIRST HALF REACHES THE HALFTIME SHEET

  BOOTH (Teo)                      SIDELINE                LOCKER ROOM
  ------------                     --------                -----------
  shell, box, one thing --------> Lyle's headset
  drive log, row by row            (quiet window)
  series note (after each drive) -> Lyle reads on   ----->  LINE 1-3:
                                     the bench               "what the
  Marisol: success by personnel,                             half told
    3rd down, explosives  ------------------------------->   us"
                                   Gus: pro marks,
                                     tablet stills  ------>  CHANGES
                                   Rome: QB's eyes,          (2-3 max)
                                     what Vance saw ------>
                                                             THEIR NEXT
  Teo, by phone at 1:30: the half in numbers  ----------->   MOVE

📊 Diagram (described): Three columns show where information comes from. In the booth, Teo gives the "shell, box, one thing" call before each play, fills the drive log on every snap, and writes a series note after each drive; Marisol tracks success rate by personnel, third downs, and explosive plays. On the sideline, Lyle hears Teo in the quiet window and reads the series notes, Gus keeps protection marks and shows tablet stills, and Rome reports what the quarterback saw. All of it flows to the locker room, where it becomes the sheet's three sections: what the half told us, the two or three changes, and the other coordinator's likely next move. Teo's phone call at about 1:30 delivers the half in numbers.

What the log said at halftime of Week 15

Here's what Teo read me over the phone when I reached the locker room in Sacramento. Chapter 21 told the start of the call: thirty-three snaps, the quick game eleven of thirteen for seventy-four yards with a long of eleven, third down three of nine, one explosive play (the back-shoulder to Sutton). We went over the quick game twice. Then it went like this.

Lyle: "Twelve."

Teo: "Nine snaps. Eleven personnel twenty-two, other two. Big nickel on every Twelve snap. The runs aren't going anywhere."

Lyle: "Ninety-four."

Teo: "Seventeen rushes, five pressures, the sack. He's winning on Ben's outside shoulder and coming back inside when Ben oversets."

Lyle: "Anything they haven't shown us?"

Teo: "Nothing on third-and-long we haven't seen. They've been quiet there."

Read that last line again. They've been quiet there. It was true. It was the most important true thing anybody said at halftime, and I heard it as good news.

🔄 Check your understanding: Why does Teo's drive log have a "?" rule, and why does he write a series note after every drive instead of waiting for halftime?

Answer

The "?" rule keeps guesses from turning into facts. Booth views of live plays are imperfect, and by halftime nobody remembers which coverage calls were guesses, so a written guess can mislead the whole staff. The series note forces the sorting to happen while the drive is fresh: one line on what the defense showed that was new. At halftime Lyle has four minutes. He can't re-watch thirty snaps, but he can read five or six series notes.


36.3 Identifying the Problem: Symptom, Cause, and Who Knows It

Most halftime mistakes are diagnosis mistakes. The coordinator sees something real, names it wrong, and fixes the wrong thing well.

The first discipline is separating the symptom (what you saw on the scoreboard or the stat sheet) from the cause (the defensive structure, the player, or the tendency that produced it). "We can't run the ball" is a symptom. "They're playing a third safety against our two-tight-end personnel, and he's fitting the alley we're running at" is a cause. You can't fix a symptom. You can only fix a cause.

The second discipline is asking what kind of problem it is, because each kind has a different level of fix. These are the five kinds we sort into.

Kind of problem What it looks like on the log The level of fix that usually answers it Week 15 example
A defender One name keeps showing up in the "one thing" column A tag or a help rule (chip, slide, double), or a call that uses his aggressiveness "94 is beating our left tackle": Rollins, 5 pressures on 17 rushes
A structure A coverage, front, or personnel answer you didn't expect, or expected less of A formation or personnel change that changes what the structure has to cover "Twelve isn't working": big nickel on all nine Twelve snaps
A price Something works, but the defense is letting it work A double move or a shot built off the thing they're conceding, ready for when they stop conceding "They're giving us the quick game, and they'll take it away"
A tendency they caught Their defenders arrive early, guess right, sit on routes Break the tendency: same look, different play; or a new look for the same play Not on our sheet. It should have been: our third-and-long protection rules
Execution The right call, the right look, a missed block or a late throw Coaching, not scheme; one sentence to one player Ben's oversetting, which Gus coached on the bench

The fourth row is the hardest, because the evidence for it isn't on your drive log. It's on their drive log. When a defense has caught one of your tendencies, it usually doesn't show it until it's sure. A good defensive coordinator will let you complete a pass three times to get the fourth one. Teo's "they've been quiet there" was exactly what a caught tendency sounds like at halftime. Nothing had happened yet.

🧩 Productive struggle: Before you read the sheet I actually wrote, write your own. You're in the visitors' locker room in Sacramento. It's 13–10, Sentinels. Here's what you know:

  • 33 snaps; quick game 11 of 13 for 74 yards, nothing longer than 11; third downs 3 of 9.
  • Twelve personnel: 9 snaps, big nickel every time, the runs going nowhere.
  • Rollins (#94): 17 rushes, 5 pressures, 1 sack, mostly against your left tackle.
  • One explosive play: a back-shoulder throw to your 6-foot-3 split end against a pressed corner.
  • On third-and-long, the Sentinels have played quiet, two-high coverage with four rushers.
  • Your third-and-long protection rules have been the same all season. They beat a six-man blitz for the winning touchdown nine weeks ago.

Write three lines describing what the half told you, then up to three changes. Then write a fourth line, the one about what they will change. Don't read on until you have all four.

Answer

My three lines were: "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 were: Trips with Sutton alone into the boundary and the tight end as #3 on a drag; chip Rollins with Ruffin; less 12 personnel. You'll see each of them examined in 36.4.

The fourth line is the one I didn't write. A strong version: "They've seen our third-and-long rules for a season, and they haven't tested them today. Assume they will." You didn't need new information to write it. You needed to ask the question. Evidence that the Sentinels were going to attack our protection didn't exist at halftime. The incentive for them to do it did: our rules had been on film for months, and they'd beaten Sacramento's best pressure once. A defense that loses to a rule studies the rule. If your fourth line named any of our long-standing tendencies, you're thinking like the other locker room.

Who knows the problem?

One more diagnostic question, and it's the one that separates good halftimes from busy ones: does the defense know about this problem yet?

If Rollins is beating our left tackle, Russ knows. He'll keep putting Rollins there. My fix has to work against a defense that expects a fix: a chip they'll see coming, a slide they'll read.

If I've found something the defense doesn't know it's giving away, such as a safety who cheats on a certain motion, I should be very careful how I use it. A halftime that announces my best discovery on the first snap of the third quarter gives the other coordinator a whole half to fix it.

And if the problem is one of my tendencies, the question flips. Does the defense know about it? On our third-and-long protection, the answer was yes, and had been for weeks. I was the only one in the building who didn't think about it at halftime.

🔄 Check your understanding: Classify each of these halftime notes as a defender, a structure, a price, a caught tendency, or execution: (a) "Their nickel is jumping every slant." (b) "We've run outside zone four times and their backside end has made the tackle three times." (c) "They're playing off coverage and giving us hitches." (d) "Our right guard has two false starts."

Answer

(a) Most likely a caught tendency: a defender jumping one route over and over usually means he knows when it's coming. Check whether the slants all came from the same formation or down. (b) A defender problem inside a structure: the backside end is chasing flat down the line, which may be a coached answer. That's a candidate for a boot or a reverse-action play off the same look. (c) A price: they're conceding the hitch, so prepare the double move for when they stop. (d) Execution: one player, one sentence, maybe a silent count. Not a scheme change.


36.4 Triage: Two or Three, Never Four

By halftime of almost any game you'll have more problems than you can fix. Week 15 gave me at least six worth writing down. Triage is the discipline of choosing the two or three that matter most, can be fixed with the time and the words you have, and won't break something else.

My rule is old and simple. Never more than three changes. A team can't absorb four. Here's why that number isn't arbitrary.

🔍 Why this works: A halftime change isn't a fact for players to remember. It's a behavior they have to carry out at full speed, tired, with a crowd in their ears, and without a single practice rep. Every change takes some of a player's attention on every snap. One change, and he's thinking about it a little. Three changes, and he may be thinking about all three on a snap where only one applies. Past that, you start getting the classic halftime errors: false starts, wrong splits, a receiver running last week's route. Most coaches who've tried a big halftime overhaul have seen the third quarter fall apart. There's also a second reason, and it's about learning. The third quarter is an experiment. Change three things, and when the first drive works or fails you can usually tell which change did it. Change seven, and you've learned nothing, because you can't separate the causes. A coordinator who changes everything at halftime gives up the one thing the third quarter can teach him.

The triage filter

Every candidate change goes through five questions. They take about ten seconds each once you're used to them.

# Question Why it matters If the answer is bad
1 Is it a pattern or a play? One snap is an event. Two from the same cause is a pattern. Leave it, and ask the booth to watch it
2 Who has to change? One player can be fixed on the sideline. A position group can be fixed between series. Only a change that needs everyone truly needs halftime. Move it to the sideline or the position room
3 Can we say it in words we already have? No new words at halftime. The players have no reps on a word they learned twenty minutes ago. If "no new protections after Friday" holds for a week, it holds twice as hard in a locker room. Rebuild the fix from existing calls, tags, and rules, or drop it
4 What does it cost? Every fix is a trade. A chip slows the back's route. Fewer tight ends means fewer run fits. Write the cost next to the change, so the booth can watch for it
5 What will they do to it? The other coordinator will see your change on the first snap it shows up. Does it survive his next move? Pre-build the answer to his answer, or choose a different change

Here's what the filter says about the three changes I actually made.

Change Pattern? Who changes? Existing words? Cost What would Russ do to it?
1. Trips, Sutton alone into the boundary, TE as #3 on a drag Yes: 13 quick throws, all against off corners All eleven (formation) Yes: Trips, Y Drag Harlan off the line: fewer run fits in that look Press and rotate. Then Sutton is alone against a pressed corner, and the drag runs under linebackers who are dropping
2. Chip Rollins with Ruffin Yes: 5 pressures on 17 rushes Ruffin and Achterberg Yes: chip is an existing rule Ruffin late into his route; a predictable side for our help Read the chip side and attack the other side, or attack the back's assignment
3. Less Twelve Yes: big nickel on all 9 snaps Sideline personnel calls only Nothing new needed Concedes the point, and makes the Twelve we still call easier to read Nothing. He wins the argument by default

Look at the last column. I filled it in for this book in the spring. At halftime in Sacramento, I didn't fill it in at all. That's the column the sheet was missing.

⚠️ Common mistake: the halftime install. A coordinator sees a new pressure in the first half and draws a brand-new protection on the whiteboard at halftime: new word, new back assignment, new hot. On the first third-and-long of the third quarter, the right guard blocks the old rule, the back blocks the new one, and the quarterback gets hit. The fix: answer a new problem at halftime with a combination of things your players already know, such as an existing protection with a different call, a hot throw you've repped, or a max-protect call from the menu. The real new protection waits for Wednesday's practice.

⚠️ Common mistake: fixing the last series. Halftime comes right after the two-minute drive, and whatever happened in the last three minutes is loudest in everybody's head. In Week 15 our last possession of the half was a punt at 0:58 after a dig and a checkdown. If I'd built my sheet around that series, I'd have spent a change on the end-of-half situation, which won't come up again until the fourth quarter. The fix: read Teo's series notes in order, starting with the first series, and write your lines before you read the last one.

📋 Try it — Make the call (triage): Hypothetical. Next season, at Birmingham. The Forge play blitz-heavy man coverage. You're down 10–3 at halftime. Your list of problems:

  • (a) On third down they've played Cover 0 (no deep safety, everyone in man, six or more rushers) on 3 of 5 snaps. Your hot throws went 2 of 3.
  • (b) Their nickel has jumped two of your slants to the field slot.
  • (c) Your right guard has two false starts.
  • (d) Ruffin has 6 carries for 11 yards, all against eight-man boxes.
  • (e) Sutton has beaten press twice on the outside, and Vance didn't throw it either time; he was holding the ball for Gaines.
  • (f) The Forge haven't blitzed on first down all half.

Pick at most three halftime changes. For each, say who changes and in what existing words. For the problems you don't take to halftime, say where they get fixed instead.

Answer

One strong answer:

  1. (d) Stop handing it into eight. This needs Vance, the line, and the back to share one rule: a Kill call from the same picture, "seven or fewer, run it; eight, take the throw." It uses existing words, and it turns their eight-man box into a reason to throw against man outside.
  2. (e) Make Sutton the first read against press on early downs. This needs the quarterback and a receiver: a back-shoulder or fade tag on Sutton's side, read first when his corner presses with no help over the top. It's an existing tag. Rome can teach it with the tablet stills of the two snaps Sutton won.
  3. (b) Break the slant tendency. The nickel is jumping slants because he's seen enough of them. Use an existing double move or a slant-and-go from the same look once, early, and keep the slant on the other side of the formation. This needs one receiver, the quarterback, and a protection that holds for the double move.

Where the others go: (a) is already working (2 of 3) and lives on the zero-bin wristband; it doesn't need halftime. Have the booth confirm the Cover 0 tells and leave it alone. (c) is one player; the line coach handles it on the bench with a silent count, if the noise warrants it. (f) is information, not a problem. Note it and watch whether it changes in the third quarter, because a blitz-heavy coordinator who hasn't blitzed on first down may be saving it.

If you picked (a) as a change, ask yourself whether you'd be fixing something that isn't broken. If you picked (c), you've spent a halftime change on one player.


36.5 Communicating the Adjustment

A change that's decided but not transmitted is just a note. Halftime has about four minutes of transmission after the decision, and every minute is shared with something else. So we've made the transmission as boring and repeatable as we can.

The order is fixed. I write the sheet. Each position coach gets about thirty seconds from me on what his room has to change. The position coaches teach it in their rooms. I walk the hallway and say the changes to the whole offense once, in about ninety seconds. Darius hears every change three times: from me, from Rome in the quarterback room, and on the copy of the sheet that goes in his game-plan binder.

Every change goes out as one sentence a player can say back. It's the same rule we use for a rookie's terminology in July. If the rookie can't say it back, it's not a word in our offense. At halftime, if the player can't say the change back, it isn't a change yet. Each position coach ends his room by having one player say the change out loud. That takes ten seconds and catches most misunderstandings before they reach the field.

The whiteboard shows the change, not the play. When Gus draws the chip for the line, he doesn't draw the whole protection. He draws Rollins, Achterberg, and Ruffin, and one arrow. When Cal Boateng, our receivers coach, talks to Devonte Sutton, he doesn't redraw the trips formation. He says: "You're alone in the boundary. They'll press you. Win the release and look for the ball over your shoulder."

The tablet shows what the new look will look like. Rome uses the halftime minutes with Darius on the still photos, not on our plays but on the defense's pictures: what a pressed corner looked like on the back-shoulder snap, what the safeties looked like before the one rotation they'd shown us. You can't practice at halftime, but you can show a quarterback the picture he's about to see.

The call sheet changes in red, and the wristband doesn't change at all. Teo prints the wristbands on Friday. Nobody reprints a wristband at halftime. If a change needs a new combination, it travels as a full call over the radio, or it travels as the band number of a call that's already on the band. On my call sheet, I cross out in red what we're not calling anymore and circle what we're leaning on. In Week 15 I crossed out most of the Twelve runs and circled every call on the sheet that came from Trips.

WEEK 15 HALFTIME SHEET (as written; Sentinels 13, Outlaws 10)

 +--------------------------------------------------------------+
 | WHAT THE HALF TOLD US                                        |
 |   1. They're giving us the quick game, and they'll take it   |
 |      away.                                                   |
 |   2. 94 is beating our left tackle.                          |
 |   3. Twelve isn't working.                                   |
 |--------------------------------------------------------------|
 | CHANGES                                                      |
 |   1. Trips: Sutton alone into the boundary, TE #3 on a drag  |
 |   2. Chip Rollins with Ruffin                                |
 |   3. Less 12 personnel                                       |
 |--------------------------------------------------------------|
 |   (no third section: what will Russ change?)                 |
 +--------------------------------------------------------------+

📊 Diagram (described): Lyle's one-page halftime sheet from Week 15, drawn as a box with two filled sections. The top section, "What the half told us," has the three lines: they're giving us the quick game and will take it away; Rollins (number 94) is beating the left tackle; Twelve personnel isn't working. The middle section, "Changes," has the three changes: trips with Sutton alone into the boundary and the tight end as the number-three receiver on a drag; Ruffin chips Rollins; less Twelve personnel. The bottom of the box is empty. There was no section for what the other coordinator would change. That section was added after this game.

Here's the first change as Cal and Rome drew it, against the shell the Sentinels had played all first half: two safeties deep, corners off at seven and eight yards.

PRE-SNAP: Halftime change 1 vs. the first-half shell
(Eleven, Gun Trips Right, 60 Right; boundary left, field right)

          F                                         $

  C                     W       M          *                 C
  (8 off)          E   N     T    E                      (7 off)
  X               LT  LG  C  RG  RT       Y      H           Z
                       R  Q
  Sutton,                               Harlan  Powers    Gaines
  alone                                  #3      #2         #1
  BOUNDARY                                                FIELD

📊 Diagram (described): The offense is in trips to the right, toward the wide side of the field: Gaines wide as number one, Powers inside him as number two, and Harlan detached from the line as number three. Sutton is alone on the left, toward the sideline that's closer to the ball. Ruffin is in the backfield on Vance's left, on the side away from the slide, where Rollins rushes. The defense shows its first-half look: two deep safeties, the boundary corner eight yards off Sutton, the field corner seven yards off Gaines, the nickel over the trips side, and two linebackers inside.

The point of the formation was to make the Sentinels declare. Three receivers to one side asks the coverage to send help there. If the safeties roll toward trips, Sutton is alone on the other side with a corner and no one over the top. If they don't roll, the trips side has numbers. Harlan as #3 on a drag is the quick game's answer to press: a crossing route under the linebackers, which a pressed corner can't touch. And Ruffin, set on Rollins's side, gets his chip on the way out.

Now here's what we actually saw in the second half, after Russ's changes, which you'll meet properly in 36.7.

POST-SNAP: Change 1 vs. the second-half look
(corners pressed; two-high rotates to one-high at the snap)

                            F  (spins to the deep middle)
   ^
   |  fade / back-                                   $ (rolls down
   |  shoulder                                          to the flat)
   |                     W        M          *
   C X (pressed,                                        C Z (pressed)
        alone)      E   N     T    E
        (E)<--R    LT  LG  C  RG  RT
       chip, then           Q
       out to the flat     <---------------------Y   drag at 5,
                                                     under W and M

📊 Diagram (described): At the snap the two deep safeties rotate: the free safety spins to the deep middle and the strong safety rolls down toward the trips side's flat. Both corners are pressed, one on Sutton alone on the left and one on Gaines on the right. Ruffin chips the left defensive end, Rollins, and then releases to the left flat. Harlan, the number-three receiver, runs a drag about five yards deep from right to left, under the two linebackers. Sutton's best route against a pressed corner with the safety in the middle is a fade or a back-shoulder throw. The routes of Powers and Gaines are not shown.

This is the picture where halftime meets halftime. Our change was built to win against an off corner who might roll coverage. Russ's change put a pressed corner on Sutton and moved a safety to the middle, where he could help on the post and still let the corner sit on the fade. Our change wasn't wrong. It was now facing a defense that had changed too. The drag was still open underneath. The isolation on Sutton was still a one-on-one. But the easy completions of the first half were gone, and the throws the formation now offered were harder ones.

📋 Try it — Draw it up: Draw halftime change 1 (Trips Right, Sutton alone left, Harlan #3 on a drag, Ruffin chipping to the left) against a different second-half answer: the Sentinels stay two high but press both corners. Show the two deep safeties' landmarks, the pressed corners, and the linebackers' drops. Then say where the ball should go first, and why.

Answer

```text POST-SNAP: Change 1 vs. press with two high

        F                                   $
        (deep half)                    (deep half)

^ | fade: safety over the top, | so this is a back-shoulder | or nothing W M * C X (pressed) C Z (pressed) E N T E (E)<--R LT LG C RG RT Q <--------------------Y drag at 5 ```

📊 Diagram (described): The two safeties stay deep, each over his half of the field. Both corners press. Ruffin chips Rollins and releases left; Harlan runs the drag right to left under the linebackers; Sutton is pressed on the left with a safety over the top of him.

With two deep safeties, Sutton's fade has a safety over it, so the isolation is weaker than it looks; it's a back-shoulder throw or nothing. The best first read is usually the drag: with both corners pressed and both safeties deep, the defenders underneath are the two linebackers and the nickel, and the drag runs across their faces. If a linebacker runs with Harlan, the area he left is where Ruffin is releasing after the chip. Two-high with press is a structure that gives you the short middle, which is exactly what the quick game wanted. It's also why Russ rotated on about half of his early downs instead of all of them: so Darius couldn't be sure, before the snap, which picture he was going to get.

🔄 Check your understanding: Name the three things that do not happen when Austin transmits a halftime change, and why each is ruled out.

Answer

(1) No new words. Players have no reps on a word learned at halftime. (2) No wristband reprints. The bands are printed Friday; a halftime change travels as a full radio call or as the number of a call already on the band. (3) No whole-play redraws on the whiteboard. The coach draws only the piece that changes, so the player sees what's different instead of re-learning what isn't.


36.6 The Second-Half Script

You've heard me say it about the first fifteen plays, a saying I got from Walt Dupree, my first NFL boss: "Script the first fifteen to learn, not to win." The second half gets a script too. It's shorter, three to five calls, and it has a sharper job. The first-half script asks, "What is their plan?" The second-half script asks two questions: "Did our changes work?" and "What did they change?"

Our rules for it:

  1. Open with a question, not your best answer. The first call of the half should be something you ran in the first half, from the same look, so that if the defense has changed, you see it against a picture you already understand.
  2. Test every change in the first series if the down and distance allow it. A change that hasn't shown up by the middle of the third quarter is a rumor. You don't know whether it works.
  3. Script the answer to their likely adjustment, not just your fix. If you think they'll press, have the press-beater on the script, ready to call the moment the booth confirms it.
  4. The booth's "one thing" for the first series of the half is always "what's different?" Teo doesn't wait to be asked. If the corners are a yard closer, he says it on the first snap.

Here's the opening drive of the third quarter in Week 15 as Teo logged it. The Sentinels kicked off to us. Watch how the drive answers both questions, in Austin's words, the way Teo wrote it on the day.

Down & spot Our call What the defense showed Result What it told us
1-10 AUS 21 Back-shoulder to Sutton Brennan in man on Sutton +19 Sutton can still win his one-on-one
1-10 AUS 40 Inside zone Two-high, light box +6 Their run fits hadn't changed
2-4 AUS 46 Ruffin run Two-high, Gaines ran off his corner +3 —
3-1 AUS 49 "Heavy, Under Wing Right, Check" → sneak Bear front +2, first down The Check menu worked
1-10 SAC 49 Powers quick out 2H → 1H, corner bailed +4 They're rotating on first down. Change confirmed
2-6 SAC 45 Dropback, Ruffin chips Rollins — Incomplete Change 2 in. Rollins never touched Darius
3-6 SAC 45 Harlan up the seam over Bell Rotation, bracket on Gaines +19 to the SAC 26 The seam is there when they bracket Gaines
1-10 SAC 26 Run 2H → 1H again +4 Second rotation on first down
2-6 SAC 22 Checkdown to Ruffin Two-high, Six +1 —
3-5 SAC 21 "Eleven, Gun Trips Right, 60 Left, Stick" Two-high, Holloway shallow, robbed the stick Incomplete Change 1 in. Holloway's depth was the tell

Field goal. 13–13, at about 9:40 of the third quarter.

On the 1-10 at the Sacramento 49, Teo's voice came in before I'd made the next call: "Two to one. They rolled on first down. That's new." It was new. In the first half, the Sentinels had rotated out of their two-high shell only now and then on early downs. After halftime they rotated on about half of them. Russ's second change had shown up on the fifth snap of the half, and the booth had it.

The third-and-5 at the 21 is told in full in Chapter 15, because it's a lesson about reading the robber. For this chapter, notice the timing. Our halftime change 1 was in the game on the tenth snap of the half, and it put Harlan at #3 on the stick exactly as designed. Holloway, aligned at nine yards instead of his usual eleven, drove on it and knocked it away. Teo had said "Holloway's shallow" in the quiet window. I heard it and didn't act on it.

So by 9:40 of the third quarter, after one drive, I knew three things. Our chip worked. Our trips change was in the game but hadn't produced anything yet. And the Sentinels had changed their early-down shell. What I didn't know was that they'd made a third change, and that they hadn't needed to show it yet. Third-and-long hadn't come up.

A script that tested the prediction: Week 10

Compare that with our third quarter in Portland, five weeks earlier. At halftime there we led 14–10. Our second-quarter touchdown had come on smash against Cover 2: the Pioneers' corner, Josh Kiel, had squatted on the hitch and Powers caught the corner route over him for 31 yards. The first line on my halftime sheet that day was a prediction: "Marchetti fixes Kiel." Leon Marchetti, the Pioneers' coordinator, was too good to let that happen twice. We expected him to teach Kiel to sink under the corner route whenever our slot released vertically.

So the second-half script had smash on it again, early, with the read already turned around for Darius: if Kiel sinks, the hitch is free. On first-and-10 at our 46, at 12:20 of the third quarter, Kiel sank. Darius saw his hips turn on the first step and threw the hitch on rhythm. Sutton, eight yards. Teo, from the booth: "He sank. Hips at the snap. That's the halftime fix."

That's the whole difference between the two halftimes. In Portland I wrote down what the other coordinator would change, and the script was waiting for it. In Sacramento I wrote down what I'd seen, and the script tested my changes but had nothing waiting for Russ's. Chapter 16 tells the Portland game from the concept's side (Chapter 16).

🔄 Check your understanding: What are the two questions a second-half script asks, and on which snap of the Week 15 third quarter did the booth answer the second one?

Answer

"Did our changes work?" and "What did they change?" The booth answered the second question on the fifth snap of the half, first-and-10 at the Sacramento 49, when the Sentinels rotated from two-high to one-high on a first down. Teo called it ("Two to one. They rolled on first down. That's new.") before the next call. That was Russ's second halftime change, rotation on about half of early downs.


36.7 Adjusting to Their Adjustments

Every halftime has two sheets of paper. You only get to see yours.

I learned what was on Russ's in the spring, at the league meetings, over a couple of coffees. In the second half of Week 15, he'd made three changes:

  1. His corners pressed on about half of our early downs, instead of playing seven and eight yards off.
  2. His safeties rotated out of the two-high shell on about half of our early downs, instead of staying two-high and letting us see the same picture every time.
  3. On third-and-7 or longer, the double-A mug with the corner fire was live. He hadn't called it in the first half.

Russ put the first two in a sentence I've repeated to every young coach since: "So at the half, I changed the price." He'd let us complete eleven of thirteen quick throws because six-yard completions weren't going to beat him on his own field. At halftime he made the quick game cost more: more pressed corners, so Darius couldn't be sure of the cushion; more rotation, so the help on the slant window could come from either side. Neither change stopped the quick game. Both made it slower to read, and the quick game can't afford slow.

Then he said the line that's in the title of this section, in his own words. "At halftime I don't fix what you did. I fix what you're going to keep doing."

Put the two halftimes side by side and you can see the whole second half before it happens.

TWO HALFTIMES, WEEK 15 (what met what in the second half)

  OUR CHANGES                        THEIR CHANGES
  -----------                        -------------
  1. Trips, Sutton alone,  <------>  1. Press ~half of early downs
     TE #3 on a drag                    (the isolated Sutton now sees
                                        press; the drag still beats it)

  2. Chip Rollins          <------>  2. Rotate ~half of early downs
     (back late into routes;            (the quick game gets slower
      our help has a side)              to read; quick out +4)

  3. Less Twelve           <------>  (nothing: he'd already won it)

  (nothing)                <------>  3. Mug + corner fire on 3rd & 7+
                                        aimed at the rules we didn't
                                        touch: 6:51 Q3, sack, -9

📊 Diagram (described): Two columns compare the halftime changes. On the left are Austin's: trips with Sutton alone and the tight end on a drag; the chip on Rollins; less Twelve personnel. On the right are Sacramento's: press on about half of early downs; rotation on about half of early downs; and the double-A mug with a corner blitz on third-and-7 or longer. Arrows pair them. Austin's trips change met the press. The chip met the rotation, which made the quick game slower. Less Twelve met nothing, because the Sentinels had already won that argument. And the Sentinels' third change, the mug, met nothing on Austin's side, because Austin's halftime didn't touch its third-and-long rules. That gap produced the sack at 6:51 of the third quarter.

The bottom row is the lesson. My halftime had a blank exactly where his had a bullet.

The sack, as a halftime adjustment

Chapters 5 and 22 take the sack apart block by block. Here I want you to see it as what it was: the third change on Russ's sheet, executed on the first snap that qualified.

The series before it, our second of the half, showed his first change. On first-and-10 at our 35, we stacked Gaines and Powers to the field so that one of them would get a free release against press. Malik Simmons pressed, the two defenders played the stack by which way each receiver broke, and Simmons broke up Powers's quick out. On second-and-10, a run for three, against a two-high look that rotated at the snap. Then third-and-7 at our 38, at 6:51 of the third quarter: the first third-and-7 or longer we'd faced since halftime.

PRE-SNAP: Week 15, Q3, 6:51, 3rd-and-7, AUS 38, 13-13
(Eleven, Gun Doubles, Harlan flexed left, "60 Left")

                                 F
                                                     $ (at 8)
 C                                         *                 C
                S   E    N  W    M    T
 X       Y           LT   LG   C   RG   RT           H           Z
                         R    Q
 Sutton  Harlan                                  Powers       Gaines
 BOUNDARY                                                    FIELD

📊 Diagram (described): Austin is in two receivers to each side. On the left, the boundary, Sutton is wide and Harlan is flexed about four yards off the left tackle. On the right, the field, Powers is in the slot and Gaines is wide. Ruffin is on Vance's left. The Sentinels show one deep safety in the middle and Holloway down at about eight yards on the field side. On the line, Weller has walked out wide on the boundary side outside Rollins; the nose is head-up on the left guard; Ames and Bell stand in the two A gaps on either side of the center; Coleman is outside the right guard; the nickel is over the field side; and both corners are off.

The rules on our card did exactly what they said, in order. The slide went left, toward the crowd and toward Rollins. That was the side our help had been going all game, which is what the chip had taught them. Ruffin set on Bell. Weller walked late. Darius moved Ruffin out to Weller. And the hot went where every A-gap hot had gone for a season: to Powers, in the field slot.

The timing is the part that belongs to this chapter. Here it is on the play clock.

THE PLAY CLOCK AT 6:51 (seconds remaining)

 40 ...... 27 ............ 15 ......... 10 ............ 3 .. SNAP
            |               |            |              |
       call in at 27   radio cuts     Weller walks   ball snapped
       ("60 Left";     (Lyle can't    out late;       (Bell free,
        Mike ID: 54)    talk after    Vance moves     hot window
                        this)         Ruffin to him   filled by
                                                      Coleman)
 |<--- Lyle can help ----->|<------ only the RULES can help ------->|

📊 Diagram (described): A play-clock timeline from 40 seconds down to the snap. The call goes in at about 27 seconds, with "60 Left" and the Mike identified as number 54, Bell. The coach-to-quarterback radio cuts off at 15 seconds. At about 10 seconds Weller walks out late, and Vance follows the rule and moves Ruffin to him. The ball is snapped with about 3 seconds left. Everything before 15 seconds is time when the coordinator can help; everything after it belongs to the rules the players carry.

I said it to Russ in the spring the way I'll say it to you: "The end walked at about ten seconds… We had to move the back, and the back was the guy who was supposed to be on the Mike."

Notice what Russ did not do. He didn't call the mug on the first third-and-long of the game, or the second. In the first half he stayed quiet on long third downs, and Teo told me so at halftime as if it were good news. Russ was letting our card stay the same until the snap where it would cost us most: tied, third quarter, at our own 38. A good defensive coordinator saves the halftime change he's surest of for the down that matters, and he lets you spend the second half not looking for it.

🛡️ The defense's view: Sit in Russ's chair at halftime. Austin's quick game is working, Austin's quarterback is comfortable, and Austin's third-and-long protection has a rule for everything, including a rule that let a back be moved after the radio cuts. Russ doesn't need to stop what's working. He needs to make it cost more, and he needs one snap where the offense's own rules carry it into a trap. So he makes changes that come in halves: press on about half the early downs, rotate on about half. A change that comes every snap is just a new tendency, and the offense adjusts to it by the second series. A change that comes half the time takes away the quarterback's pre-snap certainty, which is the thing the quick game runs on. And he aims his one big change at a rule he's watched for months, not at anything that happened in the first half. "I fix what you're going to keep doing."

The counter-adjustment: where it worked

Adjusting to their adjustments isn't only defense. In the fourth quarter we finally did it.

By then the Sentinels led 23–13. Teo had been counting Simmons's alignment on first down since the third quarter, and early in the fourth he gave it to me as his "one thing": "Simmons has pressed six of the last seven first downs."

That was Russ's first change, now so consistent it had become a tendency. A corner who presses on first down to take away the slant is a corner who'll drive on the slant. So with a little under eleven minutes left, first-and-10 at our 41, we called "Eleven, Gun Trips Right, 60 Right, Z Sluggo, H Seam, Y Drag." The Sluggo is our slant-and-go. Halftime change 1 was in the formation: trips, Sutton alone into the boundary, Harlan at #3 on a drag. Halftime change 2 was in the protection: Ruffin on Rollins's side to chip. Simmons drove on the slant fake and Gaines ran past him. Forty-four yards, caught at the Sacramento 25 and down at the 15. Three snaps later Ruffin scored from the 2 on Jumbo Power Left: 23–20, at 9:24 of the fourth quarter.

At halftime I'd told Wade that if they took the quick game away, they'd give us grass behind the corners. I'd figured we'd get two or three of those. We got one.

🎙️ On the headset: First-and-10 at our 41, down ten. Teo's "one thing" is Simmons: six of the last seven first downs pressed. That number is the other coordinator's halftime change turned into a habit. The question isn't whether Simmons will press. It's whether he'll press this snap, and the answer is probably yes, because he's been doing it for a quarter and it's been working. I don't need a new play. I need the play that looks like the thing he's pressing to stop. Slant look, go behind it. Chip on 94 so Darius has time to hold the fake. Call it.

The same lesson, five weeks later

Here's how stubborn this lesson is. In the divisional playoff, at Sacramento again, our second-quarter touchdown drive included the bender: Gaines on a six-yard hitch, Powers stemming vertically past nine yards and breaking out at twelve. Darius pumped the hitch, Simmons's eyes jumped to it, and Powers was open for fourteen. In the third quarter, on third-and-6 at the Sacramento 44, we called it again. This time Simmons's eyes stayed on Powers. Russ's staff had seen the first one and, at halftime, reminded him where his eyes belonged. Simmons broke it up and we punted.

I had a first-half winner and called it again after halftime without asking whether it had survived their halftime. That's the same mistake as Week 15, one size smaller.

📋 Try it — Whose fault? The Week 15 sack at 6:51 of the third quarter. Darius identified the Mike correctly and moved the back as the rule told him. Ruffin went where he was sent. The left side of the line blocked its men. The hot throw was where it was supposed to be. Whose fault was it, and what could have been done about it at halftime, using only words the team already had?

Answer

The coordinator's, and specifically the halftime's. No player broke a rule. The rule card did what it said, and the defense had read the card. The halftime question that would have caught it, "What will they change?", was never asked, so a set of rules the Sentinels had studied for months went into the second half untouched.

At halftime you can't install a new protection, and you shouldn't try. But you can do any of these with existing words:

  1. Subtract a rule instead of adding one. "On third-and-long, once the Mike is called, the back doesn't move." That's one sentence for two players, Darius and Ruffin. It removes the walk-up rule the Sentinels were aiming at. It might have left Weller unblocked if he'd rushed, but he was a dropper. A halftime can subtract a rule far more safely than it can add one.
  2. Call a different protection from the menu on third-and-7 or longer. A seven-man call ("70," Harlan stays in) or "80 Max" changes the count without changing anyone's technique.
  3. Put a hot on each side. A quick-game call with an answer on both sides of the ball, so a lineman who drops into one window doesn't close both.

None of these is guaranteed. Russ might have answered any of them. But each would have been a decision, where what we actually had was a default.

🔄 Check your understanding: Why did Russ's first two halftime changes come on "about half" of early downs instead of every early down?

Answer

A change that comes every snap is a new tendency, and the offense adjusts to it within a series or two. A change that comes half the time takes away the quarterback's pre-snap certainty. Darius couldn't be sure, before the snap, whether the corner would press or play off, or whether the shell would stay two-high or rotate. The quick game runs on that certainty, so taking it away made the quick game slower to read without the defense having to stop it outright.


36.8 Don't Overreact — and Don't Under-React

The most common halftime advice in football is "don't overreact." It's good advice. It's also half of the advice, and the half I've failed at is the other one.

Overreacting: treating a half as a season

A first half is roughly thirty snaps. That's not much evidence about anything.

📈 By the numbers: Suppose your offense's true success rate in a game is about 50%. Over thirty snaps, chance alone will push the number you actually see up or down by around nine percentage points fairly often. So a half at 41% and a half at 59% can come from the same offense doing the same things. When you split thirty snaps into personnel groups or concepts, the samples get smaller and the swings get bigger: nine Twelve snaps can look terrible or wonderful by luck. That doesn't mean halftime numbers are useless. It means they're leads, not laws. Marisol's halftime numbers always come with a count beside them, and the count is the first thing I read.

So when Marisol said our Twelve runs were three successful out of eight against big nickel, she said it the way she says everything: "It's a lean." What made "Twelve isn't working" a real halftime line wasn't the three of eight. It was the cause behind it: the Sentinels had a rule, big nickel against every Twelve snap since Week 6, and the third safety was fitting exactly where we were running. A small sample plus a known cause is worth acting on. A small sample with no cause behind it is usually noise.

Here are the overreactions I see most, at every level:

  • Changing a play after one failure. One snap is an event. Ask the booth what the defense did; if it was a coverage the play beats, the play didn't fail — the execution or the luck did.
  • Benching a player at halftime for one mistake. The replacement hasn't been in the game plan all week, and now you've changed a personnel group without meaning to.
  • Abandoning the run after one bad half. The run game is the credibility the rest of your offense borrows. A half without it changes how the second-level defenders play your play-action for the rest of the game. Curtis Bale, a coordinator I worked with on Walt Dupree's staff, learned this one the hard way; you'll see his halftime in Case Study 2.
  • Throwing out the game plan. The plan was built on a week of study. One half of a different look means the plan needs a branch, not a replacement.

The best halftime I ever had changed nothing the players could see

Week 6, our first game against Sacramento. We led 17–10 at the half. In the second quarter Ruffin had broken a 45-yard run on inside zone, partly because Tre Holloway, the Sentinels' strong safety, had taken a flat angle. Teo's notes said Holloway had been coming downhill hard on our inside zone ever since, like a man who'd been embarrassed and wasn't going to let it happen again.

We already had a series card on the call sheet for exactly this: inside zone, inside zone again, then play-action off the same look with Gaines on a post behind Holloway, then inside zone. Teo had built it from Holloway's film before the game. The halftime question wasn't what to call. It was when. I could have fired the play-action shot on the first two-high look of the third quarter. Instead, my halftime change was for the booth. "Teo, Holloway's first step on every run to his side. Up or back. That's your one thing until I say stop."

We didn't change a formation, a protection, or a play. In the second half we ran inside zone at Holloway's side four times before the fourth-quarter drive, and Teo's column read four for four: down. By the time we got to the drive that tied the game, I wasn't guessing. Two more inside zones from the same look, Holloway filling harder each time. Then "Inside PA, Gaines Post" from the identical picture, Holloway biting on the fake, Gaines for 35 yards to the Sacramento 4. Then inside zone again, Holloway hesitating because he didn't trust the run anymore, and Ruffin walked in. 27–27, with 7:31 left. Chapter 19 tells that sequence in full.

The halftime decision that mattered was patience. We were winning. The thing we'd found was real but not proven. So the change was to collect proof instead of spending it. A halftime adjustment can be a question you assign to the booth.

Under-reacting: the tendency that outlived its evidence

Now the other half. Five weeks after Week 15, in the divisional playoff, we trailed 10–7 at halftime. At that halftime, the box on my sheet I didn't touch was third-and-long. I had good reasons. Our new mug protection, center on the Mike, back on the other A-gap linebacker, right tackle widening to the corner blitzer, was built and repped. And Teo's page said the Sentinels had stayed two-high on every third-and-7-plus against our two-by-two formations since their bye.

The first part of that held up beautifully. On the opening drive of the third quarter, third-and-7 at our 41, Russ called the mug with the corner fire again. Zero free rushers. Gaines turned his route into a six-yard hitch when the corner came, and we had a first down at the Sacramento 49. The drive ended in the field goal that tied it 10–10.

The second part is the snap I'll carry for the rest of my career: third-and-9 at our 38, with 2:41 left and the Sentinels up 20–17. "Mills," a post with a dig under it. Two safeties at thirteen yards before the snap. At the snap, one rotated down and Ike Royce spun to the middle, and Darius threw the post into him. Interception.

The two-high streak on third-and-long had been true every time — until the time it mattered. It was a tendency, and I'd let it outlive its evidence. The evidence was all old, and no defensive coordinator keeps a streak going because his history says he should. I wrote a line in red on my whiteboard that spring, and it's still there: "Never let a tendency outlive its evidence."

That's under-reaction. The halftime sheet carried a tendency into the fourth quarter of a playoff game without asking the question that 36.7 is about: what is the other coordinator saving?

💡 Coach's point: Change what they've proven. Watch what they've hinted. Keep what's working until they've shown they've changed it, and assume that the longer something has worked, the more carefully they're planning to take it away.

🪞 Learning check-in: Pause and look at how you've been reading this book. When you read about a play that failed, did you go looking for the player who made the mistake, or for the decision that put him there? When you read a tendency table, did you treat the numbers as facts or as bets with a history? Think about the last game you watched closely, at any level. Can you name one adjustment either team made at halftime, or did you only notice plays? If you only noticed plays, that's normal, and it's the habit this chapter is trying to change. Next time, write down three defensive habits in the first quarter and check them again in the third. Notice which of the five problem types (36.3) you find easiest to spot and which you tend to miss. Most readers find "a tendency they caught" the hardest, because the evidence is on the other team's sheet.

🔄 Check your understanding: In the Week 6 game, Lyle's halftime "change" was an instruction to the booth. What made that the right move in that situation, and what would have made it the wrong one?

Answer

Austin was leading, the Holloway tendency was real but unproven (it rested on one big run and a few fills), and a play-action shot called too early would have spent the discovery before it was sure. Assigning the booth to chart Holloway's first step turned a hunch into evidence (four for four down) before Lyle bet on it. It would have been the wrong move if the offense had been failing and needed a change it could only make with everyone in the room, or if the evidence had already been strong enough to act on. Patience is a decision, not a default; it needs a reason and a plan to collect proof.


36.9 Between Series: The Adjustments That Can't Wait

Halftime gets the attention, but most adjustments in a game happen on the bench, in the minute or two while the defense is on the field. If you save everything for halftime, you've wasted the first half's best teaching moments.

After every series, I ask the same three questions. We call them the sideline three:

  1. What did they play that we didn't expect? Teo's series note answers this.
  2. What did it cost us? A play, a drive, or nothing yet.
  3. What's the first call of the next series, and what's its reason? A call that has no reason tied to what just happened is a call I made before the game, and that's fine, as long as I know it.

Then the position coaches work in parallel. Gus has the linemen on the bench with the tablet stills. Rome has Darius, sometimes with a still photo, sometimes just a sentence. Cal catches the receivers. Teo, in my headset, has one job until the defense's series ends: help me answer question 3.

What can change between series, and what waits

Can change between series Needs halftime (everyone in the room)
A single call, or which calls lead the sheet A new formation family for the second half
A tag on an existing play (a chip, an "Alert," a "Kill") A change to a rule the whole line carries
A hot or sight-adjust reminder to the QB and one receiver A personnel plan for the half (what groups we'll use, and when)
One player's alignment or technique (a split, a set) Anything that needs the QB and four receivers to read the same new picture
The booth's "one thing" Anything you'd want every player to say back

By that table, two of my three Week 15 halftime changes didn't need halftime. "Less Twelve" is a sideline decision: nobody but me and the signalers has to know. And the chip on Rollins was a tag for two players. After his second-quarter sack, on third-and-9 at our 31, with Ruffin aligned away from him, Rollins had already shown us everything we needed. I could have given Gus and Ruffin the chip on the bench, right then. Instead I saved it for halftime, which meant we played the rest of the half without it. That's a halftime change I made forty minutes late.

The honest lesson: if a fix needs only one or two players, make it between series, and save halftime for the change that needs everyone. If I'd moved the chip to the bench and the Twelve call to the sideline, my halftime sheet would have had room on it. Maybe room for the question I didn't ask.

🏫 Any level: In high school, halftime is commonly longer than the NFL's (often around fifteen minutes, set by the state association; check your rules), but you usually have far less to work with: no video, often no tablets, maybe one assistant in the press box with a legal pad and a phone. That makes the charting in 36.2 more important, not less. Give your press-box coach three columns only: safeties (one or two, and where), box count, and what happened. And your players need simpler sentences. A high-school quarterback can usually carry one new thing into the second half, sometimes two. My best halftime as a high-school coordinator in Ohio was a single sentence for my quarterback: "Ball on the hash, the field safety's coming down. Throw where he left." In college, halftime is often longer still (around twenty minutes is common), staffs are bigger, and the rules on sideline and locker-room video have changed in recent seasons, so check your current code. The principles don't change at any level: decide in a few minutes, change two or three things, use words the players already own, and ask what the other sideline is changing.

🎬 Film room: You can watch halftime adjustments on any broadcast or All-22 film. Chart the last two series before halftime and the first two series after it, and track three things: the corners' depth on first down, the shell on early downs (two-high, one-high, rotating), and the number of rushers on third-and-long. Do the same for the offense: personnel groups, formations, and the first call of the half. The first five snaps of the third quarter are the closest thing you'll ever get to reading both coordinators' halftime sheets.

🔄 Check your understanding: Use the table in 36.9 to sort Lyle's three Week 15 halftime changes: which ones truly needed halftime?

Answer

Only change 1 (trips with Sutton alone and the tight end on a drag) truly needed halftime, because a new formation emphasis changes what all eleven players do. "Less Twelve" is a sideline personnel decision. The chip is a tag for two players (Ruffin and Achterberg) and could have been made on the bench after the second-quarter sack. Moving those two out of halftime would have left room on the sheet for the question about what the Sentinels would change.


Spaced Review

Answer each question before you open it.

1. From Chapter 22 (Pass Protection): Protection is arithmetic, and the quarterback is part of the protection. On the Week 15 sack, count it: how many blockers did Austin have, how many of them blocked a rusher, and what was the "extra blocker" that was supposed to cover the free man?

Answer

Six blockers (five linemen and Ruffin), but only four of them blocked a rusher: the right guard and Ruffin both blocked air when Coleman and Weller dropped. The extra blocker was supposed to be the ball: the hot throw to Powers. Coleman dropped into that window, so the ball was covered too. Lyle's whiteboard afterward: "6 blockers + 1 ball = 7. Russ made it 4 + 0."

2. From Chapter 24 (Building the Game Plan): What's the difference between a script and a series card, and why does the dissent written in the plan's margin matter at halftime?

Answer

A script runs at the start of a game in roughly fixed order, built to ask questions. A series card is a conditional chain that waits on the sheet until its trigger fires, with each step's trigger written down. The margin dissent matters at halftime because the losing argument from Tuesday is often the next idea when the call isn't working, and it's already written down with its reasoning. In the divisional week, Rome's margin note next to the third-and-long box asked for a post-snap check on the post. It was on the page at halftime.

3. From Chapter 7 (Understanding Defensive Structure): "The shell is not the coverage." How did Russ's second Week 15 halftime change use that fact against the quick game?

Answer

The shell is the safety picture before the snap; the coverage is what the defense plays after it. By rotating out of two-high on about half of early downs, Russ made the pre-snap shell an unreliable predictor of the post-snap coverage. The quick game depends on the quarterback knowing before the snap where the throw is going, so a shell that might rotate slowed every read.


Project Checkpoint: In-Game Adjustment Protocol

Where you are. In Chapter 35 you built your QB Development Plan: the meeting schedule, the game-management card, and the confidence plan for your quarterback. Game day is where that plan gets tested. This chapter's deliverable is the system your staff uses to read the game while it's happening and to change course at halftime without losing the team.

What to build. An In-Game Adjustment Protocol, two to three pages, with four parts:

  1. The drive log. Your booth or press-box sheet: the columns you'll chart on every snap (use 36.2 as a model and cut it to what your staff can actually fill in at game speed), your "?" rule, and the one-line series note.
  2. The sideline three. Your three between-series questions, who answers each, and your list of what can change between series versus what waits for halftime.
  3. The halftime sheet. One page with four sections: What the half told us (at most three lines), Changes (at most three, each with its owner, its words, its cost, and the first call that tests it), Their next move (what the other coordinator is most likely changing, and your pre-built answer), and Keep (what you will not change, and why).
  4. The second-half script. Three to five calls: one confirm call from a first-half look, one call for each change, and one answer to their likely adjustment.

A worked example. Here is the Week 15 halftime sheet as I'd write it today, with the same first half and the same information:

HALFTIME SHEET -- WEEK 15 (rewritten)          SAC 13, AUS 10

 TOLD US   1. Quick game is a gift; price goes up in Q3.
           2. 94 vs. Ben: 5 pressures / 17.
           3. Twelve vs. Big: a lean, with a cause.
 CHANGES   1. Trips, X alone, Y drag (all 11) -- cost: fewer
              run fits -- test: series 1
           (chip on 94: already done on the bench in Q2)
           (less Twelve: sideline call, not a halftime change)
 THEIR     1. Press + rotate on early downs -> Z Sluggo ready
 NEXT      2. 3rd & 7+: the mug they haven't shown -> 3rd-long
 MOVE         rule: "Mike called, back doesn't move"; hots
              both sides
 KEEP      Inside Slant (pair it); zero bin; Rollins chip rule

📊 Diagram (described): A rewritten version of Lyle's Week 15 halftime sheet in four sections. "Told us" lists the quick game as a gift whose price will rise, Rollins's five pressures on seventeen rushes against Achterberg, and the Twelve problem as a lean with a known cause. "Changes" has only one real halftime change, the trips formation with Sutton alone and Harlan on a drag, with its cost and when it will be tested; the chip and less Twelve are noted as already handled on the bench and the sideline. "Their next move" predicts press and rotation on early downs, with the slant-and-go ready, and the unshown mug on third-and-7 or longer, answered by subtracting the rule that moves the back and putting a hot on each side. "Keep" lists what won't change.

Notice the empty space where two changes used to be. That's room to think about the other locker room.

At your level:

  • High school: Two changes maximum, and the drive log is three columns kept by one assistant. Your "Their next move" box can be one line: "What will they do about our best play?"
  • College: Add a column to the log for the opponent's substitutions, since personnel matching and tempo matter more. Plan for the longer halftime by scheduling the position-room time explicitly.
  • Pro: Build the full four-part protocol, assign every column of the drive log to a named booth coach, and add an in-game self-scout column: where your hot throws and your help have gone this game.
  • Film room: Take a game you can watch in full. Fill in a drive log for one team's first half, write the halftime sheet you'd have written, then watch the third quarter and grade your "Their next move" box. Did you predict the other coordinator's change?

Next: In Chapter 37, you'll pull everything you've built so far into a Coordinator Interview Binder, the packet you'd hand a head coach who's deciding whether to trust you with his offense.


Chapter Summary

The halftime clock (NFL, roughly 13 minutes): walk off (about 1:30) → the coordinator's decision window (about 4 minutes) → position rooms and the whole offense (about 4 minutes) → head coach → warm-up. Decide early; everything after the decision window is transmission.

The five kinds of problems and their usual fixes:

Problem Evidence Usual fix level
A defender One name in the "one thing" column Tag or help rule; use his aggressiveness
A structure An unexpected coverage, front, or personnel answer Formation or personnel change
A price Something works because they're letting it Prepare the double move or shot for when they stop
A tendency they caught Defenders early, guessing right Same look, different play; or a new look
Execution Right call, right look, missed technique One sentence to one player

The triage filter (every candidate change): pattern or play? Who has to change? Can we say it in words we have? What does it cost? What will they do to it?

The halftime sheet: Told us (≤3) · Changes (≤3, each with owner, words, cost, test call) · Their next move (with a pre-built answer) · Keep.

Rules of thumb:

  • Never more than three changes. A team can't absorb four.
  • No new words at halftime. Subtracting a rule is safer than adding one.
  • If one or two players can make the fix, make it between series.
  • The second-half script asks two questions: did our changes work, and what did they change?
  • A half is about thirty snaps: halftime numbers are leads, not laws. Act on a small sample only when you can name its cause.
  • The first half's best plan is the second half's scouting report, for both sidelines.
  • Never let a tendency outlive its evidence, theirs or yours.

What's Next

You've now got a system that can read a game while it's being played, which is most of what a head coach is hiring when he hires a coordinator. Chapter 37 steps back from the game to the career: how coaches get from graduate assistant to the play-caller's chair, what the coaching tree really is, and the crossroads Teo reaches when a college program calls and asks whether he's ready to call plays of his own.