Patriots, Week 2 — What Repeats: Nineteen Light Boxes to Six

2026 NFL Season · Thursday, Sep 24

New England's defense showed a light box on 19 of Seattle's 20 designed runs in Week 1 and on 6 of Pittsburgh's 21 on Sunday, yet allowed about four yards per designed run both times, while the offense faced the second-lowest light-box share in the league and ran for 4.96 a designed run after finishing last in Week 1.

0:007:47

Your episode · 7:47

The full episode, in writing

Last week in Seattle this defense showed a light box, six or fewer defenders, on 19 of 20 designed runs. On Sunday it showed one on six of Pittsburgh's 21. That is 95 percent, the second-highest share in the league in Week 1, turned into 29 percent, tied with Detroit for 20-third of the 32 teams this week and under a median of 44. Across all of last season New England showed a light box on 42 percent of the designed runs it faced, so it went from far above its own habit to well below it in consecutive games. Here is the part to take into Sunday. The yards per run did not move with the front. Pittsburgh's 21 designed runs gained 83 yards, just under four a run, fifteenth of the 32. Seattle's 20 gained 84, 4.2 a run, also fifteenth that week. Last season this defense allowed about 4.2 per designed run. Light front or loaded front, the run defense has landed close to four yards a run all three ways.

The rush repeated in direction, if not in size. New England sent five or more pass rushers on 16 of Pittsburgh's 46 dropbacks, 35 percent, fourteenth of the 32 this week. Last week it was 44 percent against Seattle, the fifth-highest. Across last season it was 28. Two games, both above last year. The Recap counted four sacks; the charters marked none of them as Aaron Rodgers's fault, and he left the pocket on seven of his 44 dropbacks, scrambles included. Pittsburgh ran play-action on 11 of its 46 dropbacks, 24 percent, seventeenth, close to the 25 percent opponents ran against this defense last season, a week after Seattle ran it on 41 percent.

Behind that rush, here is the ball charting on Pittsburgh's 40 throws. 26 of the 40 were charted catchable, 65 percent, tied with Philadelphia for 20-fifth of the 32, among the lower rates any defense allowed this week and under the 73 percent opponents managed against New England last season. Three were contested, a receiver and a defender arriving together, seven and a half percent, against 16 percent last season. And none were interception-worthy: New England was one of 12 defenses that did not draw a single one this week. Its interception came on a throw the charters did not mark as a defender's ball. Last week the one interception-worthy ball it drew went uncaught. Two games, and the takeaways and the charting have not lined up either way.

On offense, the run game is the week's reversal. Pittsburgh showed a light box on three of New England's 24 designed runs, 12 and a half percent, the second-lowest share any offense saw this week; only Seattle saw a lower one. Last week Seattle gave New England six or fewer on 10 of 25, 40 percent, and last season this offense saw a light box on 35 percent of its designed runs. So this was the more crowded front of the two games by a distance, and it produced the better result: a hundred 19 yards on 24 designed runs, just under five a run, the fifth-highest of the 32. Last week, against the lighter fronts, it was 65 yards on 25, 2.6 a run, last in the league. Last regular season it was about four and a half.

The passing game carries one figure that has now repeated, and it is not a good one. Drake Maye threw 22 passes and one was charted interception-worthy, a ball a defender should have caught. That is four and a half percent, the sixth-highest rate of the 32 offenses this week. Last week it was three of 33. Four across 55 throws in two games, from a quarterback who put 14 in the air on 493 throws all of last regular season. Both weeks above his own rate, the same direction twice. What reversed is the traffic. Only one of his 22 throws was contested, four and a half percent, tied with the 49ers for 20-sixth, after six of 33 last week and 13 percent across last season. And the accuracy has not moved at all: 16 of 22 catchable, 73 percent, sixteenth and on the median, the same 73 percent as last week and a few points under his 76 last regular season. One drop on the 16 catchable balls. Maye left the pocket on five of his 27 dropbacks, scrambles included, after 14 of 42 last week, and no sack was charted as his fault. Off play-action his six throws went three for 70; the other 16 went 11 for a hundred 38.

The screen game has repeated too. Three screens on 27 dropbacks, 11 percent, tenth-highest in the league this week, after six on 42 last week. That is nine on 69 dropbacks across two games, against 26 on 614 all of last season, four percent. Two weeks, both more than double last year's rate. No-huddle: none at all for the second week running, one of six offenses at zero this week, after two percent last season. Play-action on six of 27 dropbacks, 22 percent, nineteenth and just under the median, after 21 percent last week and 24 last regular season, a little lower both times, not by much. Motion on 24 of 67 charted plays, 36 percent, 20-fifth, after 44 last week and 42 last season.

The stat lines have settled, so here they are, scored half-PPR. TreVeyon Henderson led the team with 13.6, 16 carries for 76 and the touchdown. Romeo Doubs caught three of four targets for 96, 11.1, a week after no catches on three targets. Maye, eight even: 14 of 22 for 208 and the interception, +4 carries for 17. Hunter Henry three of five for 40, five and a half. Eli Raridon one catch for 30. Rhamondre Stevenson six carries for 43 one catch for three, 3.1. For Pittsburgh, Jaylen Warren led with 8.7 on 11 carries for 43 two catches for 34.

About the comparisons. For the first three weeks this show sets each game beside last regular season, since there is no within-season baseline yet. That was a different roster, a year ago, so the comparison flags what is out of the ordinary and cannot yet tell you what is real. Week 1 is last week, a single game, not a season rate. Two games settle nothing. What they allow is a first sort, into what moved the same way twice and what did not.

The thread, for anyone starting here. The Recap had last season at 14 three, with the best passing offense in the league by expected points per play and the eleventh-ranked pass defense. This season New England lost 13 to 10 in Seattle and beat Pittsburgh 20 to three. Moving the same way twice: Maye's interception-worthy rate above his own, screens at more than double last year's rate, no no-huddle, and five or more rushers on defense more often than last season. Reversed: the defense's light box, from nearly every run to under a third of them; the fronts the run game faced, and its yards per designed run with them; and Maye's contested throws. What goes in the notebook: the defensive box count has swung from one extreme to the other while the yards per designed run stayed near four, and the quarterback's loose balls, along with his catchable rate, have not come back to his own level. Two games. Written down, not called.

That's your Patriots — every game, by the numbers. This is Muffed: we do the same deep dive for all 32 teams. Follow the Patriots show wherever you're listening, so next week's recap lands the moment it's out. No theater. Just the football.

And if you play fantasy, there's a Muffed show built around your exact roster — free at muffed.ai.

_Data via nflverse. Charting: FTN Data via nflverse._

Every Patriots episode

Season reviews, draft recaps and every game recap, in one place.

More like this