Patriots 20, Steelers 3 — Week 2: A Strip Sack and Ten Stops

Second and 21 at the Pittsburgh 30, 13 46 left in the fourth quarter, New England up 13 to three. Christian Barmore sacked Aaron Rodgers at the 21 and the ball came out. Elijah Ponder picked it up at the 19 and ran it in. 10 points of win probability on one snap, the biggest swing in New England's favor all day, and the end of any suspense. The play before it, a holding penalty had erased a 15-yard Pittsburgh completion, which is how it got to be second and 21. The Patriots beat the Steelers 20 to three at home on Sunday, led from the eight 29 mark of the first quarter to the end, and by the win probability model never dipped below about a 57 percent chance of winning.

Start with the pass defense, because it is the story and because it argues with last week. Pittsburgh ran 44 pass plays and they were worth −21.8 expected points added. Expected points added is what a snap was worth given the down, the distance and the field position, and on defense a negative number is the good direction. Of the 28 teams that have played this week, only Carolina's pass defense has been better. For the comparison: at the rate this defense allowed passes last season, when it ranked eleventh in the league, those same 44 plays would have been worth about −2 and a half. And last week in Seattle this unit allowed +4.4 on 26 pass plays, 20-third of 32, without a takeaway. On Sunday it had four sacks, an interception and a touchdown of its own. Here is the thing to repeat to another fan. Those five plays, the four sacks and the interception, took 19.8 expected points from Pittsburgh. The other 39 pass plays took two. The Steelers' longest completion went 26 yards.

The sacks came at the right addresses. Third and seven at the New England 22 in the second quarter, up 10 to nothing: Christian Elliss sacked Rodgers and knocked the ball loose, and by the time Pittsburgh fell on it the ball was at the Pittsburgh 42. Fourth and 43, and a punt. One series after Ponder's touchdown, Joshua Farmer sacked Rodgers for a loss of 11 on third and five. And after Pittsburgh's 88-yard drive, first and goal at the four with five 46 left, Ponder sacked Rodgers back at the 17, a loss of 13. On the next snap Gabe Jacas intercepted a short throw to the right at the 21 and returned it 11 yards. Those were the only two snaps Pittsburgh ran inside the 20.

Third down is where it showed up drive after drive. 10 stops in 12 tries, second of the 28 teams so far this week, behind only the Jets. Five three-and-outs forced, eight punts. Seven of Pittsburgh's third downs needed between three and six yards, the manageable kind, and on the seven snaps that counted the defense won every one. This part is not new. In Seattle the defense stopped nine of 11, tied for fourth-best in Week 1, so across two games that is 19 stops on 23 third downs. Three points allowed ties Carolina and Minnesota for the fewest this week, and 278 yards allowed is fourth-fewest of the 28. Pittsburgh's three came on a 56-yard field goal with five seconds left in the half.

Against the run the day was ordinary, and that was enough. 23 rush plays, 91 yards, −2 expected points, eighteenth so far this week per play. 13 of the 23 were held to three yards or fewer, and the success rate allowed was 30 percent.

The passing game is the unit that still owes something. Four completions carried it: 29 to Hunter Henry, 30 to Eli Raridon, 29 63 to Romeo Doubs. That is a hundred 51 of the 208 passing yards. The other 10 completions went for 57. And for all of that, 25 pass plays were worth −5.3 expected points, 20-fourth per play of the 28 teams that have played this week. At the rate this offense passed last season, when it was the best in the league, those 25 plays would have been worth about +7. The gap is two giveaways. First and 10 at the Pittsburgh 34 on the opening drive, scoreless, Drake Maye threw short left to Rhamondre Stevenson, Nick Herbig forced a fumble and Pittsburgh recovered at its 31. −12 points of win probability, the biggest swing of the game in either direction. Then early in the third, up 10 to three, one play after Maye hit Doubs for 29 on third and five, he threw deep left for Henry from the Pittsburgh 33 and Patrick Queen intercepted it at the 10. −9. Those two plays cost 8.3 expected points between them, more than the passing game's whole deficit. The best snap was the 63-yarder: first and 10 from the New England eight, two 12 left in the third, Maye deep right to Doubs, out of bounds at the Pittsburgh 29. Eight points of win probability.

What the offense did not do is finish. One offensive touchdown and 13 offensive points. After Raridon's 30 yards put the ball at the Pittsburgh six, a false start, a sack, a four-yard completion and another sack turned it into a 30-yard field goal. Those three plays were the only pass or rush plays New England ran inside the 20. After the 63-yarder, a delay of game and two incompletions around a seven-yard run made it a 45-yard field goal. In between, third and one at the Pittsburgh 27 became a pass that lost two, and Andy Borregales missed from 47, wide right. Third down overall was four of 12, tied for nineteenth so far this week, and no conversions in six tries when it needed seven yards or more.

The run game gave the offense its touchdown and its steadiest work. A hundred 39 rushing yards is fifth of the 28 teams that have played this week, and 26 rush plays were worth +1.7 expected points, sixth. The score came on first and 10 from the Pittsburgh 39, eight 29 left in the first quarter, TreVeyon Henderson through left guard for 39 yards and the game's first points. Two snaps earlier Maye had been sacked for a loss of 10 on third and five, and a defensive offside wiped it out. Stevenson added 24 up the middle in the third quarter. The honest split: those two runs gained 63 yards, the other 24 rush plays gained 76, and 14 of the 26 gained three yards or fewer. The success rate was 42 percent, twelfth so far this week. Last week this run game ranked 20-fifth of 32.

Now the longer view, for anyone meeting this team this week. Last season the Patriots went 14 three with the best passing offense in the league by expected points per play, the eleventh-ranked pass defense and the sixth-best third-down offense. In Week 1 they lost 13 to 10 in Seattle with three giveaways, none forced, a pass defense that ranked 20-third and a passing game that ranked 20-second. Sunday reversed the defensive half of that: second against the pass, two takeaways, a score. It did not reverse the offensive half. The passing game has now been worth −9.9 expected points on 61 pass plays across two games, from a unit that led the league a year ago, and the offense has given the ball away five times. Two games do not settle it, and Pittsburgh's passing game had been near the bottom the week before as well. The defense looks like last year's or better. The passing game does not yet.

New England is at Jacksonville next Sunday. Watch the giveaways first, and watch what happens inside the 20.

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._