Bengals 27, Steelers 30 — Week 3: A Muffed Punt, Then a Strip Sack

11 oh four left in the fourth quarter, Cincinnati up 27 to 24, and Pittsburgh punting on fourth and 10 from its own 36. By win probability — the model's read on each team's chance of winning at that moment — the Bengals were at 67 percent, the highest they reached all afternoon. The punt went 51 yards, Williams muffed the catch at the Cincinnati 16, and Pittsburgh's Ben Skowronek recovered at the 14. 26 points of win probability on one snap, and by expected points added — what a snap was worth given the down, the distance and the field position, measured against what that situation was already worth — +5.5 for the Steelers. Pittsburgh gained −4 yards from there and still tied it with a field goal. Then it took the lead with another, and with 53 seconds left, Cincinnati down three and at the Pittsburgh 38, Joe Burrow was sacked by Harmon, fumbled, and Pittsburgh recovered. The Bengals went from 45 percent to three. They lost 30 to 27 in Pittsburgh on a day their offense produced more expected points than Pittsburgh's did.

The passing game is the contradiction. 38 pass plays — that count includes sacks, and there was one, the last — produced +14.7 expected points, 269 net passing yards and three touchdowns, and 58 percent of them were successful plays, meaning they gained what the situation asked for. By total expected points that is second of the 28 teams that have played this week, behind only San Francisco. A week earlier in Houston, 28 percent of Cincinnati's pass plays were successful, the third-lowest of the 92 team games played this season. At the rate the Bengals passed over the first two weeks, those same 38 plays would have been worth about −2; at last season's rate, when this passing game ranked 20-third, about −1. It came against a Pittsburgh pass defense that ranked first in the league in Week 1. Two of the biggest snaps were fourth and two. At 17 all, 10 11 left in the third quarter, from the Pittsburgh 42, Burrow went deep right to Ja'Marr Chase for 21, pushed out at the 21: +4.2 expected points, 17 points of win probability. Down 24 to 20 with 13 33 left in the game, from the Pittsburgh 25 after a no-gain run on third down, Burrow went deep middle to Mike Gesicki for the touchdown: +4.9, 19 points of win probability, and the lead.

The run game was a small part of the day. 18 rush plays, two of them Burrow scrambles, tied for the second-fewest of the 28 teams that have played this week. They gained 83 yards and +1.1 expected points, and half were successful — by expected points per rush play, eighth of those 28 teams, better than Week 2 and a little under Week 1. Seven of the 18 gained three yards or fewer, and the longest was 13. On third and two or shorter, Cincinnati converted three of six.

The pass defense gave back what it had built over two weeks. 35 Pittsburgh pass plays were worth +8.6 expected points to the Steelers — on defense a positive number is the bad direction. At the rate this defense allowed over the first two weeks, when it ranked eighth and then seventh in the league, those 35 would have been worth about −6 and a half. 273 net passing yards allowed is 20-fourth of the 28 teams so far this week. Pittsburgh's first two possessions both ended in touchdown passes: four plays to open the game, finished by Aaron Rodgers deep left to Roman Wilson for 38 on third and five, then Rodgers to Skowronek for 23 with 57 seconds left in the first quarter. The unit's best work came in two places. Down 17 to 10 with one 46 left in the half, Rodgers threw short middle from the Cincinnati 48 and Jordan Battle intercepted at the 36: −3.1 expected points for Pittsburgh, 10 points of win probability. On the fifth play of the next drive Burrow found Tee Higgins deep right for a 28-yard touchdown with 39 seconds left, and it was 17 all at the half. And both of Pittsburgh's fourth-quarter trips inside the Cincinnati 20 reached third and five, and both times Rodgers was sacked. Two field goals instead of touchdowns.

The run defense was the best in the league in Week 2 — by expected points per rush play, the best of all 92 team games played this season. Sunday, 21 Pittsburgh rush plays gained a hundred 40 yards and +2.7 expected points, 20-fifth by total expected points allowed of the 28 teams so far this week. At the rate of the first two weeks, those 21 would have been worth about −6. Jaylen Warren's 35 yarder at 14 to 10 in the second quarter set up a field goal, and his 26 yarder to the Cincinnati six at 27 all set up the one that won it.

Two snaps on Pittsburgh punts swung the second half. Pittsburgh lined up to punt twice all afternoon. The first, fourth and six from its own 41 with six 48 left in the third and Cincinnati up 20 to 17, was downed at the Cincinnati six — and wiped out by roughing the kicker on the Bengals, 15 yards and a Pittsburgh first down. 14 and a half points of win probability, +3.6 expected points for the Steelers, and on the seventh snap after it Rodgers hit DK Metcalf for a six-yard touchdown on third and two. The second was the muff. Of Pittsburgh's 30 points, 10 came on drives those two plays kept alive or started. The offense's ledger, eight drives: three touchdowns, two field goals, two punts — one of them after a three-and-out — and the fumble. Six of 12 on third down, two of two on fourth.

For anyone new to this show, here is where that sits. Last season Cincinnati went six and 11 with a passing game that ranked 20-third by expected points per play, a run game in the top 10, a third-down offense fifth, and a defense that ranked 20-eighth against the pass and 20-ninth against the run. Week 1 was a 33 to 27 home win over Tampa Bay in which the defense forced four turnovers. Week 2 was a 20 to six win in Houston carried by the best run-defense game any team has played this season, while the passing game ranked 20-first. Sunday flipped both. The passing game had its best day of the three, and the defense allowed more than last season's rates would have predicted: about +6 on those pass plays and about +2 on the runs, against +8.6 and +2.7 actual. Three games do not settle which defense is the real one. The Bengals are two and one, and the two biggest win-probability swings of the game were both snaps where they lost the ball.

Cincinnati is at home against the Jaguars next Sunday, October fourth.

That's your Bengals — every game, by the numbers. This is Muffed: we do the same deep dive for all 32 teams. Follow the Bengals 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._