Rams 26, Broncos 30 — Week 3: Up Sixteen at Half, Lost by Four

Third and six at the Denver 38, five oh one left in the fourth quarter, Rams up 19 to 16. Matthew Stafford threw short left for Kyren Williams, and Talanoa Hufanga took it at the Denver 34 and ran 66 yards the other way for a touchdown. 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 — that one play was −9.3 for Los Angeles. By win probability, the model's read on each team's chance of winning at that moment, the Rams went from 80 percent before the snap to 27 after it, the biggest swing of the night. Los Angeles had led 16 to nothing at halftime, and when Denver punted on its first possession of the third quarter the model gave the Rams 91 percent. They led for about 39 minutes of game clock. Denver led for about two, and won 30 to 26 on Sunday night in Denver.

The passing game is where the yards and the value came apart. 57 pass plays — that count includes sacks — the most of the 30 teams that have played this week, and 366 net passing yards, second so far this week behind only Seattle. Those 57 plays were worth −1.1 expected points, seventeenth of the 30 teams per play. Six snaps explain the gap: the two interceptions and four sacks together cost 17.8 expected points, and the other 51 pass plays were worth +16.7. For the comparison: at the rate this passing game played over the first two weeks, 57 plays would have been worth about +7; at last season's rate, when it ranked third in the league, about +13. The best of it came right after the pick-six. Down 23 to 19, Stafford found Konata Mumpfield for 21, then for nine, and on second and one at the Denver 48 with three 46 left he went deep right to Mumpfield for a 48-yard touchdown: +4.4 expected points and 29 points of win probability. 26 to 23, Rams.

The scoreboard problem lived inside the 20. Five Rams possessions reached the Denver 20 and produced one touchdown, three field goals and the interception that ended the game — 12 pass and rush plays inside the 20, one touchdown. That touchdown was the opening score, second and goal from the five with five oh oh left in the first quarter, Stafford short right to Tyler Higbee. Then the flags. Los Angeles had 13 accepted penalties for 91 yards, to Denver's five for 38. On third and two from the Denver eight early in the second quarter, a holding call wiped out a three-yard run and made it third and 12; incomplete, field goal. Later in the quarter, from the Denver 15, a delay of game and then offensive pass interference on Davante Adams erased a 16-yard catch to the Denver four, and that drive ended in a 35-yard field goal on fourth and 11. And in the third quarter, up 16 to eight, a 60-yard punt was muffed by Denver, recovered at its own 15, and the returner was tackled at the nine. An unsportsmanlike conduct penalty on the Rams wiped the play out, the re-punt came on fourth and 32 from the Rams 12, and Denver started at its own 48. Seven plays and a two-point try later it was 16 all.

The run game did its part. 23 rush plays, a hundred 16 yards, +2.3 expected points, fifth of the 30 teams per play so far this week. At the rate it ran over the first two weeks those carries would have been worth about −2.6; at last season's rate, when it was a top-10 run game, about +0.6. It faded with the lead: 13 rush plays for 87 yards in the first half, 10 for 29 after it. The longest was Williams up the middle for 30, first and 10 at the Rams 33 with one 19 left in the first quarter.

The pass defense had the first half, and then it had the last three minutes. 35 Denver pass plays — two-point tries not counted — were worth −3.2 expected points; on defense a negative number is the good direction. That was twelfth of the 30 teams in total, and a hundred 79 net passing yards allowed was eighth-fewest so far this week. In the first half Denver ran 21 pass and rush plays for 67 net yards and did not score. At last season's rate, when this pass defense ranked tenth, 35 plays would have been worth about −2.6; at the rate of the first two weeks, about −6 and a half. The takeaway came at 16 all: second and 14 at the Rams 48, 13 oh seven left, Bo Nix short left for Courtland Sutton, and Josh Wallace intercepted at the Rams 37 and brought it back 26 yards to the Denver 37 — a swing of 5.3 expected points and 16 points of win probability toward Los Angeles, which turned it into a field goal. Then the last Denver drive, from its own 35 with three 38 left, Rams up three. On second and 20 from the Denver 25, Nix hit Pat Bryant for 39 yards, 21 points of win probability. On third and six from the Rams 11 with one 15 left, Wallace intercepted at the eight and was flagged for pass interference on the same play; first and goal at the seven. On third and five from the five, an incompletion, and an illegal-use-of-hands flag on Quentin Lake made it first and goal at the two. Those two flags were worth about nine and 10 points of win probability to Denver.

The run defense held Denver to 78 yards on 25 rush plays, and 15 of the 25 gained three or fewer. Those plays were still worth +0.6 expected points, twentieth of the 30 teams so far this week; on defense a positive number is the bad direction. At the rate of the first two weeks they would have been worth about +0.3, and at last season's rate, when this run defense ranked in the top 12, about −1. The last two decided the game. From first and goal at the two, after an incompletion, Nix went around right end and was ruled in, and replay put the ball at the one. After an encroachment flag on third and one, Nix went up the middle for the touchdown with 48 seconds left. 30 to 26. The Rams still reached the Denver 14 with three seconds left, on a 20-yard catch by Williams on second and 10, but after a spike and a false start it was second and 15 from the 19, and Stafford's throw deep right for Adams was intercepted by Hufanga in the end zone.

For anyone new to this show, here is where that sits. Last season the Rams went 12 five with the third-ranked passing game in the league by expected points per play, a top-10 run game, a pass defense that ranked tenth and a run defense in the top 12. Week 1 was a 27 to seven loss to San Francisco in Melbourne, in which the passing game ranked 20-ninth of 32 seven points was the fewest any team scored that week. Week 2 was a 28 to six win over the Giants, with the passing game third and the pass defense second in the league per play. Sunday landed in between: the second-most net passing yards of the 30 teams that have played this week, and a per-play rate that ranked seventeenth. The thread that has not moved in three games is the giveaways — two in Week 1, two in Week 2, two on Sunday, and one of Sunday's was a touchdown for the other team with five minutes left. Three games do not settle what this passing game is. They do say that at one and two, the Rams have given the ball away exactly twice in every game.

Los Angeles is at Philadelphia next Sunday, October fourth.

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