Broncos 14, 49ers 24 — Week 4: Run Defense Held, a Punt Didn't

Three to three at halftime in San Francisco, and Denver's defense had the better of the half. Singleton stopped Brock Purdy's fourth-and-one sneak at the Denver 34 on the first snap of the second quarter, and Allen blocked a 54-yard field goal as the half ran out. Then, tied with nine 52 left in the third, Marvin Mims fielded a 61-yard 49ers punt at the Denver 28, got one yard, and Neal forced the fumble and recovered it for San Francisco. The 49ers scored nine snaps later, and the Broncos lost 24 to 14 to fall to two and two. Neither offense turned it over, and neither quarterback was sacked.

The passing game opens on the halves, because they were two different games. Before halftime Bo Nix's 24 pass plays were worth −3.2 expected points added, which measures what a snap was worth given the down, the distance and the field position. After it, 18 pass plays were worth +7.9. For the day, 42 pass plays, 214 yards, one touchdown, +4.7. Over the first three weeks this passing game ran at about −9 hundredths a play, a rate at which those 42 would have been worth about −3.9. So it ran well above its own rate, in a loss. Denver's biggest swing came early: third and eight at the San Francisco 43, scoreless in the first quarter, Nix short left to Mims behind the line, and Mims went 14 after the catch for 10 yards and the first down, 10 points of win probability, the model's read on each team's chance of winning at that moment. Lutz kicked a 52-yarder on the same drive. RJ Harvey caught all 10 of his targets for 68 yards. Courtland Sutton drew six targets and 98 air yards and caught one, for four. The charting answers last week's question about the arm. Not one of Nix's 42 throws was interception-worthy, after seven in the first three weeks, and none was contested, after 14 percent. 28 of 42 were catchable, 67 percent, a little better than his 65 over the first three weeks. The pressure look changed: San Francisco sent five or more rushers on 19 of his 44 dropbacks, 43 percent, against 31 percent from Denver's first three opponents, and Nix left the pocket on 16 of the 44, 36 percent, against 22. Play-action stayed in the background, eight of 44 dropbacks, 20-first of the 30 charted offenses; the run-pass option came seven times, as many as in the first three weeks combined.

The run game opens on what it got per designed run. 20 designed runs, 72 yards, 3.6 a carry, against 3.9 over the first three weeks. San Francisco showed six or fewer in the box on 10 of the 20, half of them, fourteenth of the 30 charted offenses and close to Denver's own 52 percent. In all, 22 rush plays for 82 yards, −0.02 expected points and 41 percent successful; at the first three weeks' rate, about −0.8. J.K. Dobbins carried 11 times for 40, Harvey seven for 25. The run game's best stretch was the touchdown drive midway through the fourth, down 17 to six: five runs from Dobbins and Harvey for 30 yards, Nix to Jaylen Waddle for a 13-yard touchdown on first and 10, and Tyler Badie up the middle for the two points that made it 17 to 14. That touchdown was Denver's only snap of the day inside the San Francisco 20.

The pass defense opens on one throw. 29 49ers pass plays were worth +8.1 expected points; on defense a positive number is the bad direction. At this defense's rate over the first three weeks, −6 and a half hundredths a play, those 29 would have been about −1.9. 5.5 of the eight came on a single snap: 17 to 14, two 25 left, second and eight at the San Francisco 44, Purdy short left to George Kittle three yards behind the line, and Kittle ran 59 yards after the catch for the touchdown. 15 points of win probability, and the game. Denver sent five or more rushers on 15 of Purdy's 33 dropbacks, 45 percent, sixth-highest of the 30 charted defenses this week and above its own 38 percent from the first three weeks, with no sacks. Purdy's throws were catchable on 20 of 29, 69 percent, against 66 for Denver's first three opponents, and none was contested, after 22 percent. The 49ers converted six of 13 third downs.

The run defense is the unit that turned around. 33 49ers rush plays, a hundred 42 yards, and −0.7 expected points; on defense, negative is good. Over the first three weeks this run defense allowed about +2 tenths a carry, a rate at which those 33 would have been worth about +6.7. It did it with a lighter box. Denver showed six or fewer on 16 of San Francisco's 29 designed runs, 55 percent, after 38 percent in the first three weeks, and the 29 gained 4.1 a carry, against 5.2 allowed before Sunday. Christian McCaffrey had 15 carries for 52 yards. Short yardage was where it showed: the fourth-and-one stop, a third-and-one stop at the San Francisco 21 that forced a punt, and McCaffrey stopped at the one twice. The 49ers scored on the next snap both times.

The turning point was the return, not the defense. The punt was worth +6 and a half expected points and nearly 21 points of win probability to San Francisco, the largest swing of the game. Even then the defense made the 49ers convert a fourth and one at the 19 before Purdy found Deebo Samuel from the one. Denver's last chance came down 10 with two oh seven left: Nix converted a fourth and three to Evan Engram, then threw incomplete for Sutton twice from the San Francisco 45, found Waddle for five on third and 10, and Lutz's 58-yard try went wide right with a minute 18 left.

Third downs: eight of 17. Nine drives: one touchdown, two field goals, two three-and-outs.

The stat lines, scored half-PPR: Harvey, 14.3, 10 catches for 68 +7 carries for 25. Nix, 13.6, on 25 of 42 for 214 and a touchdown. Waddle, 12 even, five catches for 35 and the score. Badie, 6.9. Dobbins and Pat Bryant, 6.2 each; Bryant caught all three targets for 47. For San Francisco, Purdy scored 19.6, Samuel 15.5 and Kittle 15.

What repeats. Last week's notebook line was the arm: three weeks of interception-worthy rates going up, and the least catchable passing in the league against the Rams. Sunday answered it, zero interception-worthy throws and a catchable rate back above his own average. The older line, heavier boxes from this defense against the run, broke the other way, and the run defense allowed a negative total for the first time this season with the lighter look. For anyone new: Denver lost in Kansas City 31 to 10, beat Jacksonville 20 to 13 and the Rams 30 to 26, and lost here 24 to 14. The one for the notebook now is the box. Watch whether Denver keeps showing the lighter front against the run, and whether the yards stay down when it does.

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

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