Darnold's game under center has held up. Everywhere else has fallen off

Darnold's success under center has held up. The gun is a different story

nflverse play-by-play, Sumer charting (computed 2026-10-12 by 2026-10-12-darnold-alignment-bootlegs.py)

nflverse play-by-play, Sumer charting (computed 2026-10-12 by 2026-10-12-darnold-alignment-bootlegs.py)

nflverse play-by-play, Sumer charting (computed 2026-10-12 by 2026-10-12-darnold-alignment-bootlegs.py). Shotgun includes pistol. Ranks are among QBs with 50+ under-center and 150+ shotgun dropbacks in 2025, and 15+ and 40+ in 2026.

Seattle calls more under-center pass plays than almost any team in the NFL. Sam Darnold lines up under center on 35 percent of his dropbacks this season. The rest of the league does it 21 percent of the time.

In 2025, under center was where Darnold did his best work. He averaged +0.32 expected points added per dropback there, 6th in the league, with a 55 percent success rate (7th) and 10.7 yards per attempt (2nd). He threw 18 touchdowns against 6 interceptions on 160 attempts, the 3rd-best touchdown rate among qualifying passers. The shotgun lagged behind on most measures, +0.04 EPA per dropback (23rd) and a touchdown rate that ranked 38th of 41, even though its interception rate was a little better (31st, against under center's 21st).

This year, under center is still where his indicators look best: a 59 percent success rate (4th in the league) and 9.9 yards per attempt (4th). Three interceptions explain most of what's gone wrong. A play-action throw intended for AJ Barner at Washington in Week 3 cost 8.7 expected points. A play-action throw intended for Emanuel Wilson against the Chargers in Week 4 cost 5.7. A deep throw intended for Elijah Arroyo against San Francisco in Week 5 cost 5.1, about 19.5 expected points combined. Take just those three plays out, and Darnold's other 36 under-center dropbacks average about +0.5 EPA each. He's also taken five sacks under center, three of them against San Francisco. It's a 110-dropback sample, worth watching rather than a verdict.

From the gun, Darnold's EPA per dropback has fallen to -0.14 (28th), with a 37 percent success rate that ranks 31st of 36. The one bright spot: six of his seven touchdown passes this season have come from the gun, five of them in the red zone, three of those to Jaxon Smith-Njigba.

Including runs, Seattle lines up under center on 50 percent of its offensive snaps this season, down slightly from 53 percent in 2025. On run plays, that share has slipped from 74 percent to 70 percent, while the rest of the league has gone from 57 percent to 65 percent.

Sumer charting

Seattle still bootlegs as much as last year. It just doesn't work anymore

nflverse play-by-play, Sumer charting (computed 2026-10-12 by 2026-10-12-darnold-alignment-bootlegs.py)

nflverse play-by-play, Sumer charting (computed 2026-10-12 by 2026-10-12-darnold-alignment-bootlegs.py). A bootleg here is a rollout with a play-action fake. Ranks are out of all 32 teams.

Seattle rolls Darnold out on a bootleg about as often as it did last year, 11 percent of dropbacks, 4th in the league both seasons. In 2025 it was the best bootleg in football: +0.60 EPA per dropback, 1st of 32, with 39 of 54 passes completed for 667 yards, 8 touchdowns and 2 interceptions. This year it ranks 28th.

The 2026 numbers: 18 bootlegs, -0.21 EPA per dropback, 12 of 16 for 118 yards, zero touchdowns and one interception. Most of that drop is a single play, the Week 3 throw intended for AJ Barner at Washington, the same interception that cost Darnold under center above. Take it out, and the other 17 bootlegs average about +0.29 EPA, still ahead of the league's +0.10.

Seattle rolls left more than almost any team, 58 percent of its bootlegs in 2025 against a league rate of 32 percent. Both directions worked that year, +0.44 EPA rolling left and +0.83 rolling right. This year the left side has collapsed: 9 bootlegs, -0.98 EPA, 6 of 9 for 37 yards, with throws averaging under a yard downfield. The right side is still good: 8 bootlegs, +0.82 EPA, 6 of 7 for 81 yards. The one bootleg that went to the middle this year was a 9-yard sack against San Francisco.

Drew Lock went 6 for 6 on his bootlegs in the first two weeks, +0.26 EPA per play. Darnold is at -0.45 on his 12, or about +0.30 without the interception. Eighteen bootlegs is a small sample, worth tracking, not a verdict yet.

Sumer charting

Darnold is holding the ball longer, and man coverage is the wall

Sumer coverage charting, nflverse EPA

Sumer charting, nflverse passer ids

Darnold is taking longer to throw the ball this season, 3.07 seconds on average, up from 2.73 seconds in the 2025 regular season. He was fastest early last year, 2.66 seconds in Weeks 1 through 10, then 2.80 seconds the rest of the way, and the share of his throws under 2.5 seconds has fallen from 51.5 percent then to 38.8 percent now. The extra time isn't coming from a clean pocket: his time to throw with no pressure has barely moved, 2.50 seconds in 2025 to 2.58 seconds in 2026. Under pressure it has jumped from 3.43 seconds to 3.81 seconds, and he's getting pressured far more often, 39.8 percent of his throws against 25.2 percent a year ago. Some of that rise is league-wide: pressure rates across the NFL on these same dropbacks have gone from about 34 percent to 43 percent in 2026.

Seattle was already a slow-dropback team. In 2025 its 2.73-second average ranked 11th-slowest of 32 teams, just above the 2.68-second league average. This year Seattle ranks 4th-slowest at 2.93 seconds, a number that includes Drew Lock, who played most of the first two games. Darnold alone, at 3.07 seconds, is slower than every team's average this season. Miami is the slowest team at 3.02 seconds.

Seattle is sending about the same number of players into routes as it did last year, 4.53 per dropback in the first half of 2025 and 4.49 in the second half, against 4.60 and 4.59 in 2026. The slower clock isn't about extra blockers staying home or extra receivers going out.

Against man coverage, almost everything about Darnold's game has fallen off. His EPA per dropback has gone from +0.275 in 2025 to -0.600 this year. His completion rate has dropped from 67.3 percent to 42.9 percent, his yards per attempt from 8.98 to 3.10, and his success rate from 57.9 percent to 20.8 percent. He's also getting pressured far more often on these throws, 58.3 percent against 28.0 percent a year ago, and sacked more too, 12.5 percent against 4.7 percent. Part of that, as above, is the same league-wide pressure rise, not just Darnold or Seattle's line.

Sumer coverage charting, nflverse results. Dropbacks include sacks and scrambles.

Against zone coverage, he's fine. His completion rate and yards per attempt have both gone up, and EPA per dropback is still positive, +0.05 this year against +0.12 last year. Zone is not the problem.

Sumer coverage charting, nflverse results. Dropbacks include sacks and scrambles.

How teams are playing man against him has changed too. Cover 0, man coverage with no deep safety, has gone from 15.0 percent of his man dropbacks to 33.3 percent. Man with a blitz has gone from 56.1 percent of his man dropbacks to 62.5 percent, and man on third down from 36.4 percent to 58.3 percent. Last year Darnold beat the blitz in man coverage, +0.362 EPA per dropback. This year it's -0.634.

Sumer coverage charting, nflverse EPA

Overall he isn't seeing much more man coverage, 20.7 percent of his dropbacks in 2025, 21.8 percent now, and the league's rate is about flat too. But his last two opponents leaned on it: the Chargers played man on 26.9 percent of his dropbacks and the 49ers on 31.4 percent, 29.5 percent across the two games. Twenty-four man dropbacks is a small sample, and one interception, a third-and-10 pick from Seattle's own 11 against San Francisco that cost 3.8 expected points, moves the number a lot.

Sumer charting

Seattle stopped max protecting, and it worked every time it didn't stop

Seattle used to max protect more than any team in football. This year it barely does, and the results have been great.

Williams outside beats both tackles inside

Does Seattle's line do better with a tackle kicked outside? The alignment it leans on most says yes, against the one it uses second most.

Denver's Alex Singleton has struggled in man coverage all season

Seattle's next opponent has a linebacker who grades out poorly in man coverage.

Three of Russell Wilson's four draft picks are starting for Seattle

Denver sent Seattle four draft picks for Russell Wilson. Here's what each one turned into, and what it costs to keep them.