Where Have All the Seconds Gone? 

Where are teams spending it?

A little of everything. With clock stoppages taken out, nearly every situation costs more than it did in 2015. You can see that in basically every situation, the amount of game clock teams use has gone up.

Game-clock seconds from one snap to the next on snaps where nothing stopped the clock, 2015 to 2025: up 1.25 for all plays, from 0.98 before a pass to 1.63 before a run. The no-huddle got 1.68 seconds faster.

Every bar but one is up between 0.98 and 1.63 seconds, biggest before a run. The exception is the no-huddle, 1.68 seconds quicker than in 2015: the teams still using it go faster with it.

Moving the start year to 2011 shrinks most bars, since 2015 was a quick season, but the huddled snap is still 1.38 seconds slower.

What about the stoppages?

The teams on the field aren't the only things controlling the clock. Officials, the rulebook, and league-wide points of emphasis can stop it between plays without a timeout. The clock stops when a runner goes out of bounds (outside the final two minutes of the first half and five of the second), for injuries, pre-snap penalties, quarter ends, replay reviews, and occasionally to re-spot the ball or move the chains.

These brief stoppages aren't recorded as stops, but they can be measured when the game clock interval and real time of day interval show a mismatch. So for example if the game clock shows 30 seconds ran, but we see on the clock it went from 3:13:25 PM to 3:14:10 PM (45 seconds), we know a stoppage has occurred.

Games in 2025 averaged about 13 such stops, down from 18.5 in 2015. Each eliminated stoppage lets more game clock run off.

Clock stops per game between clock-running snaps, both teams, 2015 to 2025: after a first down 2.4 to 0.8, after other plays 2.7 to 0.7, out of bounds 6.4 to 4.5.

That matters because the game clock determines when a game ends. If teams take the same amount of real time between snaps but the clock stops less often, more time runs off the scoreboard. That's what the “clock stopped less” bar captures: 0.58 seconds per play (95% range 0.49 to 0.68).

The data can't tell us who caused the change. It shows when the clock stopped, not why, and no rule change during this period directly explains the trend. The likeliest explanation is officials restarting the clock sooner after routine plays, a change in practice rather than the rulebook.

Let's do some accounting... again.

As with the last article, we're going to need to do some accounting. We started with the question of why teams were taking 1.6 seconds longer to run a play. We can first address two small changes that roughly cancel out (+0.22 and −0.21 seconds): the fact that teams call/run plays that result in a running clock more often adds only ~0.2 seconds, a trend that is equal and opposite to the fact that plays take just a little bit less time on average.

Most of those 1.6 seconds come from teams simply taking longer to run their next play: 1.04 seconds, roughly two-thirds of the increase. The other third, 0.58 seconds, comes from the clock stopping less often between plays, for a mix of reasons covered above. Together those two make up the 1.6 seconds.

Is it finally turning around?

Will the 5,000-yard passer ever return? At some point, one would hope so. Through Week 5 of 2026 a play costs 30.16 seconds of game clock, down from 30.58, the first real drop since 2019.

Five weeks is a small sample, but three quarterbacks are on pace for 5,000 yards: Bryce Young (5,389), Matthew Stafford (5,345) and Jared Goff (5,049). Early paces fade, though: at this point last year there was one, Stafford, and he finished at about 4,700.

As seconds might be coming back, so too will the yards.

tl;dr

At Thanksgiving, when your uncle starts lamenting the death of the 5,000-yard passer and the golden age of QBs, tell him:

“Passing yards are down, but it’s not because QBs have gotten worse or special teams have shortened the field. Drives have actually gotten longer. There are just fewer of them, because every play eats a little more clock.

Mostly that’s teams taking longer between snaps; the rest is refs stopping the clock less often. It only costs each team about three plays a game, and the 17th game roughly makes that back.

The bigger change is what teams do with the plays they get. They run more, and more dropbacks end in a scramble instead of a throw. Fewer throws per play, on fewer plays, is what keeps the best passers a few hundred yards short of 5,000. And 5,000 is a cliff: Stafford finished last season at about 4,700.

So, yes, passing yards are down. But the explanation isn’t that today’s QBs are worse. It’s that they throw less often. Now pass me the gravy.”

Notes:

You may have noticed that teams taking longer is +1.25 seconds in Figure 3 but +1.04 in Figure 5. The 1.25 only costs game clock when the clock is running, about three plays in four; after an incomplete pass it’s stopped however long teams take. That brings it to 0.92 a play. It comes back up to 1.04 because part of each gap used to be the play itself, and plays got shorter, so the teams’ own share grew by more than the gap did.