Run-Stuff Rate and the Defensive Front: Measuring Disruption at the Line

Run defense gets graded by the same lazy number that grades run offense: yards allowed per carry. And it's just as misleading on this side of the ball. A defense can give up a respectable 4.0 a carry while quietly getting shoved off the ball on most snaps, its average rescued by a few tackles in the open field. Or it can surrender that same 4.0 while winning the line of scrimmage play after play, undone by one or two long runs that leak through. Yards per carry can't see the difference, because it's an average and the front's real work happens at the extremes — specifically, at the line of scrimmage, where a defense either blows the run up or it doesn't. The stat that lives at that line is stuff rate, and it's the cleanest window we have into how a defensive front actually plays the run.

The one-sentence version: stuff rate is the share of opponent runs a defense stops at or behind the line of scrimmage, and because it counts wins at the point of attack rather than averaging yards, it measures front-seven dominance in a way yards per carry can't.

What stuff rate measures

A "stuff" is a running play stopped for no gain or a loss — the back meets the defense at or behind where the ball was snapped. Stuff rate is simply how often that happens:

Stuff rate = (opponent runs stopped at or behind the line) / (opponent rush attempts)

That's the whole definition, and its power is that it's binary and front-loaded. It doesn't care whether a stop is for a one-yard loss or a four-yard loss; it cares that the defense won the rep at the line. This makes it a near-direct readout of the front's performance, because stopping a run before it starts is mostly a function of defensive linemen and linebackers beating blocks, filling gaps, and tackling at the point of attack. It's the mirror image of the offense's opportunity rate: where the offense wants to gain that first-level hole, the defense wants to deny it, and a stuff is the most emphatic denial there is. A defense with a high stuff rate is one that consistently wins the line of scrimmage; a low stuff rate means the front is getting displaced and the back is regularly reaching the second level untouched.

Why it beats yards per carry for grading a front

Yards allowed per carry fails the front for the same reason it fails the offense: it's a skewed average, and one long run distorts it. A defense can stuff the back on eight of ten carries and still "allow 5.0 a carry" if the two it missed went for 30 and 20 — a line-of-scrimmage-dominant front that the average makes look mediocre. Stuff rate refuses to be hijacked that way. It scores each carry as a win or a loss at the point of attack and reports how often the front won, full stop. That doesn't make it the only number you need — a defense that stuffs a lot but surrenders the occasional 60-yarder has a real explosive-run problem stuff rate won't show — but it isolates the one thing yards per carry blurs: how often the front actually controlled the line. The honest read on run defense uses both: stuff rate for control at the line, plus an explosive-run measure for the chunk plays that leak through.

The relationship to havoc

Stuff rate is a close cousin of havoc rate, and it helps to see exactly how they overlap and where they don't. Havoc rate counts all the plays a defense blows up — tackles for loss, forced fumbles, and passes defensed — as a share of total snaps, run and pass alike. Stuff rate is narrower and run-specific: it's the run-defense piece of disruption, the share of rushing attempts ended at or behind the line. The two share a chunk of their DNA — a run tackle for loss is both a stuff and a havoc play — but they answer different questions. Havoc asks "how disruptive is this defense overall, against everything?" Stuff rate asks "how well does this front specifically stone the run?" A defense can post big havoc numbers built mostly on its pass rush and secondary while having a merely-fine stuff rate, or the reverse. Reading them together separates a front that wins against the run from a defense that wins mostly through the air — a distinction that matters enormously when you're projecting a matchup against a run-heavy opponent.

How front-seven play shows up in the data

The reason stuff rate is so useful is that it's where a defensive front's quality becomes legible. Good interior linemen who hold the point of attack and two-gap, edge defenders who set a hard edge, and linebackers who fill downhill and arrive on time — all of that work converts into stuffs. When a front is winning, you see it as a cluster of related signals in the play-by-play: a high stuff rate, a healthy run tackle-for-loss count, opponents falling behind schedule into passing downs, and a low rate of opponent runs reaching the second level. When a front is losing, the same signals invert — the stuff rate sags, backs get to the linebackers untouched, and opponents stay on schedule by running on early downs. Stuff rate is the single best summary of that cluster because it sits right at the line of scrimmage, the exact spot the front-seven battle is decided.

One honest caveat: stuff rate is partly a product of what it faces. A defense that sees a lot of obvious run situations — short yardage, goal line, opponents protecting a lead — will have more chances to stuff than one facing pass-first spread teams all year, and game script can inflate or deflate the raw number. That's why a careful read situates stuff rate against the run volume and down-and-distance a defense actually saw, rather than treating the bare percentage as the final word. But even raw, it's a far better front-seven gauge than the yards-per-carry average it should replace.

A worked example: same yards allowed, different fronts

Take two hypothetical run defenses, each allowing 4.2 yards per carry over a season. The figures below are invented to show the contrast, not measured from any real team:

Identical by the box-score average, opposite by construction. The Anvils win the line of scrimmage constantly — a quarter of all carries die at or behind the line — and their only blemish is the occasional long run that leaks through a gambling front; their problem is explosiveness, not control. The Sieves rarely stuff anybody; they're getting steadily pushed for three and four yards a pop but happen not to give up the back-breaker, so the average lands in the same place. Against a power running team in November, you'd vastly rather have the Anvils' front, and only stuff rate tells you that the two defenses aren't remotely the same. (Again, these figures are made up to illustrate the idea.)

How to use it

• Read stuff rate as line-of-scrimmage control. It's the share of runs the front kills before they start — the most direct gauge of front-seven dominance there is.
• Don't trust it alone. A high stuff rate with the occasional 60-yarder hides an explosive-run problem — always pair it with an explosive-run measure.
• Keep it distinct from havoc. Havoc covers all disruption, run and pass; stuff rate is the run-specific slice. Read both to see whether a defense wins up front or through the air.
• Mind the game script. Facing lots of obvious run situations inflates the chance to stuff — situate the rate against the run volume and down-and-distance the defense actually saw.
• Prefer it to yards per carry for grading the front. The average gets hijacked by a couple of long runs; stuff rate counts wins at the point of attack, where the front's work really happens.

The next time a run defense is "giving up four a carry," ask the question the average buries: four a carry how? A front that stones the back at the line and leaks one long run is doing the opposite job from a front that's quietly getting moved every snap, and only stuff rate separates them. Yards per carry tells you the result; stuff rate tells you who won the line — and on run defense, that's the part that travels.