Quarterback Pressure Rate: The Most Underused NFL Stat in Public Coverage

Second-and-eight, late third quarter, Detroit Lions at Kansas City Chiefs, season opener 2023. Patrick Mahomes drops to pass, immediately feels Aidan Hutchinson coming free off the edge from his blind side, slides up and to his right, throws across his body to Travis Kelce for an eleven-yard gain on what the broadcast describes as a “Mahomes miracle.” The replay angles run on a loop for the next three days. Mahomes draws all the credit. The play, by every public metric available, is also a quiet indictment of the Chiefs’ offensive line that night: an unblocked free runner, a quarterback under pressure within 2.4 seconds of the snap, a third-down-style scramble disguised as a designed-rollout completion. The pressure rate against Mahomes that game was 38%, the worst of his Kansas City career. The QB rating he posted was elite. Both things were true. The mainstream coverage told one of those stories. The pressure-rate ledger, sitting unmentioned in the box score, told the other.

Quarterback pressure rate — the percentage of dropbacks on which the quarterback is hurried, hit, or sacked — is, in my opinion, the single most underused stat in public NFL coverage. The metric has existed in some form since the early 2000s, became reliably measurable when PFF started charting every snap in the early 2010s, and is now produced by multiple competing sources (PFF, ESPN, Next Gen Stats). It tells a story that yards, completions, and even EPA per play cannot tell on their own: how often was the quarterback actually given a chance to operate in a clean pocket? The answer, in any given game or season, is frequently surprising. It also explains, more cleanly than any other single number, why some quarterbacks who look great in the box score age into mediocrity, and why some who look mediocre on paper turn out to be elite when the environment finally gives them clean reps.

I have been writing about football analytics since 2014, and the stat I find myself defending most often in conversations with non-analytics readers is the one this article is about. Quarterback pressure rate — what it actually measures, how it interacts with the rest of the public analytics toolkit, where it breaks, and why it deserves a place in every serious NFL conversation, is the subject of this article.

The origin: where pressure rate came from

The earliest serious charting of quarterback pressure happened inside NFL film rooms in the 1980s and 1990s, where assistant coaches assigned a “pressure” label to any play on which the defense disrupted the pocket within roughly three seconds of the snap. The definitions varied team to team, the data was proprietary, and no public-facing version existed.

The first public-facing pressure-rate tracking came from Pro Football Focus, which began commercial operations in 2007 and built its business around frame-by-frame charting of every NFL snap. PFF’s analysts, working from broadcast film, marked every pass play with metadata about the pocket: was the quarterback hurried, hit, sacked, or kept clean? The definitions were standardized internally and applied with reasonable consistency across games and seasons, which produced the first comprehensive public pressure-rate dataset.

By the mid-2010s, the NFL’s own player tracking system — chips in shoulder pads, optical tracking via stadium camera arrays — produced a parallel public dataset via Next Gen Stats. Next Gen Stats measures pressure differently, focusing on time-to-throw and time-to-pressure rather than the PFF binary of “was the QB under pressure on this snap.” The two datasets often agree on extreme cases (an elite QB with a great line versus a struggling QB with a poor line) and can disagree at the margins, which has produced occasional methodological arguments inside the analytics community.

The metric’s adoption in mainstream coverage was slow. The 2018 NFL season is generally considered the inflection point, when ESPN began surfacing PFF pressure data in broadcast graphics and the Athletic’s analytical writers (Sheil Kapadia, Aaron Schatz, others) made the stat a fixture of their weekly columns. By 2026, pressure rate is regularly cited in NFL analytics writing, though far less often than EPA, completion percentage, or yards per attempt.

How pressure rate works: in plain language

The basic mechanic is straightforward. Every NFL pass play is, after the fact, classified by chartists as either a “clean” play (the quarterback had a normal pocket without immediate disruption) or a “pressure” play (the quarterback was hurried, hit, or sacked within the snap-to-throw window). The pressure rate for a quarterback in a given game, season, or career is the percentage of his pass attempts that fell into the latter category.

League average pressure rate in recent NFL seasons has hovered around 32-35%. Elite-line teams (Kansas City’s offensive lines of 2019-2022, Cleveland’s of 2023-24 with a healthy Joel Bitonio) post pressure rates allowed in the 22-28% range. Bottom-tier offensive lines yield pressure on 40-45% of dropbacks, which is a structural problem the quarterback cannot fully overcome regardless of his individual ability.

The companion measurement is quarterback performance under pressure: the QB’s completion percentage, yards per attempt, EPA per play, and turnover rate when the chart classifies the play as a pressure situation. Elite quarterbacks — Mahomes, Allen, Hurts in his peak years — maintain reasonable efficiency even under pressure, with EPA per play that, while degraded, remains positive. Most QBs see their EPA per play collapse into negative territory under pressure, often by a full point per dropback or more.

The conceptual insight is that quarterback evaluation needs to disentangle the player from the protection. A QB posting elite efficiency behind elite protection is doing something good; a QB posting comparable efficiency behind league-average or worse protection is doing something extraordinary. The reverse is also true: a QB with mediocre numbers behind a great line may not be the cause of his team’s offensive problems.

The critical component: pressure rate vs sack rate

The most common mistake in pressure-rate writing is to conflate it with sack rate. The two are related but materially different.

Sack rate measures the percentage of dropbacks that end in a sack. League average sack rate is about 6-7%. A great sack rate (suppressed) is in the 3-5% range; a bad one is 9-11%. Sacks are a downstream consequence of pressure but not the only consequence — a quarterback under pressure who completes a short throw, takes a hit and falls down, throws the ball away, or scrambles for a small gain has all “survived” the pressure without taking a sack.

Pressure rate is the much larger universe — typically four to six times the sack rate. A team that allows pressure on 35% of dropbacks but holds opponents to a 5% sack rate is allowing significant disruption that the box score does not fully capture. The 30 percentage points between pressure and sack are a quarterback’s scrambling, throw-away, and hot-route survival skills — and a measure of how much work the QB is doing to keep the offense functional.

Pressure rate vs the alternatives: a comparison

Pressure rate sits in a constellation of stats that measure different parts of the QB and offensive line’s joint output. A short comparison:

The honest reading uses two or three of these in concert. A QB with low TTT, low pressure rate allowed, and elite clean-pocket EPA is the platonic ideal of a modern quarterback (peak Patrick Mahomes). A QB with high TTT, high pressure rate, and decent EPA under pressure is a different kind of player (vintage Aaron Rodgers, holding the ball waiting for receivers to win). Neither is wrong; they are different jobs.

What the data needs: inputs

Pressure rate is, fundamentally, a charting stat. The minimum inputs are frame-by-frame video of every NFL pass play and a trained chartist or computer-vision model that can classify each play as clean or under pressure. PFF’s commercial operation employs hundreds of chartists who work through every NFL snap during and after each week. Next Gen Stats uses optical tracking data and player-position chips to derive a similar signal automatically.

The two methodologies can disagree by a few percentage points on any given team’s season-long pressure rate. PFF’s manual charting is, in my experience, slightly more conservative — they classify fewer borderline plays as pressure. Next Gen Stats’ automated approach is more liberal but applies the threshold consistently across all teams. For public-facing writing, citing the source matters: a “32% pressure rate” can be a different number depending on which feed you pulled it from.

For writers who want to do this work without commercial data access, the public version is more limited. ESPN publishes some PFF-derived pressure metrics in its game summaries and weekly graphics. nflfastR includes sack data and some pressure-adjacent fields (pass attempts under pressure flagged in certain seasons), though not with the depth that PFF provides. For most serious pressure-rate writing, the PFF subscription remains close to mandatory.

Building the analysis: a working framework

The practical workflow:

1. Start with pressure rate allowed by the team’s offensive line, season-to-date and last 4 games. Identify whether the OL is improving, declining, or stable.
2. Pull the QB’s split numbers: clean-pocket EPA per play, pressured EPA per play, completion %% under pressure, sacks-per-pressure rate. The differential between clean and pressured tells you how much the QB is being asked to overcome.
3. Compare to the league baseline. A QB whose pressured EPA per play is -0.6 (close to league average) is performing about as well as the average QB does in those situations. A QB whose pressured EPA is -0.2 is doing something exceptional.
4. Check the time-to-throw metric. A quarterback with elite pressured-EPA but a 3.2-second TTT is creating his own pressure problem; the same player at 2.5 TTT is genuinely beating the rush with quick reads.
5. Cross-reference with film for scheme context. Some offenses are designed to take longer to develop (the Andy Reid scheme historically); others are quick-game-driven (the Mike McDaniel scheme). The pressure-rate context is partly a function of play design.

Where this gets weird: common mistakes

The pitfalls writers fall into with pressure data.

Attribution confusion. A high pressure rate can be the offensive line’s fault, the play-caller’s fault, or the quarterback’s fault. A QB who holds the ball too long generates pressure that, in the chartist’s eyes, is on him; the offensive line metrics will not necessarily reflect that. Public coverage often credits or blames the OL alone, ignoring the joint nature of the outcome.

Small samples are noisy. A single game’s pressure rate is highly variable. A QB facing the 2024 Detroit defense for sixty minutes will look catastrophically different from the same QB facing a mediocre defense the following week. Pressure-rate writing should always aggregate at least three games of data before drawing strong conclusions.

Cross-source comparison. PFF pressure rate and Next Gen Stats pressure rate are different numbers. Cross-referencing them as if they measure the same thing introduces noise. The careful version names the source.

Garbage-time pollution. Pressure rates spike in obvious passing situations (trailing late, long-and-distance). Filtering for competitive game-state is essential for meaningful comparison across teams whose typical situations vary.

The play-action confound. Play-action passes, particularly those with hard runs faked, produce different pressure dynamics than standard dropbacks. A team that runs heavy play-action volume will appear to have lower pressure rates partly because the rush is delayed by the run-fake action. The metric does not always disentangle this cleanly.

When pressure rate shines: use cases

The applications where the stat earns its rent:

Quarterback evaluation across different offensive environments. Comparing a QB on a great line to a QB on a poor line on raw counting stats is fundamentally unfair. Pressured-EPA per play normalizes the environment and produces evaluation that survives a roster change. Justin Herbert’s 2022 season, for example, looks meaningfully different through the pressure-adjusted lens than the raw box score suggested.

Offensive line evaluation. The pressure rate allowed by an offensive line is, in my opinion, a better season-long evaluation of OL play than the traditional metrics (sacks allowed, hurries allowed by individual lineman). The aggregated number captures the unit’s effectiveness in a single statistic.

Identifying mismatches in upcoming games. A defense with a high pressure rate generated, against an offense with a high pressure rate allowed, is a matchup the underdog has a structural path to win. The 2024 Houston-Cleveland playoff game was a textbook example — Cleveland’s defensive pressure rate was elite, Houston’s OL had been mediocre, and the game played out roughly as the pressure-rate gap suggested.

Roster construction. A team building around a quarterback with poor pressured-EPA needs to invest heavily in offensive line. A team with a quarterback who excels under pressure can deprioritize OL investment relative to other roster needs. The math, applied carefully, drives meaningful draft and free-agency decisions.

A working example: Joe Burrow’s 2022-2024 trajectory

Joe Burrow’s three-season window from 2022 through 2024 is one of the cleaner pressure-rate case studies of the recent era. In 2022, Cincinnati’s offensive line was a near-bottom-tier unit; Burrow faced pressure on 38% of dropbacks, sixth-highest among starting quarterbacks. His clean-pocket EPA per play was +0.42 (elite); his pressured EPA per play was -0.41 (better than league average, given the context). The differential — 0.83 EPA per play between clean and pressured — was significant but in the typical range. The team made the AFC Championship game on the back of Burrow’s clean-pocket performance and what was, by the metric, a roughly average performance when pressure was applied.

In 2023, Cincinnati’s line improved meaningfully and Burrow’s pressure rate dropped to 30%. His clean-pocket EPA fell slightly to +0.36 (a normal year-over-year fluctuation) and his pressured EPA rose to -0.19, much better than league average. The team’s offensive efficiency improved as a function of both fewer pressures and better performance under the ones that did happen. The 2024 season, which ended with the team underperforming on the win-loss ledger, was nonetheless a pressure-rate continuation of 2023 — Burrow was facing fewer pressures and managing them better than ever. The wins did not follow because the defense regressed, not because the QB had taken any step backward.

That kind of granularity — separating the QB from the line from the defense — is exactly what pressure rate, used carefully, makes possible. Mainstream coverage of those three Cincinnati seasons frequently conflated all three storylines into a single “Burrow is great” or “Burrow is struggling” narrative depending on the week. The pressure-rate ledger told a steadier, more coherent story.

The limits: what pressure rate cannot tell you

The honest version of pressure-rate writing names the limits.

Pressure rate cannot, on its own, tell you whether a quarterback is good. It tells you about the environment the QB is operating in and how he responds to it. A QB facing little pressure can still post poor numbers if he makes bad decisions in clean pockets; a QB facing constant pressure can still post elite numbers if his skill in the chaos is sufficient. The metric is one ingredient.

Pressure rate cannot fully distinguish player skill from scheme. A coordinator who designs an offense that gets the ball out in 2.3 seconds will produce a quarterback whose pressure rate looks artificially low; a coordinator who runs deep concepts with 3.0+ second timing will produce the opposite. Scheme matters in the data, and the writers who use pressure rate well incorporate scheme context.

Pressure rate cannot replace film study for receiver and route evaluation. The metric captures the pocket experience and what happens to it. It does not capture, for example, whether the receivers are getting open quickly or running clean routes. A QB with elite pressured-EPA may be partly the beneficiary of receivers who beat their coverage; cross-checking with separation data fills that gap.

Pressure rate, finally, cannot eliminate the joint-product problem at the heart of football analytics. The QB, the OL, the receivers, the play-caller, and the defense are all contributing to every play in interacting ways. Pressure rate isolates one specific dimension — the pocket experience — with reasonable accuracy. It does not resolve the broader attribution question, which is the genuinely hardest problem in public NFL analysis.

One final limit, less methodological. Pressure rate data is still primarily commercial. PFF’s subscription is the canonical source; Next Gen Stats provides a partial public version; nflfastR includes some derived fields. For writers without paid access, the version of pressure-rate analysis that is possible in 2026 is meaningfully less rich than what’s available to ESPN columnists or club analytics departments. The democratization of public NFL data, which has been impressive in the EPA era, still has the pressure-rate gap to close.

Frequently asked questions

What is a good pressure rate for an NFL offensive line?

League average pressure rate allowed has hovered around 32-35% in recent seasons. An elite line keeps pressure rate to 22-28%. A struggling unit yields 40-45%. The Kansas City lines of 2019-2022 and the Detroit lines of 2023-24 are recent examples of units that consistently kept their quarterbacks clean. The Carolina and New England lines of recent seasons are examples of units that have struggled to do so.

How do PFF and Next Gen Stats define pressure differently?

PFF’s chartists classify a play as pressure when the quarterback is hurried, hit, or sacked within the snap-to-throw window, based on frame-by-frame video review. Next Gen Stats uses the optical tracking and chip data to flag pressure when a defensive player is within a threshold distance of the QB within a threshold time after the snap. The two definitions overlap heavily but produce slightly different season-long numbers — usually within 2-3 percentage points for the same team.

Does pressure rate correlate with winning?

Yes, but not as cleanly as EPA per play. Teams that allow low pressure rates and generate high ones tend to win, but the relationship is mediated through quarterback play and defensive scheme. A team with elite pressure rate generated and mediocre QB play can still struggle; a team with high pressure rate allowed and an elite scrambling quarterback can still win. The metric is a strong predictor but not a deterministic one.

Where can I see public pressure rate data?

Pro Football Focus’s website publishes summary tables behind a paywall; their public Twitter/X account surfaces weekly leaders. NFL Next Gen Stats provides a partial public version. ESPN’s game pages occasionally include PFF-derived pressure data. The fullest analysis requires a PFF subscription, which is, for serious football analytics writing in 2026, a near-essential tool.

Sources and further reading

• Pro Football Focus — the canonical commercial source for pressure-rate data, with weekly leaders and historical archives.
• NFL Next Gen Stats — the league’s official tracking-data feed, including pressure-adjacent metrics.
• Sheil Kapadia at The Ringer — one of the public-facing writers who has consistently integrated pressure-rate analysis into mainstream NFL coverage.
• nflfastR documentation — the open-source backbone for EPA analysis, with some pressure-related fields available.
• Bill Barnwell’s writing — long-form NFL analysis that consistently uses pressure-rate context as part of QB evaluation.

The Mahomes-to-Kelce scramble that opened this article — second-and-eight, edge pressure, miracle completion — was the highlight of that Monday morning’s broadcast loop and the kind of play that, if you write only what the box score knew, becomes a one-paragraph note about the quarterback’s escape ability. The pressure-rate ledger, sitting unmentioned, told the deeper story: that Mahomes was working harder for his completions that night than the box score reflected, and that the offensive line in front of him had been the actual story of the game. For the broader frame on how to read modern football efficiency without missing context, our guide to EPA is the natural companion piece.