How Football ChangedThe long arc
NFL Quarterbacks Are Getting Sacked Like It’s 1999
In the 1960s and ’70s, the league’s old season totals show a little over 8 sacks in every 100 NFL pass plays. The play-by-play logs, which start in 1999, show the rate falling to about 6 in 2014–16, then climbing back to about 7, the 1999 level. Scrambles rose with it.
By Edgehalla ResearchPublished 10 min read
On Monday night, Oct 2, 2023, the Seattle Seahawks sacked the New York Giants eleven times at MetLife Stadium, tying their franchise record. Daniel Jones took ten of them. Seattle won 24–3.
“We take responsibility for those 11 sacks,” the Giants’ right tackle, Evan Neal, said afterward, the Associated Press reported. “There is no excuse whatever.”
By the end of that season the Giants had been sacked 85 times, the most of any team since play-by-play records begin in 1999. League-wide, 2023 set the high mark in those records too, 7.2 sacks in every 100 pass plays, a little above the 7.0 of 1999.
On this page
In the 1960s and ’70s, a little over 8 pass plays in 100 ended in a sack
A season of 85 sacks makes it easy to believe quarterbacks have never had it worse. The league’s records go back to 1963, and the seven worst seasons all came in the first 15.
Sacks fell unevenly for five decades, to lows in 1994 and 2016, then climbed back to the 1999 level
Sacks per 100 pass plays (throws plus sacks; scrambles left out). Regular season.
Show data
| Year | Old season totals, 1963–98 | Play-by-play logs, 1999–2025 |
|---|---|---|
| 1963 | 8.5% | – |
| 1964 | 10.0% | – |
| 1965 | 8.8% | – |
| 1966 | 8.8% | – |
| 1967 | 7.8% | – |
| 1968 | 8.3% | – |
| 1969 | 8.1% | – |
| 1970 | 8.2% | – |
| 1971 | 7.5% | – |
| 1972 | 7.8% | – |
| 1973 | 8.9% | – |
| 1974 | 7.9% | – |
| 1975 | 8.3% | – |
| 1976 | 9.1% | – |
| 1977 | 8.9% | – |
| 1978 | 7.9% | – |
| 1979 | 7.5% | – |
| 1980 | 7.0% | – |
| 1981 | 6.7% | – |
| 1982 | 7.8% | – |
| 1983 | 8.0% | – |
| 1984 | 8.4% | – |
| 1985 | 8.3% | – |
| 1986 | 7.6% | – |
| 1987 | 7.6% | – |
| 1988 | 6.8% | – |
| 1989 | 7.2% | – |
| 1990 | 7.3% | – |
| 1991 | 6.6% | – |
| 1992 | 7.8% | – |
| 1993 | 6.8% | – |
| 1994 | 5.9% | – |
| 1995 | 6.1% | – |
| 1996 | 6.5% | – |
| 1997 | 7.4% | – |
| 1998 | 7.2% | – |
| 1999 | – | 7.0% |
| 2000 | – | 7.0% |
| 2001 | – | 6.9% |
| 2002 | – | 6.4% |
| 2003 | – | 6.3% |
| 2004 | – | 6.8% |
| 2005 | – | 6.7% |
| 2006 | – | 6.6% |
| 2007 | – | 6.1% |
| 2008 | – | 5.9% |
| 2009 | – | 6.1% |
| 2010 | – | 6.2% |
| 2011 | – | 6.4% |
| 2012 | – | 6.2% |
| 2013 | – | 6.7% |
| 2014 | – | 6.4% |
| 2015 | – | 6.1% |
| 2016 | – | 5.8% |
| 2017 | – | 6.4% |
| 2018 | – | 6.8% |
| 2019 | – | 6.7% |
| 2020 | – | 6.0% |
| 2021 | – | 6.3% |
| 2022 | – | 6.7% |
| 2023 | – | 7.2% |
| 2024 | – | 6.9% |
| 2025 | – | 6.9% |
- 1978: 1978 blocking rules
- 1994: 1994: 5.9
- 2016: 2016: 5.8
The line starts high in the 1960s and falls in fits. It touches lows in 1994 and 2016, then climbs.
Picture Jones on one of those dropbacks. If he gets the throw away, that is a pass play. If Seattle gets to him first, that is a pass play too, and a sack. If he tucks the ball and runs, the play is left out, because the old records never counted scrambles.
In 1964, by the league’s old season totals, quarterbacks went down on 10.0 of every 100 pass plays, about one in ten. No season since, in either the old totals or the play-by-play logs, is higher. Counting all 15 seasons from 1963 to 1977 together, the rate was a little over 8 in 100 (8.4).
Each of those old season tables lists every offensive play, plus the passes and the runs. Take away the passes and the runs, and what is left is the sacks. Before 1970 the tables cover NFL teams only, not the AFL.
The fall came in fits, with big bounces back
From there the rate fell, though never in a straight line. Decade by decade, the old season totals show it lower every time. They have about 8.6 sacks in every 100 pass plays in the 1960s, 8.2 in the 1970s, 7.5 in the 1980s and 6.8 in 1990–98.
Season by season, it bounced. The rate was nearly 9 in 1977. In 1978 the league let offensive linemen block with their arms out and their hands open. The rate was 7.9 that year and just under 7 by 1980. By 1984 it was back above 8. It dropped to 5.9 in 1994. By 1997 it was back above 7.
In the old season totals, bounces like those were normal. A typical season moved about half a sack per 100 pass plays from the one before. The long fall was much bigger. From 1963–77 in the old totals to 2014–16 in the play-by-play logs, the rate dropped by more than 2 sacks per 100 pass plays, about five times a typical bounce.
The play-by-play logs pick up the last leg down, to 2016
From 1999 every snap is logged, so Jones’s ten sacks at MetLife each have their own line. Those play-by-play logs carry the last leg of the fall. In 1999, the one season both the old totals and the logs cover, the two agree to within a tenth of a sack per 100 pass plays.
The logs start at 7.0 sacks in every 100 pass plays in 1999. The rate averaged about 6.5 in 1999–2009 and 6.4 in the 2010s, and hit bottom at 5.8 in 2016, the lowest season in the logs. The old season totals had their own low, 5.9 in 1994. Those two lows are too close to rank. One comes from the old totals and one from the logs, so neither gets called the record.
Even at the bottom, teams sat far apart. In 2016 the Raiders’ offense was sacked 18 times, about 3 in every 100 pass plays and the lowest rate in the league. Cleveland was sacked 66 times, more than 10 in 100, the highest.
The table lists every period from both sources. Read the old totals against each other and the logs against each other; a row from one source shouldn’t be lined up against a row from the other.
| Years | Source | Sacks per 100 pass plays |
|---|---|---|
| 1963–69 | Old season totals | 8.6 |
| 1970–79 | Old season totals | 8.2 |
| 1980–89 | Old season totals | 7.5 |
Continues below
| Years | Source | Sacks per 100 pass plays |
|---|---|---|
| 1990–98 | Old season totals | 6.8 |
| 1963–77 | Old season totals | 8.4 |
| 1978–98 | Old season totals | 7.2 |
| 1999–2009 | Play-by-play logs | 6.5 |
| 2010–19 | Play-by-play logs | 6.4 |
| 2020–25 | Play-by-play logs | 6.6 |
| 1999–2001 | Play-by-play logs | 6.9 |
| 2014–16 | Play-by-play logs | 6.1 |
| 2023–25 | Play-by-play logs | 7.0 |
Since 2016 the sack has climbed back to the 1999 level
Then it turned, and every season since 2016 has been higher. Single seasons bounce, so take three at a time, all from the logs. The rate was 6.1 sacks in every 100 pass plays in 2014–16 and 7.0 in 2023–25. In 1999 it was 7.0. The sack is back to where the play-by-play logs began. The 1960s and ’70s, when the old season totals had it above 8, are still a long way off.
Seasons in the logs bounce less than they did in the old totals. A typical change from one season to the next is about a quarter of a sack per 100 pass plays, or about one and a half sacks for one team over a season. The climb from 2014–16 to 2023–25 was about 0.9, about three times that, and it has lasted three seasons, so it is more than a bounce.
In sacks, the climb added about 530 over 2023–25, about one more every three games for each of the 32 teams. That count compares two numbers: the sacks the league actually had in 2023–25, and the sacks the same pass plays would have produced at each down’s 2014–16 rate.
Third downs supplied about half of the extra sacks
About half of those 530 extra sacks came on third down. Every other down added some.
Every down got more sacks, and third downs supplied about half the extras
Sacks per 100 pass plays (throws plus sacks), play-by-play. Third downs split by yards to go: short 1–3, medium 4–6, long 7+.
Show data
| 2014–16 | 2023–25 | Change | |
|---|---|---|---|
| 1st down | 4.9% | 5.3% | +0.5 pts |
| 2nd down | 5.3% | 6.0% | +0.7 pts |
| 3rd and short | 6.1% | 8.0% | +1.9 pts |
| 3rd and medium | 8.2% | 10.3% | +2.1 pts |
| 3rd and long | 9.8% | 11.2% | +1.4 pts |
| 4th down | 6.2% | 8.1% | +1.9 pts |
Every down rose between 2014–16 and 2023–25. Third down went from under 9 sacks in every 100 pass plays to over 10. Short, medium and long third downs all rose, by between about 1.4 and 2.1 sacks per 100 pass plays, so the climb didn’t come from one kind of third down.
Scrambles rose too, so fewer pass plays end in a throw
Didn’t quarterbacks just learn to run instead? A quarterback with a rusher in his face can take the sack or take off. If quarterbacks had simply learned to run, sacks would have dropped as scrambles rose. The chart shows both climbing together.
Since 2016, sacks and scrambles both rose, and about 12 pass plays in 100 now end in one or the other
Per 100 pass plays, counting scrambles as pass plays too (throws, sacks and scrambles). Regular season.
Show data
| Year | Sack or scramble | Sacks | Scrambles |
|---|---|---|---|
| 2006 | 9.4% | 6.5% | 3.0% |
| 2007 | 8.6% | 5.9% | 2.7% |
| 2008 | 8.7% | 5.8% | 3.0% |
| 2009 | 8.6% | 5.9% | 2.7% |
| 2010 | 9.3% | 5.9% | 3.4% |
| 2011 | 9.9% | 6.2% | 3.8% |
| 2012 | 9.5% | 6.0% | 3.5% |
| 2013 | 10.2% | 6.4% | 3.8% |
| 2014 | 9.7% | 6.1% | 3.6% |
| 2015 | 9.5% | 5.9% | 3.6% |
| 2016 | 8.9% | 5.6% | 3.3% |
| 2017 | 9.9% | 6.2% | 3.8% |
| 2018 | 10.5% | 6.5% | 4.0% |
| 2019 | 10.4% | 6.4% | 4.0% |
| 2020 | 10.0% | 5.7% | 4.3% |
| 2021 | 10.4% | 6.0% | 4.4% |
| 2022 | 10.9% | 6.4% | 4.5% |
| 2023 | 11.8% | 6.8% | 5.0% |
| 2024 | 11.8% | 6.5% | 5.3% |
| 2025 | 12.0% | 6.5% | 5.5% |
- 2016: 2016: lowest sack rate
This section counts a different way, so its numbers won’t match the rest of the story. Every throw, sack and scramble counts as a pass play here, which makes each sack rate a little lower than elsewhere. On this count, sacks went from about 5.6 to 6.5 in every 100 pass plays between 2016 and 2025, and scrambles from about 3.3 to 5.5.
Plays that ended in one or the other went from nearly 9 to about 12 in 100, the most since this count starts in 2006 (9.4 that year). In 2025, about one in eight pass plays ended without a throw. Across 715 single seasons by a quarterback since 2006, the ones who scrambled a lot also tended to get sacked a lot.
The scramble count starts in 2006 because that is when the logs become reliable on scrambles. From 2006 on, every play the logs mark as a scramble says "scrambles" in its written description. Before 2006, at most about two in three do, so the older scramble counts can’t be trusted.
Teams that trail get sacked most, and each sack costs about a foot more
In every stretch of the logs, teams that were behind got sacked most. In 2019–25, an offense down by nine or more was sacked 7.4 times in every 100 pass plays. In close games it was 6.5, and with a lead of nine or more, 6.1.
Each sack also costs a little more ground. In 1999–2005 the average sack lost about 6.3 yards; in 2019–25 it lost 6.7, about a foot more. The average peaked at 6.9 yards in 2021 and was 6.5 in 2025. Some offenses take far more of those sacks than others, year after year.
Sacks follow the offense more than the defense
When sacks pile up, fans credit the pass rush, and so does the box score, which puts each sack in a defender’s column. Follow teams into the next season, though, and the sacks stick to the offense. That tells you which teams get sacked. It can’t tell you why the whole league rose.
Start with the offenses. The 2023 Giants were sacked on 14.2 of every 100 pass plays. The next year they were at 7.6, still above the 1999–2025 league rate of 6.5. The 2002 Texans, at 14.5, had the only higher team rate since 1999 in our data. A year later they were at 7.6.
Now the defenses. The 2022 Eagles had 70 sacks, 11.3 per 100 of their opponents’ pass plays, the second-highest rate of any defense since 1999 in our data. The next season they were at 6.2, below the league rate. The 2008 Chiefs managed 10 sacks, 1.9 per 100, the lowest. A year later they were at 4.3.
Four teams can’t settle it, so put one number on all of them, from 0 to 1. Suppose the Giants had been sacked on 14.2 of every 100 pass plays again in 2024, and every other offense had matched its own year before just as exactly. That would score 1. If last year told you nothing about this one, the number would be 0. Below 0 would mean teams tended to flip, high one year and low the next.
An offense’s sack rate repeats from one season to the next far more than a defense’s
Bar height: 1 = every team’s sack rate came out exactly where it was the year before; 0 = last year told you nothing. 1999–2025. Quarterback bars use quarterbacks with at least 200 pass plays in both seasons; starter bars use teams whose main quarterback had 200; the every-offense and defense bars use every team.
An offense that keeps its starter carries over 0.53, and so does a quarterback who stays put (0.51). A defense carries over 0.22, and a team that changes its starter 0.12.
On that scale, offenses that kept their starting quarterback score 0.53 and defenses 0.22. Allowing for luck, the offense figure could be anywhere from 0.47 to 0.59 and the defense figure from 0.16 to 0.29. Those don’t overlap, so luck can’t explain the gap. Count every offense, new quarterback or not, and the figure is 0.40, still clear of the defenses.
Some of every season is bounces: a tipped pass, a slipped block, a quarterback who holds the ball a beat too long. Make every offense exactly league average, and those bounces alone would still scatter their season sack rates about 1.0 per 100 pass plays either side of the average. Real offenses scatter about 2.0, about twice as far. Twice the scatter works out to four times the room to vary, so about a quarter of the differences between offenses is the kind luck makes. Defenses scatter 1.4, somewhat more than bounces would on their own, so more than half of their differences could be luck.
An offense is a line, a play-caller and a quarterback. The quarterback is the easiest of the three to follow, because when he changes teams the logs follow him by name.
Inside the offense, the quarterback carries some of it
Some quarterbacks get dropped season after season. In 2002 Houston’s David Carr was sacked on 14.6 of every 100 pass plays, and no quarterback since 1999 has been sacked more times in a season than his 76. Three seasons later he was sacked 68 times (13.9 per 100). The highest rate in our data belongs to Tommy DeVito of the 85-sack Giants, sacked 37 times in 215 pass plays in 2023, 17.2 per 100.
At the other end is Peyton Manning. In 2009 Indianapolis’s quarterback was sacked 10 times in 581 pass plays, 1.7 per 100, the lowest of 969 single seasons by a quarterback with at least 200 pass plays. His career rate since 1999, 3.1, is the lowest of the 55 quarterbacks with 3,000 or more.
| Quarterback | Pass plays | Sacks per 100 |
|---|---|---|
| Peyton Manning | 9,054 | 3.1 |
| Drew Brees | 10,922 | 3.8 |
| Brett Favre | 6,680 | 4.4 |
| Patrick Mahomes | 4,885 | 4.5 |
| Tom Brady | 12,589 | 4.5 |
| Michael Vick | 3,521 | 9.0 |
| Drew Bledsoe | 3,619 | 8.6 |
| Russell Wilson | 6,662 | 8.6 |
Continues below
| Quarterback | Pass plays | Sacks per 100 |
|---|---|---|
| Daunte Culpepper | 3,479 | 8.5 |
| Geno Smith | 3,368 | 8.2 |
At the far ends of the table, Michael Vick was sacked almost three times as often as Peyton Manning, 9.0 sacks per 100 pass plays against 3.1. Over 3,000 pass plays, bounces alone would move a career rate by only about 0.45.
Now follow quarterbacks who change teams. Baker Mayfield was sacked on 9.7 of every 100 pass plays in 2022, with Carolina and the Rams, and on 6.6 with Tampa Bay the next year. In 2025, Sam Darnold arrived in Seattle from Minnesota and went from 8.1 to 5.4. The quarterback who left Seattle, Geno Smith, went from 8.0 to 11.0 in Las Vegas.
Those three moves point in different directions, so look at all 87 moves since 1999. On the same 0-to-1 scale, a quarterback’s rate with his new team scored 0.27. When a team changed starters, its own rate scored only 0.12. So a quarterback’s old rate tells you a bit more about his new team than the team’s old rate does. The two are close enough that the lean could be luck. A quarterback who stays put scores 0.51, against the movers’ 0.27. Movers’ seasons are a little shorter, which explains only a small part of that gap.
On most teams the line, the play calls and the quarterback change together, and this data has no way to pull them apart. Franchises outlast all three.
Whole franchises differ, but the gaps mostly didn’t carry over
Indianapolis has the lowest franchise rate since 1999, 4.7 sacks per 100 pass plays. Cleveland has the highest, 7.5. Split the years, though, and Indianapolis went from 3.6 in 1999–2012 to 5.9 in 2013–25.
Across all 32 franchises, 1999–2025 rates spread about three times as far as bounces alone would, about 0.7 sacks per 100 pass plays against 0.2. So the differences are real. They mostly didn’t carry over. On the same 0-to-1 scale, a franchise’s 1999–2012 rate scores 0.17 against its 2013–25 rate. With only 32 franchises to go on, the true figure could be anywhere from −0.19 to 0.49, so the data can’t say whether the gaps last.
| Franchise | 1999–2025 | 1999–2012 | 2013–2025 |
|---|---|---|---|
| Indianapolis | 4.7 | 3.6 | 5.9 |
| New Orleans | 5.1 | 5.1 | 5.2 |
| Green Bay | 5.6 | 5.2 | 6.0 |
| Denver | 5.8 | 5.5 | 6.0 |
| Cleveland | 7.5 | 7.3 | 7.6 |
| Seattle | 7.5 | 7.0 | 7.9 |
| Minnesota | 7.3 | 7.6 | 7.1 |
| Houston | 7.3 | 7.5 | 7.2 |
The table is a record of where each franchise has been. It says little about where one is going. Since 1999, sacks have stuck to offenses, and a little to quarterbacks, more than to the pass rushers who got the credit.
Big nights make it feel like sacks are everywhere
A record or a ten-sack night is what makes the sack feel back. Nights like that turn up all through the play-by-play logs, from 2001 to 2025.
In Week 18 of 2025, Cleveland’s Myles Garrett sacked Cincinnati’s Joe Burrow for his 23rd sack of the season. The Browns’ site called it a new NFL record, past the 22.5 of Michael Strahan in 2001 and T.J. Watt in 2021. "I want to be remembered as one of the very best at the position," Garrett said. Across the league that season, about 18,600 pass plays produced 6.9 sacks in every 100.
Donovan McNabb was sacked twelve times by the Giants on Sep 30, 2007, the most by one quarterback in a game in our play-by-play data. The league rate that season was 6.1, below 1999’s 7.0. David Carr’s 76 sacks in 2002, which NBC Sports Boston lists as the NFL single-season record, came in a 6.4 season.
Daniel Jones’s ten at MetLife was one of seven times since 1999 that a quarterback was sacked exactly ten times in a game. The Giants’ eleventh sack that night came on a trick play. Jones pitched the ball back to receiver Parris Campbell, who looked to throw, and Devon Witherspoon brought him down, the play-by-play shows.
Garrett’s 23 were a handful of the roughly 1,300 sacks the whole league recorded that season, and every one of those counts toward the league rate.
The MetLife season was the top of a climb back to 1999
Geno Smith was the other quarterback at MetLife that night. The play-by-play has him sacked twice, and he came back for the second half after a tweaked knee and X-rays, FOX 13 reported. On Nov 23, 2025, Cleveland sacked him ten times in Las Vegas, three of them by Garrett.
Daniel Jones’s ten sacks that Monday night came in 2023, which finished at 7.2 sacks in every 100 pass plays, the high mark in the play-by-play logs. Two seasons later, in 2025, the league was at 6.9, about the 7.0 of 1999.
How we measured it
- Sack rate is sacks divided by pass attempts plus sacks, regular season. Scrambles are left out so the number matches the older league totals, which have no scramble count. Rates for several seasons add up the sacks and pass plays first, then divide.
- 1963–1998: league season totals from Pro-Football-Reference. A sack is offensive plays minus pass attempts minus rush attempts. The 1963–69 rows cover NFL teams only, not the AFL. The strike-shortened 1982 season (nine games) and 1987, with its replacement games, are in the series.
- 1999–2025: nflverse play-by-play, 519,753 pass plays including scrambles and 32,399 sacks. The 1999 file has 247 games and the 2000 file 246; a full season then had 248. Team and passer counts can differ by a sack or two from official counts.
- Where the books meet: in 1999 the league totals give 6.94 and the play-by-play 6.98. Over 1999–2025, Pro-Football-Reference’s per-game figures match the play-by-play to between −0.32 and +0.21 every season. Pro-Football-Reference counts spiked balls as pass attempts and the play-by-play leaves them out, which puts the play-by-play rate higher by at most about 0.04 a season. So this story compares the two books only where the gap is far bigger than the per-season disagreement between them (−0.32 to +0.21), and never ranks close values across them (1994 vs 2016, for one).
- The NFL began keeping track of individual sacks only in 1982, as Wikipedia’s list of career sack leaders notes. This story uses team and passer totals.
Season-to-season moves. Within the league totals (1963–98), the typical move was 0.48 sacks per 100 pass plays (35 moves). Within the play-by-play (1999–2025), it was 0.27 (26 moves). The 1998-to-1999 step crosses books, so it is left out of both. The 1978 rule change is timing only. Wikipedia’s summary of that season says, "The pass blocking rules were extended to permit extended arms and open hands." The rate fell after it and rose again by 1984.
Carry-over is the season-to-season correlation of a sack rate, with 95% ranges. "Same starter" means the team’s most-used passer was the same in both seasons. The bounce benchmark is the spread pure chance would give at the 1999–2025 league rate over each team’s or franchise’s pass plays.
Quarterback tools
Live Edgehalla tools on the same subject.
Show the work
- Tests written down
- Not in advance (exploratory)
- Results computed
League totals 1963–1998 (36 seasons, Pro-Football-Reference) and every regular-season dropback 1999–2025 in nflverse play-by-play (519,753 dropbacks, 32,399 sacks, 969 passer-seasons with 200+ attempts plus sacks, 861 team-seasons, 17,140 passer-games). Multi-season rates are pooled (sacks / attempts plus sacks). Repeatability is season-to-season Pearson r with 95% ranges. The luck benchmark compares the spread of team, franchise and career rates with the spread pure chance would give at the league rate (6.5%). Franchise persistence compares 1999–2012 with 2013–25. Descriptive; no model and no betting test.
| Test | Result | Verdict |
|---|---|---|
| PFR and nflverse agree on the overlap | 1999: 6.94% vs 6.98%; per-game gap -0.32 to +0.21 points, 1999–2025 | Holds |
| League-total sack rate fell every decade, 1960s to 1990s | 8.60 / 8.18 / 7.53 / 6.83% (1963–69, 1970s, 1980s, 1990–98, pooled); single seasons 1984 8.40%, 1994 5.86%, 1997 7.37% | Holds, with large single-season bounces |
More tests below
| Test | Result | Verdict |
|---|---|---|
| Back to the 1999 level (headline band ±0.3) | 2023–25 pooled 6.992% vs 1999 6.978% (gap +0.014); 1999–2001 6.946% | Holds |
| The 2016–25 climb is bigger than a typical play-by-play season move | 2014–16 6.084% to 2023–25 6.992% (+0.908); median season move within play-by-play 0.271 (n 26), ratio 3.35 | Holds |
| Scrambles replaced sacks | Both rose after 2016; passer-level r 0.44 (n 715) | Rejected |
| Offense sack rate repeats more than defense | Team, same starter r 0.53 (0.47 to 0.59) vs defense r 0.22 (0.16 to 0.29) | Holds |
| Team-season rates spread wider than chance | Offense SD 1.97 vs chance 1.03 (luck share 27%); defense 1.36 vs 1.03 (57%) | Offense yes, defense barely |
Continues below
| Test | Result | Verdict |
|---|---|---|
| Sack rate follows the quarterback | QB new team r 0.27 (0.06 to 0.45) vs team new QB r 0.12 (0.01 to 0.23) | Leans, not clear |
| Franchise sack rates persist | Spread 0.66 vs chance 0.20; 1999–2012 vs 2013–25 r 0.17 (-0.19 to 0.49, n 32) | Real gaps, do not persist |
| Sacks cost more yards now | 6.32 yards (1999–2005) vs 6.67 (2019–25); r 0.73 with season | Small rise |
Frequently asked questions
Are NFL sacks going up?
Yes, since 2016. On the play-by-play count the rate was 5.8 sacks in every 100 pass plays that season and 6.9 in 2025, about the same as 1999 (7.0). The league’s season totals had it above 8 in the 1960s and ’70s.
What is the sack rate?
Sacks for every 100 pass plays, where a pass play is a throw or a sack. Scrambles are left out so the number matches the older league totals, which have no scramble count.
Are sacks the quarterback’s fault or the offensive line’s?
A team’s sack rate carries into the next season far more on offense than on defense. Inside the offense, the rate leans toward the quarterback when he changes teams, but that lean could be luck, and the data can’t cleanly separate line, scheme and passer.
Did scrambles replace sacks?
No. Since 2016 both have risen, and quarterbacks who scramble a lot also tend to get sacked a lot.
Which quarterback was sacked the most in one season?
David Carr, sacked 76 times for Houston in 2002, which NBC Sports Boston lists as the NFL record. Since 1999 the highest sack rate in a season with 200 or more pass plays is Tommy DeVito’s 17.2 per 100 for the 2023 Giants.
Which NFL team allowed the most sacks since 1999?
In our 1999–2025 play-by-play, the 2023 Giants, sacked 85 times. The 2002 Texans had the highest rate (14.5 per 100 pass plays).
More in The long arc
Sources and method
- nflverse play-by-play, rosters, combine and schedules (CC-BY 4.0)
- Pro-Football-Reference: NFL season-by-season team offense (league totals, 1963–1998)
- Associated Press via News 12, "Daniel Jones sacked 10 times as Giants show little in 24-3 loss to the Seattle Seahawks" (Oct 3, 2023)
- FOX 13 Seattle, "Devon Witherspoon pick-six, team record 11 sacks carry Seahawks to 24-3 win over Giants" (Oct 2, 2023)
- NBC Sports Boston, "Which QB has been sacked the most in a single NFL season?" (Sep 8, 2022)
- Cleveland Browns, "Myles Garrett breaks NFL single-season sack record" (Jan 4, 2026)
- Cleveland Browns, "Browns defense registers 10 sacks in win over Raiders: By the Numbers" (Nov 2025)
- Wikipedia, "1978 NFL season" (rule changes)
- Wikipedia, "List of NFL career sacks leaders" (sacks tracked since 1982)
Research and analysis, not betting advice. Past results do not guarantee future returns; bet only what you can afford to lose.