How Football ChangedThe long arc
Is the NFL Backing Away From the Shotgun?
In 2007 NFL offenses ran a little over one play in four from the shotgun. By 2023 it was about seven in ten. Then it fell two seasons running, and 21 of 32 teams used it less. A dip in 2017 lasted one season, so it’s too soon to say this one will stick.
By Edgehalla ResearchPublished 10 min read
In 2016, Chip Kelly’s San Francisco 49ers took 98.4 percent of their plays with the quarterback in the shotgun. No team since 1999, as far back as the play-by-play goes, has come closer to never lining up under center.
They went 2–14, and the 49ers fired Kelly after one season. The next coach in San Francisco was Kyle Shanahan.
In 2025, Kyle Shanahan’s 49ers used the shotgun on about 54 plays in 100. Since 2023 the rest of the league has been edging the same way, more slowly. Nobody knows yet whether it will keep going.
On this page
The shotgun spent 16 years taking over
Kelly’s team was the extreme, but the whole league was heading his way. In the shotgun, the quarterback stands a few yards behind the center and catches a long snap. Under center, he takes the ball straight from the center’s hands.
In 2007, NFL offenses ran a little over one play in four from the shotgun. By 2013 it was about 59 in 100, more than double. The 2013 jump alone, about 11½ points, is the biggest one-season rise in our data. By 2023 it was about 72 in 100, the peak. In 2007 teams were in the shotgun on about 56 of every 100 third-down plays but only about 16 of every 100 first downs. By 2023 they used it on about 63 of every 100 first downs.
The shotgun rose to 72.2% of NFL plays in 2023, then fell two seasons in a row, in close games too
Share of regular-season runs and passes (no kneel-downs, spikes or two-point tries) with the quarterback in the shotgun. The grey dashed line, before 2007, undercounts the shotgun (see the box below).
Show data
| Year | All plays, undercounted before 2007 | All plays | Close games |
|---|---|---|---|
| 1999 | 11.9% | – | – |
| 2000 | 12.8% | – | – |
| 2001 | 13.9% | – | – |
| 2002 | 13.8% | – | – |
| 2003 | 14.2% | – | – |
| 2004 | 13.0% | – | – |
| 2005 | 15.5% | – | – |
| 2006 | 19.4% | – | – |
| 2007 | 26.9% | 26.9% | 21.3% |
| 2008 | – | 32.1% | 27.0% |
| 2009 | – | 36.7% | 29.6% |
| 2010 | – | 37.7% | 31.6% |
| 2011 | – | 40.4% | 33.9% |
| 2012 | – | 47.3% | 41.2% |
| 2013 | – | 58.8% | 54.3% |
| 2014 | – | 60.7% | 55.8% |
| 2015 | – | 62.5% | 57.7% |
| 2016 | – | 64.2% | 59.5% |
| 2017 | – | 59.3% | 54.2% |
| 2018 | – | 63.9% | 59.2% |
| 2019 | – | 64.7% | 60.8% |
| 2020 | – | 66.1% | 61.4% |
| 2021 | – | 66.2% | 61.2% |
| 2022 | – | 67.9% | 64.1% |
| 2023 | – | 72.2% | 69.3% |
| 2024 | – | 70.6% | 67.3% |
| 2025 | – | 66.2% | 62.7% |
- 2017: 2017 dip: 59.3%
- 2023: Peak: 72.2%
Follow the solid line from 2007. It goes up almost every year, dips once in 2017, then steps down twice at the end. The lower line, close games, does the same. The shotgun share rose in 21 of the 24 seasons from 2000 to 2023, even counting the undercounted early years. It fell only in 2002, 2004 and 2017, and never twice in a row.
Then it fell two seasons in a row, even in close games
Then the line turned. In 2024 the shotgun share slipped to about 71 plays in 100. In 2025 it fell to about two in three, where it stood in 2020 and 2021. That is about 6 fewer shotgun plays in every 100 than in 2023, a 6-point drop. It was the first time in our data, which starts in 1999, that it fell two seasons running.
A team down 17 in the fourth quarter has to throw, and most NFL passes come from the shotgun. So the drop could just mean fewer teams were chasing games. The close games rule that out. In games still in doubt, where each team had at least a 1-in-5 chance to win and more than two minutes were left in the half, the shotgun fell from about 69 plays in 100 in 2023 to about 63 in 2025. That is the lower line on the chart above. On first and second down in those games it fell further, from about 64 to about 56.
Runs were the last to join the shotgun and the most to leave it
Runs have led the way back out. They were also the last ones in. Passing moved into the shotgun first. Count every time a quarterback dropped back to pass, sacks and scrambles included, and by 2013 about three of every four came from the shotgun. Runs came later. In 2007 only about 6 of every 100 designed runs, meaning handoffs and planned quarterback runs, started from the shotgun. That reached about 32 in 2013 and about half in 2023.
After 2023, the shotgun share of runs fell about twice as far as the share of passes
Share from the shotgun. Dropbacks = passes, sacks and scrambles. Designed runs = handoffs and planned quarterback runs. Dashed line: every first-down play, run or pass.
Show data
| Year | Dropbacks | Designed runs | 1st down, all plays |
|---|---|---|---|
| 2007 | 41.6% | 6.2% | 16.3% |
| 2008 | 48.0% | 10.5% | 20.3% |
| 2009 | 54.6% | 11.6% | 23.3% |
| 2010 | 55.5% | 11.7% | 24.4% |
| 2011 | 58.3% | 13.9% | 27.3% |
| 2012 | 65.8% | 19.1% | 33.4% |
| 2013 | 75.6% | 32.1% | 47.0% |
| 2014 | 78.0% | 33.6% | 48.6% |
| 2015 | 79.9% | 34.1% | 50.3% |
| 2016 | 80.6% | 37.5% | 52.1% |
| 2017 | 77.9% | 31.0% | 44.8% |
| 2018 | 80.0% | 37.7% | 51.9% |
| 2019 | 80.5% | 38.9% | 52.3% |
| 2020 | 81.2% | 42.1% | 54.5% |
| 2021 | 82.4% | 40.8% | 54.3% |
| 2022 | 83.2% | 44.9% | 56.8% |
| 2023 | 86.1% | 50.3% | 62.8% |
| 2024 | 84.8% | 48.9% | 59.2% |
| 2025 | 82.2% | 41.9% | 53.0% |
On the chart, the runs line climbs for 16 years and then bends down hardest. Since 2023, shotgun runs fell about 8 points, to about 42 in 100. Shotgun dropbacks fell about 4, to about 82.
First down moved most. Teams were in the shotgun on about 63 of every 100 first-down plays in 2023 and about 53 in 2025. Third down hardly moved, from about 90 to about 88. When a team has to throw, it still reaches for the gun.
Twenty-one teams backed off, from Seattle to Philadelphia
Between 2023 and 2025, 21 of the 32 teams used the shotgun less and 11 used it more. The middle team fell about 4 points. Sean McVay’s Rams used it least in 2025, on about 41 plays in 100. Dan Quinn’s Commanders used it most, on almost 9 in 10.
Most offenses took snaps under center more often in 2025 than in 2023
Shotgun share of each team’s plays, 2023 vs 2025. Dots below the line used the shotgun less in 2025.
Shotgun share, 2025
Shotgun share, 2023
Show data
| Shotgun share, 2023 | Shotgun share, 2025 | |
|---|---|---|
| SEA | 72% | 46% |
| CHI | 78% | 52% |
| BUF | 75% | 50% |
| LA | 62% | 41% |
| JAX | 81% | 61% |
| BAL | 85% | 65% |
| IND | 91% | 72% |
| PHI | 94% | 78% |
| CAR | 83% | 67% |
| NE | 66% | 55% |
| LAC | 81% | 72% |
| GB | 73% | 63% |
| MIA | 81% | 72% |
| TB | 74% | 66% |
| SF | 61% | 54% |
| DET | 56% | 52% |
| PIT | 70% | 67% |
| CIN | 86% | 82% |
| KC | 84% | 81% |
| DAL | 64% | 63% |
| ARI | 72% | 72% |
| MIN | 55% | 56% |
| HOU | 61% | 62% |
| CLE | 66% | 68% |
| NYG | 74% | 77% |
| ATL | 78% | 81% |
| WAS | 82% | 89% |
| DEN | 58% | 66% |
| TEN | 64% | 73% |
| LV | 56% | 66% |
| NYJ | 64% | 74% |
| NO | 63% | 79% |
- No change: from (40%, 40%) to (95%, 95%)
Every dot below the diagonal is a team that used the shotgun less in 2025 than in 2023. The table below has the six biggest falls and the three biggest rises.
| Team | Coach | 2023 | 2025 | Change |
|---|---|---|---|---|
| Seahawks | Mike Macdonald | 72.2% | 46.4% | -25.8 |
Continues below
| Team | Coach | 2023 | 2025 | Change |
|---|---|---|---|---|
| Bears | Ben Johnson | 77.5% | 52.1% | -25.4 |
| Bills | Sean McDermott | 74.5% | 50.2% | -24.3 |
| Rams | Sean McVay | 61.9% | 41.0% | -20.9 |
| Jaguars | Liam Coen | 81.3% | 60.9% | -20.4 |
| Ravens | John Harbaugh | 84.8% | 65.0% | -19.8 |
| Saints | Kellen Moore | 63.3% | 79.0% | +15.7 |
| Jets | Aaron Glenn | 63.5% | 73.7% | +10.2 |
| Raiders | Pete Carroll | 56.3% | 66.3% | +10.0 |
Three of the six biggest falls came with a head coach hired since 2023. Mike Macdonald’s Seahawks fell about 26 points, Ben Johnson’s Bears fell about 25 points and Liam Coen’s Jaguars fell about 20 points. The other three came under the same coach as in 2023. Sean McDermott’s Bills fell about 24, Sean McVay’s Rams fell about 21 and John Harbaugh’s Ravens fell about 20. McDermott and Harbaugh were both fired after the 2025 season. The three biggest risers, the Saints, Jets and Raiders, also had head coaches hired since 2023. But most of the teams that fell kept the same coach, 14 of the 21. These numbers can’t show whether coaches caused any of it.
Even Philadelphia backed off. The Eagles led the league in shotgun share in four seasons (tied with Detroit for the most), under Andy Reid in 2009, Chip Kelly in 2013 and 2015 and Nick Sirianni in 2023. Kelly’s 2015 Eagles ran about 94 plays in 100 from the shotgun. Nick Sirianni’s 2023 Eagles ran about 94, the league high in the peak year. By 2025 his Eagles were down to about 78, a 16-point fall under the same coach.
The Shanahan–McVay offenses fell more than the rest of the league, mostly because of the Rams
The obvious suspects are five head coaches from one coaching family who run NFL offenses: Kyle Shanahan, Sean McVay, Matt LaFleur, Kevin O’Connell and Mike McDaniel. We fixed that list of five before looking at any numbers. It leaves out coordinators and other head coaches with ties to the family, such as Cincinnati’s Zac Taylor, a former McVay assistant whose Bengals stayed deep in the shotgun. So the family’s reach could be wider.
The Shanahan–McVay offenses were always under center more, and the gap widened in 2025
Shotgun share of plays: offenses whose head coach is Shanahan, McVay, LaFleur, O’Connell or McDaniel, against every other team.
Show data
| Year | Every other team | Shanahan–McVay offenses |
|---|---|---|
| 2017 | 60.3% | 44.5% |
| 2018 | 65.4% | 42.4% |
| 2019 | 66.2% | 50.0% |
| 2020 | 67.5% | 53.2% |
| 2021 | 66.8% | 60.3% |
| 2022 | 69.1% | 60.9% |
| 2023 | 73.3% | 66.3% |
| 2024 | 71.7% | 64.3% |
| 2025 | 67.8% | 56.8% |
As a group, those offenses used the shotgun less than the rest of the league in every season from 2017 to 2025. In 2017, McVay’s Rams and Shanahan’s 49ers ran a little under 45 plays in 100 from the shotgun. Everyone else ran about 60. The gap held at about seven to eight points in 2022–24, then widened to about 11 in 2025. On the chart, that is where the family’s line drops away.
After 2023 they also cut back more than the rest of the league, mostly because of the Rams. As a group their shotgun share fell about 9½ points. Everyone else’s fell about 5½. McVay’s Rams alone fell about 21. Team by team, though, the Seahawks, Bears and Bills each fell more than any of the five.
The family is also small. Those five offenses ran about 5,000 of the league’s 33,000 plays in 2025. That is about 15 plays in every 100. Spread across the whole league, their 9½-point drop comes to about 1.4 points. The other 27 teams ran the other 28,000 plays. Their 5½-point drop comes to about 4.6 points. Together those make up the league’s fall of about 6 points. So the five supply less than a quarter of it, and the rest of the league more than three quarters.
Under center, teams mostly run, and when they pass they usually fake a handoff first
FTN, a charting company, keeps its own count. It logs where the quarterback stands on nearly every snap. The league’s official play-by-play, its written log of every play, marks shotgun snaps separately. The two agree. In every season from 2022 to 2025, at least 99 of every 100 snaps FTN charted as shotgun are marked shotgun in the play-by-play too. By FTN’s count, quarterbacks lined up under center on about 27 plays in 100 in 2023 and about one in three in 2025.
Quarterbacks lined up under center on one play in three in 2025, the most in the four seasons of FTN charting in our data (2022–25)
Share of plays by where the quarterback lined up, from FTN’s charting.
Under-center snaps fell to 27.3% of plays in 2023 and rose to 33.6% in 2025, by FTN’s charting.
- 2022
- 2023
- 2024
- 2025
The one odd bar is the pistol, a shotgun cousin with a running back lined up right behind the quarterback. It jumps to about 9 in 100 in 2024, then falls back to 5. That looks like a change in how FTN charted it, though we can’t confirm it. Shotgun and pistol together came to about 70 plays in 100 in 2024, within a fraction of a point of the play-by-play, which counts the pistol as shotgun.
Back on the field, most under-center snaps are still runs, about 7 in 10 in 2025. Passes from under center are growing too. They were about 13 of every 100 dropbacks in 2023 and about 17 in 2025.
When a quarterback under center does drop back, he usually fakes a handoff first. He turns, pushes the ball toward the running back’s belly, pulls it back and looks downfield. FTN charted that fake on about 8 in 10 under-center dropbacks in 2025. From the shotgun it was about 1 in 10. ESPN’s Bill Barnwell counted play-action on more than 3 in 4 under-center dropbacks in Week 1 of 2026.
Under-center passes have gained more for years, long before the turn
Did teams go back because passes from under center work better? Under center has gained more per pass since 2013. Teams kept moving to the shotgun for ten more years anyway. So that edge alone can’t explain a turn that came in 2024, and these numbers can’t tell us what did. Analysts grade each pass by what it does for the drive. A 9-yard catch on third-and-8 keeps a drive alive. The same catch on third-and-12 leaves the offense facing fourth down. The first catch did more for the offense’s chances of scoring, so it scores higher in expected points, the yardstick on the chart below.
On early downs, passes from under center have out-gained passes from the shotgun every season since 2013
Expected points per dropback on 1st and 2nd down, by where the quarterback lined up.
EPA
Show data
| Year | Under center | Shotgun |
|---|---|---|
| 2007 | 0.06 EPA | 0.03 EPA |
| 2008 | 0.06 EPA | 0.06 EPA |
| 2009 | 0.05 EPA | 0.07 EPA |
| 2010 | 0.09 EPA | 0.06 EPA |
| 2011 | 0.08 EPA | 0.07 EPA |
| 2012 | 0.07 EPA | 0.10 EPA |
| 2013 | 0.09 EPA | 0.08 EPA |
| 2014 | 0.11 EPA | 0.07 EPA |
| 2015 | 0.17 EPA | 0.09 EPA |
| 2016 | 0.17 EPA | 0.08 EPA |
| 2017 | 0.12 EPA | 0.05 EPA |
| 2018 | 0.19 EPA | 0.08 EPA |
| 2019 | 0.14 EPA | 0.06 EPA |
| 2020 | 0.18 EPA | 0.10 EPA |
| 2021 | 0.12 EPA | 0.07 EPA |
| 2022 | 0.12 EPA | 0.05 EPA |
| 2023 | 0.06 EPA | 0.03 EPA |
| 2024 | 0.16 EPA | 0.07 EPA |
| 2025 | 0.14 EPA | 0.06 EPA |
On first and second down in 2025, a dropback from under center was worth about 0.14 expected points. A dropback from the shotgun was worth about 0.06. Under center has come out ahead every season since 2013.
Teams also choose when to line up under center, and most of their passes from there are play-action. So the two lines on the chart aren’t a fair fight.
Two seasons is the longest fall yet, and 2017 shows why that isn’t proof
So is the NFL backing away from the shotgun? Yes, for now. But in 2017 the shotgun share fell about 5 points in one season, more than it fell in 2025 alone, and in 2018 it was back to about 64 in 100.
Each season has about 33,000 plays. With that many plays, chance alone can’t move the league’s share anywhere near 6 points, so the drop itself is real. But 2017 shows that a real drop can still reverse.
Looking two seasons at a time gives the same picture. Take every pair of seasons two years apart, such as 2006 to 2008, and average each team’s change. The average team cut its shotgun use in only three of those 18 overlapping stretches: 2015–17 and 2016–18, both around the 2017 dip, and 2023–25. The 2023–25 drop, about 6 points, is the biggest.
In 18 overlapping two-season stretches since 2006, the average team cut its shotgun use only three times, and 2023–25 was the biggest cut
Average change in a team’s shotgun share over each two-season stretch, in percentage points.
Of the 18 two-season stretches from 2006–08 to 2023–25, the average team’s shotgun share fell in 3 (2015–17, 2016–18 and 2023–25); the 2023–25 change, -5.9 points, was the largest fall.
Single teams bounce around much more. From 2023 to 2025 the Seahawks fell 26 points and the Saints rose 16. A typical team landed about 11½ points off the league’s average move, one way or the other. That is far more than luck alone moves a team, about 3 points, so single teams really do change. But one team’s swing can come from a new quarterback, injuries or how its games went. The league’s 6-point fall across about 33,000 plays is far too big for luck, and it showed up in most teams (21 of 32).
Kelly’s 49ers ran 98 of every 100 plays from the shotgun. Shanahan’s ran 54. For two seasons the rest of the league has been walking toward Shanahan’s end of that line. In Week 1 of 2026, Barnwell counted offenses under center at the highest rate since 2012, “back where we were in 2013.” Our numbers stop at 2025, and his count is one week. In our count, 2013 was the first season the shotgun ran more than half of the NFL’s plays.
Quarterback tools
Live Edgehalla tools on the same subject.
Show the work
- Tests written down
- Not in advance (exploratory)
- Results computed
Every regular-season scrimmage play (runs and passes; kneels, spikes and two-point tries left out) in nflverse play-by-play, 1999–2025: 867,908 plays, 861 team-seasons. The shotgun flag is parsed from the official play description. FTN charting matched to 133,636 plays from 2022–2025. Close games: win probability 20–80% for the offense with more than two minutes left in the half. League shares count every play; the two-season windows average each team’s change, so each team counts once (2023–25: 5.9 points by team, 6.1 by play, from unrounded shares). The Shanahan–McVay share of the league’s fall is an approximate split that holds each group’s share of plays at its 2025 level. Expected points are nflverse EPA per dropback on 1st and 2nd down. Descriptive only; no model and no holdout.
| Test | Result | Verdict |
|---|---|---|
| Shotgun flag against the play text | Agrees on every play, every season | Same source |
More tests below
| Test | Result | Verdict |
|---|---|---|
| Shotgun flag against FTN charting (2022–2025) | 99.0–99.5% of FTN shotgun snaps flagged; 0.4–1.0% of under-center snaps | Agree |
| Close games only (win probability 20–80%, more than two minutes left in the half) | 69.3% (2023) vs 62.7% (2025); 1st and 2nd down 63.6% vs 55.8% | Drop holds |
| Teams that fell, 2023 to 2025 | 21 of 32; median -4.0 points; team-to-team spread (sample SD) 11.5 points | League-wide |
| Share of the drop from the five Shanahan–McVay offenses | 1.4 of 6.1 points (group play shares held at 2025: 15.2% of plays × -9.5) | Minority |
| League fall 2023–2025: every play vs the average team | 6.1 play-weighted vs 5.9 team average | Same size |
Continues below
| Test | Result | Verdict |
|---|---|---|
| 1999 without teams coded at zero shotgun plays | 15.3% vs 11.9% | Early series too low |
Frequently asked questions
How often do NFL teams use the shotgun?
On 66.2% of regular-season runs and passes in 2025, down from a peak of 72.2% in 2023. In 2007 it was 26.9%.
Are NFL teams going under center more?
Yes, since 2023. 21 of 32 teams used the shotgun less in 2025 than in 2023. FTN charted 33.6% of 2025 plays with the quarterback under center, up from 27.3% in 2023. A one-season dip in 2017 reversed, so it is too soon to call it a lasting trend.
Which team uses the shotgun the least?
In 2025, the Rams: 41.0% of their plays, under Sean McVay. The Commanders used it most, on 88.5%.
Which NFL team used the shotgun the most?
Chip Kelly’s 2016 49ers, on 98.4% of plays, the highest team-season in play-by-play since 1999. John Harbaugh’s 2020 Ravens (96.8%) are next, and his Ravens ran 95.6–96.8% in every season from 2019 to 2021. Below those Ravens seasons, Chip Kelly’s 2015 Eagles (93.8%) and Nick Sirianni’s 2023 Eagles (93.7%) each led their season.
Did the Shanahan and McVay offenses cause the move back under center?
They were always under center more, and as a group they fell more than the rest of the league after 2023, mostly because of the Rams. The Seahawks, Bears and Bills each fell more than any one of them. They run about 15 of every 100 plays, so their own plays account for only about 1.4 points of the league’s 6.1-point drop directly. The split can’t measure other teams copying them.
Why go under center?
Mostly to run: 68.9% of 2025 under-center snaps were runs. When teams do pass from under center, 81.7% of the dropbacks use a play fake. On early downs, under-center dropbacks also gained more expected points per play than shotgun dropbacks, though teams choose when to use them, so that isn’t a fair test.
More in The long arc
Sources and method
- nflverse play-by-play, rosters, combine and schedules (CC-BY 4.0)
- FTN Data via nflverse: play charting 2022–2025 (quarterback alignment, play-action), licensed CC-BY-SA 4.0
- Charting data provided by FTNFantasy.com/data
- Licence for the FTN Data used here: Creative Commons Attribution-ShareAlike 4.0 (CC-BY-SA 4.0)
- nflfastR source, helper_add_nflscrapr_mutations.R (shotgun flag from the play text)
- 2016 San Francisco 49ers season (Wikipedia)
- “49ers officially hire Kyle Shanahan as new head coach”, CBS News Texas, Feb. 7, 2017
- Bill Barnwell, ESPN, “NFL offenses are going under center more often in 2026” (via ABC7), Sept. 17, 2026
- NFL.com, “Ravens fire HC John Harbaugh after his 18th season ends with missing playoffs”, Jan. 6, 2026
- CBS Sports, “Bills fire Sean McDermott”, Jan. 19, 2026
- Zac Taylor (Wikipedia): Rams assistant wide receivers coach 2017 and quarterbacks coach 2018 under Sean McVay; Bengals head coach from 2019
Research and analysis, not betting advice. Past results do not guarantee future returns; bet only what you can afford to lose.