How Football ChangedThe long arc
NFL Red-Zone Touchdowns Rose in Two Jumps, Then Stopped
In 1999–2005, 51% of NFL trips inside the opponent 20 ended in a touchdown. In 2021–25 it was 57%. The rise came in two steps, around 2002 and 2012, and the rate has been flat for 14 seasons. The trips themselves didn’t get easier. About four in ten of the extra touchdowns came on drives where a coach went for it on fourth down and made it.
By Edgehalla ResearchPublished 17 min read
On Oct. 6, 2013, Peyton Manning’s Denver Broncos beat the Dallas Cowboys 51–48 at AT&T Stadium. Seven times they reached the Dallas 20. Six times they scored a touchdown.
That was close to normal for them. Denver finished the 2013 regular season with 55 touchdowns on 70 red-zone trips, the most red-zone touchdowns of any team in play-by-play data back to 1999. The Broncos set the NFL record for points in a season, 606.
Around the league that year, 56 trips in 100 ended in a touchdown. That number had just reached a new level. It has hardly moved since.
On this page
Offenses finished more trips in two jumps, then stopped
Denver’s season came just after the league’s last jump. A red-zone trip is any drive that reaches the opponent’s 20-yard line or closer, and NFL offenses now finish more of those trips with touchdowns than they used to. Three questions follow from that: when it happened, why, and how much of any one team’s red-zone record is real.
Start with when. From 1999 to 2001, NFL offenses turned about 49% of their red-zone trips into touchdowns. From 2002 to 2011 they turned about 53%. Since 2012 the figure has been about 57% (leaving out 2020), and it hasn’t climbed again. Play-by-play data starts in 1999, so this story can’t say what came before.
NFL red-zone touchdowns jumped around 2002 and 2012, then held steady
Share of trips to the opponent 20 or closer that ended in a touchdown, each regular season. Shaded: how far each season’s rate could move on luck. Dashed: the average of each stretch between the jumps.
Show data
| Year | TD rate | TD rate (range) | 1999–01 level | 2002–11 level | 2012–25 level |
|---|---|---|---|---|---|
| 1999 | 47.9% | 45.3% to 50.4% | 49.2% | – | – |
| 2000 | 50.6% | 48.1% to 53.1% | – | – | – |
| 2001 | 49.0% | 46.5% to 51.5% | 49.2% | – | – |
| 2002 | 53.3% | 50.9% to 55.7% | – | 52.6% | – |
| 2003 | 52.1% | 49.6% to 54.6% | – | – | – |
| 2004 | 54.6% | 52.1% to 57.1% | – | – | – |
| 2005 | 51.4% | 49.0% to 53.9% | – | – | – |
| 2006 | 51.5% | 49.0% to 54.0% | – | – | – |
| 2007 | 52.2% | 49.7% to 54.6% | – | – | – |
| 2008 | 53.0% | 50.6% to 55.5% | – | – | – |
| 2009 | 52.7% | 50.2% to 55.1% | – | – | – |
| 2010 | 53.2% | 50.7% to 55.6% | – | – | – |
| 2011 | 52.3% | 49.9% to 54.7% | – | 52.6% | – |
| 2012 | 55.0% | 52.5% to 57.4% | – | – | 56.6% |
| 2013 | 56.0% | 53.6% to 58.4% | – | – | – |
| 2014 | 54.1% | 51.7% to 56.6% | – | – | – |
| 2015 | 56.6% | 54.2% to 59.0% | – | – | – |
| 2016 | 56.0% | 53.6% to 58.4% | – | – | – |
| 2017 | 53.9% | 51.4% to 56.4% | – | – | – |
| 2018 | 59.1% | 56.7% to 61.5% | – | – | – |
| 2019 | 57.4% | 55.0% to 59.7% | – | – | – |
| 2020 | 61.6% | 59.3% to 63.8% | – | – | – |
| 2021 | 59.1% | 56.8% to 61.3% | – | – | – |
| 2022 | 57.1% | 54.7% to 59.4% | – | – | – |
| 2023 | 55.2% | 52.8% to 57.6% | – | – | – |
| 2024 | 57.9% | 55.6% to 60.2% | – | – | – |
| 2025 | 57.7% | 55.3% to 60.0% | – | – | 56.6% |
- 2020: 2020 (outlier)
A season is only so many trips. Turn a handful of touchdowns into fumbles at the 1, or a handful of end-zone drops into touchdowns, and the rate moves. The shaded band shows how far luck like that could plausibly move each season. It is a 95% range: 19 times in 20, a season’s real level sits inside it.
Before asking why, check that the jumps are real, and not good and bad years bouncing around one steady rate. Set 2020 aside. The other 26 seasons sit too far apart to be one steady rate plus luck. The biggest jump comes at 2012. We tested it by shuffling the seasons into random order 2,000 times. If the order didn’t matter, a jump that sharp would turn up in a fair share of shuffles. None did. Split the years at 2002 and 2012 and you get three stretches. Inside each one, the seasons differ only as much as luck would explain. The earlier jump is the less certain of the two. It rests on just three seasons, 1999–2001, before it.
The level reached in 2012 has held. From 2021 to 2025, seasons ran between about 55% and 59%. The 2013–19 seasons averaged about 56%, and 2021–25 ran about a point higher. A gap that small could be luck.
This season hasn’t broken the pattern either. In weeks 1–6 of 2026, the complete weeks when our data was pulled on Oct. 26, 2026, offenses scored on about 61% of 602 trips, against about 58% in the same weeks of 2021–25. Over that many trips, a gap that size could be luck.
The table puts both checks side by side, each gap next to how big or small it could really be once luck is allowed for. Both ranges run from below zero to above it, so either gap could be nothing at all. Put the five seasons from 2021 to 2025 together, and the league’s real rate is most likely between about 56% and 59%.
| Comparison | Earlier | Later | Gap (points) | Likely range of the gap (95%) |
|---|---|---|---|---|
| 2013–19 against 2021–25 | 56.2% | 57.4% | +1.2 | −0.2 to +2.6 |
| Weeks 1–6: 2021–25 against 2026 | 57.5% (3,004 trips) | 60.6% (602 trips) | +3.1 | −1.2 to +7.4 |
Moving the start of the red zone in by a yard lifts every season but leaves the shape alone. We also counted only drives that reached the 19 or closer, so a drive whose deepest snap came at the 20 itself drops out. Counted that way, every season rises by about a point (0.6 to 1.3), and the two jumps stay where they are.
The same trips, finished better
The obvious suspect for the jumps is a friendlier field, with more drives reaching first-and-goal at the 3 and fewer stalling at the 19. The trips didn’t get easier. In 1999 the average trip first reached the red zone at the 13.8-yard line. In 2025 it was the 14.2.
Picture a drive that first reaches the red zone at first-and-goal from the 8. There is no first down left to gain, only the goal line. Trips that start like that, goal to go, edged down, from about 24% of trips in 1999–2005 to about 23% in 2021–25. Short fields fell too. Drives that began at the 20 or closer went from about 9% of trips to about 5%.
| Kind of trip | 1999–2005 | 2021–25 |
|---|---|---|
| First red-zone snap came with goal to go | 23.9% | 22.6% |
| Drive began at the 20 or closer (short field) | 8.5% | 5.1% |
The trips stayed the same. The finishing changed
Actual red-zone touchdown rate, against the rate each season would have had if only its mix of trips (where the drive first reached the 20 or closer, and on which down) had changed.
Show data
| Year | Actual | Trip mix only |
|---|---|---|
| 1999 | 47.9% | 54.8% |
| 2000 | 50.6% | 54.8% |
| 2001 | 49.0% | 54.4% |
| 2002 | 53.3% | 54.7% |
| 2003 | 52.1% | 54.3% |
| 2004 | 54.6% | 55.1% |
| 2005 | 51.4% | 54.5% |
| 2006 | 51.5% | 54.8% |
| 2007 | 52.2% | 53.7% |
| 2008 | 53.0% | 54.1% |
| 2009 | 52.7% | 54.4% |
| 2010 | 53.2% | 54.9% |
| 2011 | 52.3% | 55.4% |
| 2012 | 55.0% | 54.9% |
| 2013 | 56.0% | 54.9% |
| 2014 | 54.1% | 55.2% |
| 2015 | 56.6% | 54.3% |
| 2016 | 56.0% | 54.8% |
| 2017 | 53.9% | 54.6% |
| 2018 | 59.1% | 54.5% |
| 2019 | 57.4% | 54.9% |
| 2020 | 61.6% | 54.9% |
| 2021 | 59.1% | 54.3% |
| 2022 | 57.1% | 54.4% |
| 2023 | 55.2% | 54.1% |
| 2024 | 57.9% | 54.5% |
| 2025 | 57.7% | 54.3% |
The chart makes the same check a second way, in three steps. First, sort the trips into kinds by where the drive first reached the red zone and on which down. One kind is a drive that first got there between the 6 and the 10 on first down. Second, work out how often each kind scored over all 27 seasons together. Third, give every season those same rates, so the only thing that differs from year to year is how many trips of each kind it had. Done that way, the league rate would have stayed between about 54% and 55% throughout, while the real rate climbed. The check knows only the yard line and the down. It doesn’t know the score, the clock or the weather.
Offenses got better instead, from every starting point. The gains from the 16 to 20, the 11 to 15 and the 6 to 10 are bigger than luck would usually produce; the biggest came from the 6 to 10. From the 1 to 5, where most trips already ended in touchdowns, the gain is borderline and could be luck. The last column shows how far each gain could swing with luck. Only for the 1 to 5 does that range dip below zero.
| Drive first reached | TD rate 1999–2005 | TD rate 2021–25 | Change (points) | Likely range of the change (95%) |
|---|---|---|---|---|
| 16–20 yard line | 43.3% (5,259) | 49.6% (4,351) | +6.3 | +4.3 to +8.3 |
| 11–15 yard line | 52.3% (2,985) | 58.8% (2,419) | +6.6 | +3.9 to +9.2 |
| 6–10 yard line | 58.4% (1,436) | 67.3% (1,137) | +8.9 | +5.1 to +12.6 |
| 1–5 yard line | 77.5% (1,111) | 81.1% (809) | +3.6 | −0.1 to +7.2 |
More coaches passed up the three
Part of the better finishing came from the sideline. In 1999–2005, coaches went for it on fourth down inside the 20 on about 7% of red-zone trips. In 2021–25 they did on about 11%. They also made it more often: about 46% of tries then, about 55% now. Why coaches changed their minds is the fourth-down boom, a story of its own. This one counts what it did to the red zone.
In 1999–2005, about 2.5% of all trips ended in a touchdown on a drive that needed a converted red-zone fourth down. In 2021–25, about 5% did. That is about 2.5 points of the roughly 6-point rise, or about 42% of it. The exact share is less certain than that: it could be anywhere from about 31% to 57%.
Even without the fourth-down touchdowns, the red-zone rate rose
The lower line removes every touchdown that came on a drive where the offense went for it on fourth down inside the 20 and converted, as if those trips had all ended without one.
Show data
| Year | All touchdowns | Without fourth-down TDs |
|---|---|---|
| 1999 | 47.9% | 46.2% |
| 2000 | 50.6% | 48.1% |
| 2001 | 49.0% | 47.2% |
| 2002 | 53.3% | 50.5% |
| 2003 | 52.1% | 48.6% |
| 2004 | 54.6% | 51.3% |
| 2005 | 51.4% | 48.2% |
| 2006 | 51.5% | 48.8% |
| 2007 | 52.2% | 49.2% |
| 2008 | 53.0% | 49.2% |
| 2009 | 52.7% | 48.1% |
| 2010 | 53.2% | 50.3% |
| 2011 | 52.3% | 50.2% |
| 2012 | 55.0% | 51.8% |
| 2013 | 56.0% | 52.9% |
| 2014 | 54.1% | 51.8% |
| 2015 | 56.6% | 53.9% |
| 2016 | 56.0% | 53.0% |
| 2017 | 53.9% | 50.9% |
| 2018 | 59.1% | 55.0% |
| 2019 | 57.4% | 54.2% |
| 2020 | 61.6% | 57.5% |
| 2021 | 59.1% | 53.7% |
| 2022 | 57.1% | 52.5% |
| 2023 | 55.2% | 50.2% |
| 2024 | 57.9% | 53.2% |
| 2025 | 57.7% | 51.2% |
Strip those touchdowns out, as if every coach had kicked, and the rate still rose about 3.5 points, from about 49% to 52%, more than luck would usually produce. That lower line can only undercount, since some of those drives would have scored anyway.
The table lines up each step: how often coaches went for it, how often they made it, and how many touchdowns needed it. Its last row takes every one of those touchdowns out, and the rate still climbs. The brackets show how far each era’s make rate could swing with luck.
| Fourth downs inside the 20 | 1999–2005 | 2021–25 |
|---|---|---|
| Trips where the offense went for it | 6.8% | 11.0% |
| Tries converted (likely range) | 45.7% (42.2–49.3%) | 55.1% (52.0–58.2%) |
| Trips ending in a TD that needed the conversion | 2.7% | 5.2% |
| Red-zone TD rate, all touchdowns | 51.3% | 57.4% |
| Red-zone TD rate without those touchdowns | 48.7% | 52.2% |
Play by play, the change is small. Inside the 10, passes scored on about 31% of attempts in 1999–2005 and 34% in 2021–25. Runs by players other than the quarterback went from about 26% to 27%. Teams also threw more near the goal line. Passes made up about 44% of plays inside the 10 in 1999–2005. That climbed to about 53% in 2013–19, then eased back to about 50% in 2021–25. Inside the 2, quarterback runs rose from about 8% of plays to 12%, which is the sneak and the tush push. Touchdowns per play rose a little and touchdowns per drive rose more, which fits drives getting extra downs.
| Play | 1999–2005 | 2013–19 | 2021–25 |
|---|---|---|---|
| Inside the 10: passes that scored | 30.7% | – | 33.7% |
| Inside the 10: non-quarterback runs that scored | 25.9% | – | 27.2% |
| Inside the 10: passes, as a share of all plays | 43.5% | 52.7% | 50.3% |
| Inside the 2: quarterback runs, as a share of all plays | 8.4% | – | 11.8% |
The new touchdowns replaced turnovers and missed kicks more than made field goals
Each extra touchdown took a trip away from some other ending. In 2025, 113 trips reached fourth down inside the 20 and still ended in a touchdown, not a field-goal try or a turnover on downs. Teams settled for field goals less often, but not by much. Field-goal attempts on a trip fell from 36.7% to 32.3%, a drop of under five points.
Touchdowns rose as field goals, misses and turnovers fell
Share of red-zone trips by result, 1999–2005 against 2021–25.
Show data
| 1999–2005 | 2021–25 | Change | |
|---|---|---|---|
| Touchdown | 51.3% | 57.4% | +6.1 pts |
| Field goal made | 32.4% | 30.4% | −2.0 pts |
| Turnover | 8.0% | 5.4% | −2.6 pts |
| Turnover on downs | 3.6% | 4.9% | +1.3 pts |
| Field goal missed | 4.3% | 1.8% | −2.5 pts |
Turnovers fell from about 8% of trips to 5%. Missed field goals fell from about 4% to 2%, as kickers got far more accurate. Going for it has a price. Turnovers on downs, fourth-down tries that came up short, rose from about 4% to 5%.
The scoreboard gain is smaller than the touchdown rate suggests. A trip was worth about 4.6 points in 1999–2005 and 4.9 in 2021–25 (4.55 and 4.90), counting the try, the extra point or two-point play after a touchdown. A touchdown with its try stayed worth about 7 points both times (6.97 and 6.95). The extra point was moved back in 2015, so kickers made fewer of them, while two-point tries rose. At 3.2 trips per team per game, better finishing is worth about 1.1 points a game.
For one team, finishing drives is mostly luck. Getting there is steadier
A single team’s rate is far shakier than the league’s. Take the best team-season in our data, the 2020 Green Bay Packers under Matt LaFleur. With Aaron Rodgers at quarterback, they scored on 48 of 58 trips (about 83%). Sports Illustrated’s Bill Huber, who counted it as 80 percent, later called it “perhaps the best mark in NFL history.” The next season the Packers scored on about 60%.
That kind of drop is normal. Red-zone rate is one of the most quoted team stats, and one of the noisiest a team posts. A team that finishes drives well one year usually lands much closer to the league average the next. Of the team measures we checked, red-zone touchdown rate repeats from one year to the next least of all. Point margin per game and offense outside the red zone each repeat about twice as well. So does getting to the red zone in the first place, measured as trips per drive.
To compare them, give each measure a score from 0 to 1. At 0, this season tells you nothing about the next. At 1, next season is an exact repeat. Red-zone rate scores around 0.2. The others all score around 0.4. Offense outside the red zone is judged here by expected points per play, which gives a 9-yard catch on third-and-8 more credit than the same catch on third-and-20. One keeps the drive going. The other sets up a punt.
| Team measure | How closely next season matches (0 to 1) | Likely range (95%) |
|---|---|---|
| Red-zone touchdown rate | 0.21 | 0.15–0.28 |
Continues below
| Team measure | How closely next season matches (0 to 1) | Likely range (95%) |
|---|---|---|
| Point margin per game | 0.40 | – |
| Offense outside the red zone (expected points per play) | 0.44 | – |
| Getting to the red zone (trips per drive) | 0.38 | – |
A team’s red-zone number bounces because trips are scarce. A team gets only about 51 a season, and over that few trips the bounce of the ball moves the rate a lot. In a single season, teams’ rates spread about 8 points either side of the league. Coin-flip luck alone would spread them about 7 points, nearly as much as the spread we actually see. Take the luck out, and the real differences between offenses come to about 4 points either way. So only about a quarter of the spread in a single season is real; the rest is luck.
Some teams really are better at it. Of 861 team-seasons, 88 finished so far from their league that luck alone would put a team there only one time in 20. To see how many luck would produce on its own, we had a computer replay all 27 seasons 500 times with every offense equally good, so only the bounces differed. Those make-believe leagues usually had about 43 such seasons. Nine in ten of them had between 33 and 54. The real league has about twice as many.
Luck also writes the top of the list. In those make-believe leagues, where no offense was better than another, the luckiest team-season of 30 or more trips usually finished around 78%, and nine times in ten between 75% and 83%. So Green Bay’s 82.8% is within reach of pure luck. Denver’s 55 of 70 is different, because it came over far more trips. In 2,000 replays of all 27 seasons with every offense equal, only about 3 in 100 had any team-season that far above its league. So Denver’s 2013 red zone was probably not just luck.
The 2012 Kansas City Chiefs are the other end. Under Romeo Crennel, with Matt Cassel and Brady Quinn splitting the starts, they went 2–14 and scored on 10 of 36 trips, the furthest below its league of any team-season in our data. In nine games they reached the red zone and came away with no touchdown, including the finale on Dec. 30, 2012, a 38–3 loss at Denver. The next year, under Andy Reid with Alex Smith, Kansas City scored on 34 of 57 (59.6%).
Crennel also coached the 2005 Cleveland Browns, 11 of 39 (28.2%); two of the three lowest rates in our data are his. A coach’s name sticks to these numbers only loosely. John Fox’s 2010 Carolina Panthers scored on 10 of 31, the sixth-lowest rate in our data. By 2013 Fox was coaching Denver, with Manning at quarterback: the record team from the start of this story. He wasn’t even on the sideline for all of that record. He missed four games, weeks 10 to 13, while he recovered from heart surgery, and Jack Del Rio stood in.
The bottom of the list has a regular, and it isn’t one coach or one quarterback. The New York Jets keep turning up there. Three of the eight team-seasons furthest below their league are theirs: 2016 (Todd Bowles, Ryan Fitzpatrick, 34.6%), 2020 (Adam Gase, Sam Darnold, 40.5%) and 2023 (Robert Saleh, Zach Wilson, 32.4%). Gase’s 2020 team managed it in the league’s best season.
| Season | Team | Coach | QB (most starts) | Record | TDs / trips | TD rate | League | Next season |
|---|---|---|---|---|---|---|---|---|
| 2020 | Packers | Matt LaFleur | Aaron Rodgers | 13–3 | 48 of 58 | 82.8% | 61.6% | 60.3% |
Continues below
| Season | Team | Coach | QB (most starts) | Record | TDs / trips | TD rate | League | Next season |
|---|---|---|---|---|---|---|---|---|
| 2013 | Broncos | John Fox | Peyton Manning | 13–3 | 55 of 70 | 78.6% | 56.0% | 64.5% |
| 2005 | Seahawks | Mike Holmgren | Matt Hasselbeck | 13–3 | 43 of 55 | 78.2% | 51.4% | 51.1% |
| 2003 | Chiefs | Dick Vermeil | Trent Green | 13–3 | 45 of 58 | 77.6% | 52.1% | 65.1% |
| 2024 | Ravens | John Harbaugh | Lamar Jackson | 12–5 | 50 of 66 | 75.8% | 57.9% | 48.3% |
| 2013 | Bengals | Marvin Lewis | Andy Dalton | 11–5 | 36 of 48 | 75.0% | 56.0% | 55.8% |
| 2006 | Raiders | Art Shell | Andrew Walter | 2–14 | 9 of 33 | 27.3% | 51.5% | 55.3% |
| 2012 | Chiefs | Romeo Crennel | Cassel / Quinn (8 each) | 2–14 | 10 of 36 | 27.8% | 55.0% | 59.6% |
| 2005 | Browns | Romeo Crennel | Trent Dilfer | 6–10 | 11 of 39 | 28.2% | 51.4% | 47.4% |
| 2000 | Cardinals | Vince Tobin and Dave McGinnis | Jake Plummer | 3–13 | 14 of 45 | 31.1% | 50.6% | 44.2% |
| 2008 | Rams | Jim Haslett and Scott Linehan | Marc Bulger | 2–14 | 10 of 32 | 31.3% | 53.0% | 33.3% |
Continues below
| Season | Team | Coach | QB (most starts) | Record | TDs / trips | TD rate | League | Next season |
|---|---|---|---|---|---|---|---|---|
| 2010 | Panthers | John Fox | Jimmy Clausen | 2–14 | 10 of 31 | 32.3% | 53.2% | 56.9% |
Every one of the top six fell the next season, and every one of the bottom six rose. The same thing happens across all teams. The top fifth of team-seasons ran about 11 points above their league and only about 2 above it the next year. The bottom fifth ran about 12 below, then about 3 below.
| Team-seasons | This season, against the league | Next season, against the league |
|---|---|---|
| Top fifth | +11.1 | +2.1 |
| Bottom fifth | −12.0 | −3.1 |
So to guess a team’s red-zone rate next year, look past this year’s red-zone rate to its offense outside the red zone. That is the better guide, about twice as good, though neither explains even a tenth of the differences between teams the next season. For next year’s wins, this year’s point margin is the guide, and adding red-zone rate adds almost nothing. In the table below, a guide that called every team’s next season exactly would score 100%.
| Next season | Guide from this season | How much of next season it explains |
|---|---|---|
| Red-zone TD rate | Red-zone TD rate | 4.5% |
Continues below
| Next season | Guide from this season | How much of next season it explains |
|---|---|---|
| Red-zone TD rate | Offense outside the red zone (expected points per play) | 8.8% |
| Win percentage | Point margin | 13.3% |
| Win percentage | Point margin plus red-zone TD rate | 13.4% |
Over a long career, the best quarterbacks beat the noise
Luck evens out over enough trips, so the place to look for real red-zone skill is a long career. Season to season, a quarterback’s red-zone rate carries over about as weakly as a team’s: 0.20 on the same 0-to-1 scale, across 465 pairs of seasons with 30 or more trips. Over a career, though, the small edges add up. Of the 45 quarterbacks who started on 300 or more trips, ten finished so far from average that luck alone would put a quarterback there only one time in 20. Chance alone would put about two there. All of them are above average.
Manning and Brees trade the top spot, depending on how you count. In the games Manning started, his offenses scored on about 62% of 901 trips, about 81 touchdowns more than an average offense in the same seasons. For the number of trips he had, that puts him furthest above average. Drew Brees (about 62% of 1,079) has more touchdowns above average, about 84, over more trips. Tom Brady (about 61% of 1,269) is third on both. We picked these names after looking, and the figures carry their coaches and teammates with them, so read them as a modest signal.
Over Manning’s 901 trips, luck alone usually moves a career about 15 touchdowns either way. His 81 is 5.4 times that. A career more than twice luck’s usual swing from average, above or below, turns up by luck only about one time in 20. Even the bottom four stay inside that.
| Quarterback | Seasons | Trips | TD rate | TDs above average | Times luck’s usual swing |
|---|---|---|---|---|---|
| Peyton Manning | 1999–2015 | 901 | 61.6% | +81 | +5.4 |
Continues below
| Quarterback | Seasons | Trips | TD rate | TDs above average | Times luck’s usual swing |
|---|---|---|---|---|---|
| Drew Brees | 2002–2020 | 1,079 | 62.4% | +84 | +5.2 |
| Tom Brady | 2001–2022 | 1,269 | 60.9% | +78 | +4.4 |
| Kurt Warner | 1999–2009 | 424 | 60.6% | +39 | +3.8 |
| Trent Green | 2000–2008 | 353 | 61.2% | +32 | +3.4 |
| Aaron Rodgers | 2008–2025 | 905 | 61.4% | +48 | +3.2 |
| Alex Smith | 2005–2020 | 502 | 52.0% | −14 | −1.2 |
| Joe Flacco | 2008–2025 | 668 | 52.5% | −17 | −1.3 |
| Drew Bledsoe | 1999–2006 | 342 | 47.7% | −14 | −1.5 |
| Michael Vick | 2001–2015 | 382 | 49.2% | −15 | −1.5 |
2020 ran hot for one season
The timeline has one exception, and it is why the tests of when the level changed set 2020 aside. In 2020 offenses scored on 61.6% of 1,773 trips, the highest league rate in our data. That was about 4 points above 2021–25, more than luck would usually produce. It was a season played, in Wikipedia’s words, “with either a greatly reduced audience or no fans at all,” and with no preseason.
In 2020, home and road offenses finished trips at the same rate, 61.6% at home and 61.6% away. In every other season combined, home offenses converted about 2 points more, a home edge bigger than luck would usually produce.
Each gap below comes with how far luck could stretch it. The 2020 jump and the usual home edge stay above zero even at the low end. One season is too few trips to pin down the 2020 home edge. It could have been anything within about 4.5 points either way, so the data can’t tell it from no edge at all.
| Comparison | Gap (points) | Likely range (95%) |
|---|---|---|
| 2020 against 2021–25 | +4.2 | +1.7 to +6.7 |
| Home minus road, every season but 2020 | +2.2 | +1.3 to +3.2 |
| Home minus road, 2020 | −0.1 | −4.6 to +4.5 |
Officials also called less holding. By ESPN’s count, 95 offensive holding flags were thrown in the first 48 games, 59% fewer than in 2019. Yahoo Sports’ Henry Bushnell tallied the full season: “They called 2.14 offensive holds per game, a sharp decline from 3.42 and 3.45 the two seasons prior.”
Empty stadiums, fewer flags and no preseason all came in the same year. The data can’t tell which of them mattered, or whether any did. The rate fell back the next year, so we treat 2020 as one season and leave it out of the tests for when the level changed.
In 2025, coaches went for it inside the 20 on 228 trips
Back to the sideline, and the last full season. No season in our data had more trips with a fourth-down try inside the 20 than 2025’s 228. Of those, 113 ended in a touchdown that needed the conversion, the highest share of trips in any season. The league scored on 57.7% of trips, inside the range it has held since 2012.
Now strike those 113 touchdowns, and the same kind from every other season, as if every coach had kicked. 2025 drops to about 51%. The real figure can only be higher, since some of those drives would have scored anyway. Of the other seasons since 2012, leaving out 2020, only two would sit lower. Counted that way, the 13 seasons from 2012 to 2025, 2020 aside, ran from about 50% to 55%, and 2002 to 2011 from about 48% to 51%. Without its fourth downs, 2025 looks most like 2004, also about 51%, a season from before the league’s last jump, though 2017 and 2023 sat lower still.
The league still finishes trips at about the level it reached the year Manning’s Broncos went six for seven in Dallas. It gets there with more fourth downs than ever.
Share this article
Running back tools
Live Edgehalla tools on the same subject.
Show the work
- Tests written down
- Not in advance (exploratory)
- Results computed
nflverse play-by-play, regular season 1999–2025 (43,715 trips), with 2026 weeks 1–6 (the complete weeks in the file pulled Oct. 26, 2026) held out as a check. A trip is a drive (nflverse game and fixed_drive) with at least one offensive snap at or inside the opponent 20; plays wiped out by penalties and two-point tries don’t count. Overtime trips and trips that ended with the half are left out. A touchdown is an offensive passing or rushing touchdown on the drive. We don’t use nflverse’s td_team field alone: it credits the visiting team with Jacksonville’s touchdowns in 16 Jaguars home games in 2001–02, plus one 2011 play. After that fix the 2001 Jaguars score on 27 of 53 trips; counting only drives that got strictly inside the 20, 24 of 46. The team’s own season review gave 24 of 48: the touchdowns match on that definition and the trips differ by two. Coaches and starting quarterbacks come from the nflverse games file (one per team-game; John Fox is listed for all 16 Denver games in 2013, though Jack Del Rio coached weeks 10–13). Shaded bands in the season chart are 95% ranges; its dashed lines are the levels of the three stretches between the two breaks the tests below pick out, fitted without 2020. Ranges in the tests are 95% unless marked 90%. Descriptive only: no causal claim.
| Test | Result | Verdict |
|---|---|---|
| One steady rate, 1999–2025 (2020 out) | χ² 142.0 on 25 df | Rejected |
More tests below
| Test | Result | Verdict |
|---|---|---|
| Best single break in the season series | 2012: 51.9% before, 56.6% after; permutation p = 0.0005 (2,000 shuffles) | Real |
| Two breaks: is each stretch stable? | 2002 and 2012: 49.2%, 52.6%, 56.6%; within-stretch χ² 30.9 on 23 df, p = 0.13 | Stable |
| 2013–19 vs 2021–25 | 56.2% vs 57.4%; +1.2 points (−0.2 to +2.6) | Flat within noise |
| 2021–25 level from the five season rates (t-interval) | 55.6–59.1%; seasons 55.2–59.1% | Plateau |
| Holdout: 2026 weeks 1–6 vs same weeks 2021–25 | 60.6% (n = 602) vs 57.5% (n = 3,004); +3.1 (−1.2 to +7.4) | Noise; consistent with the plateau |
Continues below
| Test | Result | Verdict |
|---|---|---|
| Trip-mix-only rate (entry yard line and down at pooled rates) | 53.7–55.4% in every season | Mix did not drive the rise |
| Entry at the 16–20: 1999–2005 vs 2021–25 | 43.3% → 49.6%; +6.3 (+4.3 to +8.3) | Real gain |
| Entry at the 11–15: 1999–2005 vs 2021–25 | 52.3% → 58.8%; +6.6 (+3.9 to +9.2) | Real gain |
| Entry at the 6–10: 1999–2005 vs 2021–25 | 58.4% → 67.3%; +8.9 (+5.1 to +12.6) | Real gain |
| Entry at the 1–5: 1999–2005 vs 2021–25 | 77.5% → 81.1%; +3.6 (−0.1 to +7.2) | Borderline |
Continues below
| Test | Result | Verdict |
|---|---|---|
| Share of the rise from TDs after a converted red-zone fourth down | such TDs 2.7% → 5.2% of trips; 2.6 of 6.1 points = 42.0% (the body rounds these four to the nearest half point, so they add up); bootstrap 95% 31.4–57.1% (2,000 draws) | About four in ten |
| Rise without those TDs (kick-only bound) | 48.7% → 52.2%; +3.5 (+2.1 to +4.9) | Real gain remains |
| 2020 vs 2021–25 and vs 2013–19 | +4.2 (+1.7 to +6.7); +5.4 (+3.0 to +7.8) | One-season outlier, no cause assigned |
| Home minus road TD rate: 2020 vs all other seasons | −0.1 (−4.6 to +4.5) vs +2.2 (+1.3 to +3.2) | Home edge absent in 2020 |
| Team red-zone TD rate (vs league), year to year | r = 0.21 (0.15–0.28, n = 829); without 2020 pairs 0.20; split-half 0.11, full-season 0.21 | Weak |
Continues below
| Test | Result | Verdict |
|---|---|---|
| Comparison: year-to-year r of other team measures | red-zone TD rate 0.21; EPA/play outside the 20 0.44; trips per drive 0.38; point margin 0.40; points per trip 0.24 | Red zone repeats least |
| Signal share of single-season spread | observed SD 8.2, luck SD 7.1, true SD 4.1 points (~51 trips per team-season); 25% signal = true variance ÷ observed variance (SDs squared, which is why 4.1 of 8.2 is about a quarter, not a half); defense 12% | Mostly luck |
| Team-seasons beyond ±1.96 vs luck-only simulation | 88 observed vs 43 (90% 33–54), 500 simulations | Some real skill |
| Best team-season rate (30+ trips) in a luck-only league | median 78.0% (90% 74.6–82.9%); actual best 82.8% | Best rate within luck’s reach (small samples set the top) |
Continues below
| Test | Result | Verdict |
|---|---|---|
| Any team-season as far above its league as the named one (z), luck-only leagues | 2003-KC z 3.9: 2.3% of runs; 2005-SEA z 4.0: 1.9% of runs; 2013-DEN z 3.8: 3.3% of runs; 2020-GB z 3.3: 22.6% of runs (2,000 runs) | Denver, Seattle, Kansas City rare by luck; Green Bay not |
| Next-season red-zone rate: R² from this year’s red-zone rate, outside-the-20 EPA, both | 4.50% / 8.75% / 9.41% (n = 829) | Offense elsewhere predicts better |
| Next-season win %: point margin, then plus red-zone rate | R² 13.30% → 13.36%; red-zone coefficient 95% −0.11 to 0.24 | Adds nothing visible |
| Quarterback careers (300+ trips) beyond ±1.96 | 10 of 45 vs 2.3 by luck; QB-season r 0.20 (n = 465) | Signal over a career; names picked after the fact |
Continues below
| Test | Result | Verdict |
|---|---|---|
| Strictly inside the 20 instead of at or inside | 1999 48.9%, 2025 58.7% | About a point higher; same shape |
Frequently asked questions
What is a good red zone touchdown percentage in the NFL?
The league average was 57.4% in 2021–25, with seasons between 55.2% and 59.1%. The top fifth of team-seasons since 1999 ran about 11 points above their league’s rate. Most of that edge doesn’t last. The same teams were about 2 points above the next year.
Which NFL team had the best red zone percentage ever?
In play-by-play data since 1999, the 2020 Green Bay Packers: 48 of 58 trips, 82.8%. The 2013 Denver Broncos scored the most red-zone touchdowns, 55 on 70 trips. Measured against their own league, the 2005 Seahawks (78.2%, coached by Mike Holmgren) stand furthest above average.
Is red zone efficiency predictive?
Weakly. A team’s red-zone touchdown rate one season tells you little about its rate the next (a correlation of 0.21). Its offense outside the red zone predicts next year’s red-zone rate better, and red-zone rate adds almost nothing to point margin when predicting next season’s wins.
Why did NFL red zone touchdown rates go up?
The trips didn’t get easier. Entry yard lines and downs barely changed. About 42% of the rise from 1999–2005 to 2021–25 came on drives where a coach went for it on fourth down inside the 20 and converted. Fewer turnovers and missed field goals account for much of the rest of the shift in outcomes.
Why did NFL scoring spike in 2020?
In the red zone, 2020 was the best season in our data, 61.6%. Home and road offenses converted at the same rate, offensive holding calls fell sharply, and there was no preseason. The data can’t say which of these, if any, caused it, and the rate fell back in 2021.
More in The long arc
- Verdict: Data story
NFL Offenses Got Better on Third Down Around 2014. Since 2020, Third Downs Have Been Shorter.
- Verdict: Data story
Whatever Happened to the No-Huddle?
- Verdict: Data story
NFL Coaches Throw About Half as Many Red Flags as They Did in 2007
- Verdict: Data story
A Quarter Fewer NFL Deep Throws Go Down the Middle
- Verdict: Data story
About Seven NFL Head Coaches Lose Their Jobs Every Year, and Always Have
- Verdict: Data story
Good NFL Teams Stay Good for Less Time Than They Used To
Sources and method
- nflverse play-by-play and players (CC-BY 4.0)
- nflverse games file (nflverse/nfldata): results, head coaches and starting quarterbacks, 1999–2025
- 2013 Denver Broncos season (Wikipedia): 606 points; Fox’s heart surgery; Del Rio interim, weeks 10–13
- 2020 NFL season (Wikipedia): reduced or no crowds; preseason canceled
- Bill Huber, “Red zone has Packers singing the blues”, Sports Illustrated (All Packers), Nov. 7, 2022
- Kevin Seifert, “How the NFL is manipulating scoring in 2020”, ESPN, Sept. 30, 2020
- Henry Bushnell, “The NFL broke scoring records in 2020”, Yahoo Sports, Jan. 5, 2021
- Jacksonville Jaguars, 2001 season review (Jaguars.com), Jan. 9, 2002
Research and analysis, not betting advice. Past results do not guarantee future returns; bet only what you can afford to lose.