How Football ChangedEndangered roles
The Running-Back Cliff Barely Moved. Fewer Backs Reach It.
In 2003, backs aged 28 or older got a third of NFL running-back carries. By 2022 they got one in eight. The work went to younger backs. Backs still slow down in their late 20s. They run about a fifth to a quarter of a yard a carry worse than backs of 22 to 25, much as they did in 1999–2010. What changed is how many are still carrying the ball at 27.
By Edgehalla ResearchPublished 12 min read
On July 22, 2023, some of the NFL’s best running backs got on a private Zoom call about their shrinking market. Austin Ekeler set it up. Christian McCaffrey, Nick Chubb, Saquon Barkley, Josh Jacobs and Derrick Henry were among the backs who joined him.
The next day Chubb talked about it. “We’re the only position that our production hurts us the most,” he said. “If we go out there and run 2,000 yards with so many carries, the next year they’re going to say, you’re probably worn down.”
The season before, backs 28 and older took 12.3% of running-back carries, the smallest share in our data, close to 2019 (12.4%).
On this page
The carries went young
Chubb was talking about contracts, and you can see his point in who carried the ball. In 2003, backs who were 28 or older on September 1 got 33.5% of NFL running-back carries, about a third. By 2022 they got one in eight. In 2025 they were back up to 18.0%.
So do backs now fall off at 28 sooner than they used to, or do fewer of them keep the ball long enough to find out? The numbers point to the second. A plain count of the backs who reached 27 with a full workload settles it, further down. To compare, we split the seasons into two eras, 1999–2010 and 2014–2025, and leave out the three seasons in between.
Meanwhile the young backs got busier. Backs 24 and under got about 36 of every 100 running-back carries in an average season from 1999 to 2010, and about 47 from 2014 to 2025. Their share peaked at 61.2% in 2019. The 2017 draft class alone took about a quarter of all running-back carries that year (24.5%), and the 2018 class about a fifth (21.2%). A year later, in 2020, the young backs’ share was down to 46.1%.
Backs 28 and older fell from a third of the carries to one in eight, then won a little back
Regular season, 1999–2025. Age on September 1 of each season.
Show data
| Year | Backs 28 and older | Backs 24 and younger |
|---|---|---|
| 1999 | 21.6% | 35.2% |
| 2000 | 28.2% | 29.1% |
| 2001 | 30.5% | 37.3% |
| 2002 | 31.1% | 38.8% |
| 2003 | 33.5% | 35.6% |
| 2004 | 30.9% | 32.8% |
| 2005 | 26.2% | 34.9% |
| 2006 | 26.1% | 31.5% |
| 2007 | 31.7% | 32.5% |
| 2008 | 30.9% | 32.9% |
| 2009 | 24.1% | 42.8% |
| 2010 | 21.4% | 50.5% |
| 2011 | 26.2% | 37.1% |
| 2012 | 15.1% | 37.3% |
| 2013 | 24.5% | 33.7% |
| 2014 | 25.2% | 43.7% |
| 2015 | 28.5% | 45.9% |
| 2016 | 25.4% | 48.4% |
| 2017 | 21.1% | 47.7% |
| 2018 | 16.6% | 60.6% |
| 2019 | 12.4% | 61.2% |
| 2020 | 14.1% | 46.1% |
| 2021 | 15.1% | 45.0% |
| 2022 | 12.3% | 45.9% |
| 2023 | 18.9% | 45.4% |
| 2024 | 20.5% | 37.9% |
| 2025 | 18.0% | 40.0% |
- 2011: New labor deal
- 2021: 17-game season
The solid line, backs 28 and older, wanders through the 2000s, drops hard from 2017 and climbs a little after 2022. The dashed line for backs 24 and under moves roughly the other way.
The great exception was Frank Gore. In 2020 he was 37 and carried 187 times for the New York Jets, the oldest season with 100 or more carries in our data. That same year backs 28 and older took 14.1% of the league’s running-back carries. Gore had 100+ carries in sixteen straight seasons, from 2005 to 2020, the longest run in our data (Thomas Jones is next with twelve). Carry counts come from play-by-play and can run a few below official totals.
His coach that season, Adam Gase, put it this way in Tom Junod’s 2023 ESPN profile of Gore: “I hate to say this, but Frank was 37 years old and the best player on the team.” Gore averaged 0.89 yards a carry below the league’s running-back average that year. That number compares him with every running back in the league. Gase was comparing him with his own teammates.
Thirty-year-olds lost the most
Split the backs into four age groups: 24 and under, 25 to 27, 28 and 29, and 30 and older. Most of the work lost by backs 28 and older came out of the 30-and-older group. In an average season, backs 30 and older went from about 13 of every 100 carries in 1999–2010 to about 7 in 2014–2025. Backs 28 and 29 went from about 15 to about 12. Backs 30 and older got about 15% of the carries in 2003. In 2023 they got about 4%. Backs 25 to 27 swung from season to season but averaged about the same in both eras, about 36% and 34%.
Backs 30 and older lost the most ground; backs 25 to 27 held steady
Share of all running-back carries each season. The four lines add to 100%.
Show data
| Year | 24 and under | 25–27 | 28–29 | 30 and older |
|---|---|---|---|---|
| 1999 | 35.2% | 43.2% | 8.5% | 13.1% |
| 2000 | 29.1% | 42.7% | 19.2% | 9.0% |
| 2001 | 37.3% | 32.2% | 19.9% | 10.5% |
| 2002 | 38.8% | 30.1% | 15.4% | 15.7% |
| 2003 | 35.6% | 30.9% | 18.6% | 14.9% |
| 2004 | 32.8% | 36.2% | 14.5% | 16.5% |
| 2005 | 34.9% | 38.9% | 8.5% | 17.8% |
| 2006 | 31.5% | 42.3% | 15.2% | 10.9% |
| 2007 | 32.5% | 35.8% | 22.2% | 9.6% |
| 2008 | 32.9% | 36.1% | 19.9% | 11.1% |
| 2009 | 42.8% | 33.1% | 12.0% | 12.2% |
| 2010 | 50.5% | 28.1% | 11.2% | 10.2% |
| 2011 | 37.1% | 36.7% | 20.0% | 6.2% |
| 2012 | 37.3% | 47.7% | 10.0% | 5.0% |
| 2013 | 33.7% | 41.8% | 14.8% | 9.7% |
| 2014 | 43.7% | 31.1% | 18.7% | 6.5% |
| 2015 | 45.9% | 25.6% | 17.6% | 10.9% |
| 2016 | 48.4% | 26.2% | 13.8% | 11.6% |
| 2017 | 47.7% | 31.1% | 10.0% | 11.1% |
| 2018 | 60.6% | 22.8% | 7.7% | 8.9% |
| 2019 | 61.2% | 26.4% | 7.1% | 5.3% |
| 2020 | 46.1% | 39.8% | 7.8% | 6.2% |
| 2021 | 45.0% | 40.0% | 9.0% | 6.1% |
| 2022 | 45.9% | 41.8% | 6.3% | 6.0% |
| 2023 | 45.4% | 35.7% | 15.2% | 3.6% |
| 2024 | 37.9% | 41.6% | 15.8% | 4.7% |
| 2025 | 40.0% | 42.0% | 11.0% | 7.0% |
The dashed line for backs 30 and older sinks toward the floor. The line for 25 to 27 bounces around without going anywhere.
In 2004 the busiest back in the league was 31 years old. Curtis Martin carried 370 times for the New York Jets and ran for 1,697 yards. Emmitt Smith, 35, carried 267 times for the Arizona Cardinals, and Jerome Bettis, 32, carried 250 times for the Pittsburgh Steelers. Three backs aged 30 or older reached 250 carries that season.
The 250-carry thirty-year-old all but vanished
Martin’s kind of season, 250 carries after 30, nearly disappeared. There were 18 of them from 1999 to 2006 and 4 from 2007 to 2014. Then came 5 from 2015 to 2022, and 2 in the three seasons since.
250-carry seasons by backs 30 and older fell from 18 in 1999–2006 to 2 in 2023–25
Backs aged 30+ with 250 or more carries, count per period, regular season
Thirty-year-old workhorses went from 18 in 1999–2006 to 5 in 2015–2022, then 2 in 2023–2025.
Each of the first three bars counts eight seasons; the last counts three. The last 250-carry season by a back 30 or older before the gap was Adrian Peterson’s, at 33, with 251 carries for Washington in 2018. From 2019 to 2023 no back aged 30 or older reached 250 carries, even after the 17th game arrived in 2021 and made 250 a little easier to reach. Derrick Henry did it in 2024 and again in 2025.
The table lists every one of them. Of the 29 rows, 18 come from the first eight seasons.
| Season | Back | Team | Age | Carries | Rush yards |
|---|---|---|---|---|---|
| 1999 | Terry Allen | NE | 31 | 250 | 887 |
| 1999 | Emmitt Smith | DAL | 30 | 326 | 1,392 |
| 1999 | Ricky Watters | SEA | 30 | 323 | 1,198 |
| 2000 | Emmitt Smith | DAL | 31 | 292 | 1,197 |
| 2000 | Ricky Watters | SEA | 31 | 275 | 1,223 |
| 2001 | Garrison Hearst | SF | 30 | 252 | 1,206 |
| 2001 | Emmitt Smith | DAL | 32 | 261 | 1,021 |
| 2001 | Lamar Smith | MIA | 30 | 313 | 968 |
| 2002 | Antowain Smith | NE | 30 | 252 | 982 |
| 2002 | Emmitt Smith | DAL | 33 | 254 | 975 |
| 2003 | Curtis Martin | NYJ | 30 | 323 | 1,308 |
| 2004 | Jerome Bettis | PIT | 32 | 250 | 941 |
| 2004 | Curtis Martin | NYJ | 31 | 370 | 1,697 |
| 2004 | Emmitt Smith | ARI | 35 | 267 | 937 |
| 2005 | Tiki Barber | NYG | 30 | 357 | 1,860 |
| 2005 | Warrick Dunn | ATL | 30 | 280 | 1,416 |
Continues below
| Season | Back | Team | Age | Carries | Rush yards |
|---|---|---|---|---|---|
| 2006 | Tiki Barber | NYG | 31 | 327 | 1,662 |
| 2006 | Warrick Dunn | ATL | 31 | 286 | 1,140 |
| 2008 | Thomas Jones | NYJ | 30 | 290 | 1,312 |
| 2009 | Thomas Jones | NYJ | 31 | 332 | 1,402 |
| 2013 | Frank Gore | SF | 30 | 276 | 1,128 |
| 2014 | Frank Gore | SF | 31 | 255 | 1,106 |
| 2015 | Frank Gore | IND | 32 | 260 | 967 |
| 2015 | Adrian Peterson | MIN | 30 | 327 | 1,485 |
| 2016 | Frank Gore | IND | 33 | 263 | 1,025 |
| 2017 | Frank Gore | IND | 34 | 260 | 961 |
| 2018 | Adrian Peterson | WAS | 33 | 251 | 1,042 |
| 2024 | Derrick Henry | BAL | 30 | 325 | 1,921 |
| 2025 | Derrick Henry | BAL | 31 | 307 | 1,595 |
Backs do fade from about 27
So the older backs stopped getting the ball. Now look at the ones who kept it, and whether they faded sooner than backs used to.
“When you turn 28, you’re going to hear as a running back, you’re declining,” Gore said in the same profile. He meant it as a complaint: the label sticks even after a good season. The per-carry numbers show a real fade starting in the late 20s, though it is a small one.
The chart measures each back against the average running back that season. A back at zero was exactly average. In both eras the lines sit mostly above zero in the early 20s and slide below it through the late 20s.
In both eras backs run best in their early 20s and slide through their late 20s
Yards a carry against that season’s league running-back average (0 = average). Backs with 50+ carries. Shaded bands show where the true value probably lies.
yds
Show data
| 1999–2010 | 1999–2010 (range) | 2014–2025 | 2014–2025 (range) | |
|---|---|---|---|---|
| 22 | 0.07 yds | −0.12 yds to 0.26 yds | 0.13 yds | −0.02 yds to 0.28 yds |
| 23 | 0.05 yds | −0.08 yds to 0.19 yds | 0.04 yds | −0.09 yds to 0.16 yds |
| 24 | 0.14 yds | 0.00 yds to 0.28 yds | −0.02 yds | −0.13 yds to 0.09 yds |
| 25 | 0.07 yds | −0.07 yds to 0.21 yds | 0.09 yds | −0.02 yds to 0.21 yds |
| 26 | −0.03 yds | −0.18 yds to 0.12 yds | −0.01 yds | −0.14 yds to 0.11 yds |
| 27 | −0.03 yds | −0.16 yds to 0.10 yds | −0.13 yds | −0.30 yds to 0.03 yds |
| 28 | −0.08 yds | −0.25 yds to 0.08 yds | −0.15 yds | −0.32 yds to 0.03 yds |
| 29 | −0.22 yds | −0.36 yds to −0.08 yds | −0.23 yds | −0.38 yds to −0.09 yds |
| 30 | −0.05 yds | −0.35 yds to 0.24 yds | −0.27 yds | −0.55 yds to 0.00 yds |
| 31 | −0.03 yds | −0.24 yds to 0.18 yds | −0.05 yds | −0.40 yds to 0.31 yds |
The fade is about the same size it was
If backs wear down sooner now, that slide should be bigger than it used to be. It has barely grown. In 1999–2010, backs aged 28 to 30 ran about a fifth of a yard a carry worse than backs aged 22 to 25. In 2014–2025 they ran about a quarter of a yard worse. The two eras differ by about five hundredths of a yard, and luck alone could easily make a gap that size.
Yards alone can mislead. A 3-yard run on 3rd-and-2 keeps the drive alive. The same 3 yards on 3rd-and-10 brings out the punter. Expected points, a stat that scores the first run higher, gives the same answer. On it, the older backs trailed by a hair in both eras, and the difference between the eras could be luck.
Backs 28 to 30 trail backs 22 to 25 by about the same in both eras
Difference in yards per carry against the league average, by era
The gap was 0.21 yards a carry in 1999–2010 and 0.25 in 2014–2025. The 95% ranges overlap mostly, so the decline did not get clearly gentler or steeper.
Each bar shows how far the older backs trailed in one era, and the two are nearly the same length.
Following the same backs gives the same answer, with one hint
Those averages flatter the old backs. They count only backs who kept their jobs. A 30-year-old cut after a bad year drops out. A Gore, still carrying at 37, stays in. So this check follows each back from one season to his next.
Between 27 and 30, a back lost about a tenth of a yard a carry a year in 1999–2010 and about a fifth of a yard in 2014–2025. Those are small groups, 166 pairs of back-to-back seasons in the early era and 104 in the recent one, so that gap could be luck.
Backs aged 22 to 25 dropped too, about a fifth of a yard a year in both eras. That is more than you would expect from backs in their prime. A back needs 50 carries in both seasons to be in this check. The backs who earn the second season have often just had a big first one. A big season is hard to repeat at any age. So the fair test is the same age group in two eras.
| Ages | Era | Change per season | Likely range | Pairs of seasons |
|---|---|---|---|---|
| 27–30 | 1999–2010 | −0.09 | −0.21 to 0.04 | 166 |
| 27–30 | 2014–2025 | −0.20 | −0.36 to −0.05 | 104 |
| 22–25 | 1999–2010 | −0.18 | −0.28 to −0.08 | 246 |
| 22–25 | 2014–2025 | −0.18 | −0.27 to −0.09 | 311 |
Every row loses ground. Backs 22 to 25 lost almost exactly the same in both eras. For 27 to 30 the rows differ, but their ranges overlap.
On expected points, recent backs aged 27 to 30 did fade a little faster, by more than luck usually makes. Measured against young backs’ own change, the gap is about the same size but less certain, so we wouldn’t lean on it.
Fewer backs reached 27 with 100 carries
The fade barely changed, so the older backs went missing before 27. In the 12 seasons from 1999 to 2010, 69 backs had 100+ carries at 27, about 6 a season. In the 10 seasons from 2014 to 2023, only 39 did, about 4 a season. The count stops at 2023 because a back who was 27 in 2024 or 2025 has not yet played his season at 29.
The backs who did get there kept going about as often as before. In 1999–2010, 46.4% still had 100+ carries at 29. In 2014–2023, 41.0% did. With this few backs, that gap is noise. Some of the drop may be teams choosing younger backs, and some may be supply, from a deep draft class or a veteran’s injury. These numbers can’t split the two.
The table puts two seasons side by side. In 2008 backs 28 and older got 30.9% of the carries, led by LaDainian Tomlinson with 292. 2025’s top eight look a lot like 2008’s at the top, led by McCaffrey and Henry. Further down, 2008 had 17 backs aged 28 or older with 100+ carries; 2025 had 9. The 2025 names are part of the comeback of the last three seasons, which gets its own section below.
| Season | Back | Age | Carries |
|---|---|---|---|
| 2008 | LaDainian Tomlinson | 29 | 292 |
| 2008 | Thomas Jones | 30 | 290 |
| 2008 | Jamal Lewis | 29 | 279 |
Continues below
| Season | Back | Age | Carries |
|---|---|---|---|
| 2008 | Brian Westbrook | 28 | 233 |
| 2008 | Justin Fargas | 28 | 218 |
| 2008 | Larry Johnson | 28 | 192 |
| 2008 | Warrick Dunn | 33 | 186 |
| 2008 | Derrick Ward | 28 | 182 |
| 2025 | Christian McCaffrey | 29 | 311 |
| 2025 | Derrick Henry | 31 | 307 |
| 2025 | Saquon Barkley | 28 | 280 |
| 2025 | Tony Pollard | 28 | 242 |
| 2025 | Kareem Hunt | 30 | 163 |
| 2025 | David Montgomery | 28 | 157 |
| 2025 | Aaron Jones | 30 | 132 |
| 2025 | Alvin Kamara | 30 | 131 |
The lasting fall began six years after the 2011 pay rules
Many fans blame the 2011 labor deal for the backs who never reach 27. It created a rookie wage scale that, as the former agent Joel Corry wrote for CBS Sports, “drastically reduced salaries for early first-round picks”. Draft picks sign four-year contracts, and for first-round picks only the team can add a fifth year.
Cheap rookie deals may have nudged teams toward young backs slowly, as those deals spread through rosters. Our numbers can’t check that. They do show when the drop came. Through 2010 the older backs’ share sat between about 21.4% and 33.5%. It dipped to 15.1% in 2012, the year after the deal, and recovered. Then it fell hard from 2017, to 12.4% in 2019 and 14.1% in 2020.
In season averages, older backs got about 27 of every 100 carries in 2005–10 (26.8%). They got about 22 in each of the next two stretches, 2011–13 (21.9%) and 2014–19 (21.5%), and about 16 since 2020 (16.5%). The drop that lasted arrived in 2017, six years after the deal.
Derrick Henry got the ball past 30, and kept running
Derrick Henry kept getting the ball anyway. In 2022, the season backs his age got their smallest share of the work, the league’s busiest running back was 28 years old. Henry carried 347 times for the Tennessee Titans that year, by the play-by-play count. Two years earlier, at 26, he had run for 2,027 yards on 377 carries.
On March 12, 2024, at 30, he agreed to a two-year deal with Baltimore, worth $16 million and up to $20 million with incentives, according to CBS Sports. He called it “a no-brainer for me.” That season he carried 325 times for 1,921 yards. StatMuse lists that as the most rushing yards in a season by a player aged 30 or older, ahead of Tiki Barber’s 1,860 in 2005.
Per carry, Henry’s 2024 was 1.55 yards better than the league’s running-back average. That is the best of the 136 seasons in our data by a back aged 28 or older with 200+ carries. In 2025, at 31, he was still 0.85 yards a carry better than average on 307 carries.
Even Henry’s great years came and went. In 2020 he ran about a yard a carry better than the league’s average back. In 2021 and 2022 he was barely above average. In 2024 he was about a yard and a half better again.
Picture a general manager whose back, at 28 or older, just had a great year. If years like that don’t carry over, the team has less reason to keep paying for the next one. Across all backs, they mostly don’t. Great seasons by backs 22 to 25 don’t repeat reliably either. On a 0-to-1 scale of how much one season predicts the next, where 0 means no link and 1 a perfect one, older backs score 0.13 and backs 22 to 25 score 0.21 (from 233 and 626 pairs of seasons). Those two scores are too close to call one lower than the other. That is a reason a team might give; these numbers can’t show that it is why teams moved on.
The comeback belongs to one draft class
Since 2023 older backs have won back a little work. They got 18.9% of the carries in 2023, and 19.1% on average over 2023–25, against 13.8% in 2020–22. That rise is about 5 more carries in every 100 (5.3). A single season has moved the share by that much or more six times since 2000, so a rise this size could be luck.
Most of that work went to one group of players. The 2017 draft class, which took about a quarter of all running-back carries in 2019, grew old together. In 2024 it took 62.8% of all carries by backs 28 and older, more than three in five, through Aaron Jones (Minnesota Vikings, 29, 255), Joe Mixon (Houston Texans, 28, 245), James Conner (Arizona Cardinals, 29, 236), Alvin Kamara (New Orleans Saints, 29, 228) and Kareem Hunt (Kansas City Chiefs, 29, 200).
In 2024 one draft class took more than three in five of the older backs’ carries
Largest single draft class’s share of carries by running backs 28 and older, each season
Show data
| Year | Largest class |
|---|---|
| 1999 | 36.8% |
| 2000 | 23.5% |
| 2001 | 27.2% |
| 2002 | 21.7% |
| 2003 | 27.0% |
| 2004 | 37.0% |
| 2005 | 35.0% |
| 2006 | 32.9% |
| 2007 | 33.5% |
| 2008 | 21.7% |
| 2009 | 19.1% |
| 2010 | 20.9% |
| 2011 | 32.8% |
| 2012 | 27.9% |
| 2013 | 30.0% |
| 2014 | 26.0% |
| 2015 | 42.0% |
| 2016 | 26.8% |
| 2017 | 19.9% |
| 2018 | 18.8% |
| 2019 | 23.7% |
| 2020 | 28.6% |
| 2021 | 32.9% |
| 2022 | 37.0% |
| 2023 | 49.3% |
| 2024 | 62.8% |
| 2025 | 43.3% |
- 2024: 2017 class
- Reference: 1999–2022 average (28.4%)
The line spikes in 2024, far above the usual. In an average season from 1999 to 2022 the biggest single class took 28.4% of the older backs’ carries, and the most any class managed was 42.0% (the 2008 class, in 2015). The 2017 class took 49.3% in 2023 and 43.3% in 2025. One big class turning 28 at once fits the timing. The data can’t say whether teams also changed their minds about older backs.
The backs on that 2023 call show up here too. Ekeler carried 179 times at 28 that season. Chubb had 102 at 28 in 2024. In 2025 McCaffrey, 29, carried 311 times and Barkley, 28, carried 280. Add Henry, and five of the six backs named in reports of the call appear among the busiest backs 28 and older.
At 28, a back still loses about a fifth of a yard
Backs of 28 to 30 still run about a fifth to a quarter of a yard a carry worse than backs of 22 to 25. A great year at that age says little about the next. And a back with 100+ carries at 27 has a little under even odds of 100+ at 29, much as in the 2000s.
In 2025 the youngest back to carry 250 times was Ashton Jeanty of the Las Vegas Raiders, 21, with 266 carries. He turns 28, the age Gore talked about, in December 2031.
Henry was 28 in 2022 and led the league in carries that year. He was on that call in 2023. In 2025 he carried 307 times for the Baltimore Ravens. At 31, he was the oldest back to carry 250 times.
How we counted
- The data is the league’s play-by-play record, collected by the free nflverse project. Carries are regular-season run plays, minus kneels, two-point tries and aborted snaps.
- Running backs are players listed at RB, HB or FB on that season's nflverse roster. Quarterback and receiver runs are left out.
- Age is whole years on September 1 of the season, from the birth date in the nflverse players file. A back who turns 28 in October counts as 27 all season.
- Per-carry numbers are measured against that season's league running-back average, so changes in league-wide rushing cancel out. Only backs with 50+ carries count.
- Expected points per carry credits each run by how much it changed the points the offense could expect to score, given the down, distance and field position before and after the play.
- Draft class is the draft year in the nflverse players file, or the rookie season for undrafted backs. Team is the team a back had most carries for that season.
- Shares, not counts, so the 17-game seasons since 2021 do not inflate anything, except the 250-carry count, which 17 games make easier to reach since 2021.
- “In our data” means the 1999–2025 regular seasons; play-by-play starts in 1999, so earlier careers are cut off (Gore’s run is complete, Emmitt Smith’s is not).
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: 322,293 running-back carries, 1,840 player-seasons with 50+ carries. Positions from each season's roster, ages and draft classes from the players file. Era comparisons use 95% intervals; the 2017-class share is read against the largest single-class share in every season; per-carry persistence is the season-to-season correlation for the same backs, and the two age groups' correlations are compared on Fisher's z scale. Carry counts come from play-by-play and can run a few below official totals. Descriptive history; no holdout because nothing is being predicted.
| Test | Result | Verdict |
|---|---|---|
| Share of RB carries by backs 28+: 31.5% (2008), 14.4% (2020), 18.8% (2023) | 30.9%, 14.1%, 18.9% | Held (within a point) |
| The age cliff did not move | 28–30 minus 22–25: -0.21 (−0.34 to −0.08) vs -0.25 (−0.38 to −0.13) yards per carry; late minus early −0.05 ± 0.18 | Mostly held; one measure borderline |
More tests below
| Test | Result | Verdict |
|---|---|---|
| Same, in expected points per carry | -0.01 vs -0.03 | Held: no difference beyond noise |
| Same backs, year over year, ages 27–30 | -0.09 (−0.21 to 0.04, n 166) vs -0.20 (−0.36 to −0.05, n 104) yards per carry per year; late minus early −0.11 ± 0.20. Ages 22–25: -0.18 (n 246) vs -0.18 (n 311) | Overlapping ranges |
| Same backs, year over year, ages 27–30, expected points | late minus early −0.038 ± 0.032; net of 22–25 −0.036 ± 0.038 | Borderline; not robust |
| Shift tied to the 2011 rookie wage scale (hunch) | 28+ share: 26.8% (2005–10), 21.9% (2011–13), 21.5% (2014–19), 16.5% (2020–25) | No step at 2011; not testable as a cause |
Continues below
| Test | Result | Verdict |
|---|---|---|
| Backs with 100+ carries at 27 still lead at 29 | 46.4% (n 69) vs 41.0% (n 39) | Noise |
| The 2023–25 rebound is real | 19.1% vs 13.8% (2020–22), a rise of 5.3 points; standard deviation of single-season moves 4.7 points; 6 of 26 single-season moves at least as big as the rise | Unproven; could be luck |
| The 2023–25 rebound is one draft class | 2017 class: 49.3%, 62.8%, 43.3% of 28+ carries; largest class 1999–2022 averaged 28.4%, max 42.0% | Held as timing; not a cause |
| An older back’s per-carry edge carries over | r 0.13 (n 233) for 28+; r 0.21 (n 626) for 22–25; difference 1.2 standard errors on Fisher's z | Weak in both groups; the two are not clearly different |
Continues below
| Test | Result | Verdict |
|---|---|---|
| Henry 2024 is the best per-carry season by a 28+ back with 200+ carries | 1.55 yards above the league RB average; best of 136 | Held, in our data (1999–2025) |
Frequently asked questions
At what age do NFL running backs decline?
From about 27. Backs aged 28–30 ran 0.25 yards a carry worse than 22–25 year-olds in 2014–2025, against the league average, and 0.21 worse in 1999–2010.
Do older running backs get fewer carries than they used to?
Yes. Backs 28 and older got 33.5% of running-back carries in 2003 and 12.3% in 2022. In 2025 it was 18.0%.
Did the running-back age cliff move?
Not clearly. The per-carry decline with age is about the same in 1999–2010 and 2014–2025 (backs aged 28–30 trailed 22–25 year-olds by 0.21 yards a carry, then by 0.25). One measure, same-back expected points, leans toward a slightly steeper decline recently, but it could be luck. What changed most is how many backs still have 100+ carries in their late 20s.
How many 30-year-old running backs had 250 carries?
18 from 1999 to 2006, 4 from 2007 to 2014, 5 from 2015 to 2022 and 2 from 2023 to 2025 (Derrick Henry twice).
Who ran for the most yards in a season after turning 30?
Derrick Henry, with 1,921 yards for Baltimore in 2024 at age 30, according to StatMuse. Tiki Barber’s 1,860 in 2005 is next.
Who is the oldest NFL running back with 100 carries in a season?
In our data (1999–2025), Frank Gore: 187 carries for the New York Jets in 2020, at 37.
Did the 2011 rookie wage scale end the veteran running back?
The data cannot say. The older backs’ share did not drop in one step at 2011: it dipped in 2012, recovered, and fell hardest from 2017. A slower effect is possible but untested here.
More in Endangered roles
Sources and method
- nflverse play-by-play, rosters and players (CC-BY 4.0)
- CBS Sports, "NFL's top running backs hold private meeting to discuss next steps in diminishing market, per report" (July 23, 2023)
- Tom Junod, ESPN, "Was Frank Gore the last NFL running back?" (republished by ABC7 New York, Dec 20, 2023)
- CBS Sports, Ravens sign Derrick Henry to a two-year deal (March 12, 2024)
- StatMuse, most rushing yards in a season by a player aged 30 or older
- Joel Corry, CBS Sports, "Agent's Take: Fixing the rookie wage scale, plus a look at its history and how it works" (April 24, 2019)
Research and analysis, not betting advice. Past results do not guarantee future returns; bet only what you can afford to lose.