SITE NEWS: We are moving all of our site and company news into a single blog for Sports-Reference.com. We'll tag all PFR content, so you can quickly and easily find the content you want.

Also, our existing PFR blog rss feed will be redirected to the new site's feed.

Pro-Football-Reference.com ยป Sports Reference

For more from Chase and Jason, check out their work at Football Perspective and The Big Lead.

2007 and the Workhorse Running Back: Trend or Anomaly?

Posted by Jason Lisk on October 17, 2007

If it seems like the top running backs are rushing the ball fewer times so far this year, it's because they are.

You would have to go back a decade to see as few high carry performances through the first six weeks. The two highest carry games so far this year are Larry Johnson with 31 on Sunday against Cincinnati, and Steven Jackson with 30 in his final game played so far this year. The last season in which no running back had at least 33 carries in a game in the first six weeks was 1996 (Thurman Thomas, 32 attempts, week 1).

The number of games where a running back had at least 27 carries in a single game is at its lowest rate for all the years available. Meanwhile, the percentage of games with a running back between 15 and 20 attempts is at its highest rate.

Here are the individual game, individual running back rushing attempt patterns for the years which we have game by game data (1995-2007), through the first six weeks of each season:

Year     33+  30-32  27-29  24-26  21-23  18-20  15-17     Tm Gms
===================================================================
2007      0     2      7     21     23     39     44         178
2006      1     5      9     19     29     35     36         174
2005      2     3      9     18     31     35     32         176
2004      4     7      5     18     38     30     35         176
2003      4     3      9      9     26     23     39         176
2002      1     7      7     16     24     26     37         176
2001      2     6      5     20     25     31     35         170
2000      3     5      9     15     27     22     32         172
1999      1     5     10     15     32     29     35         172
1998      3     6     12     15     35     32     25         162
1997      5     3     11     15     17     22     38         164
1996      0     3      9     17     18     25     30         162
1995      1     3      5     15     19     34     35         166
===================================================================

While there was a slight decline in the last two years from the peak of 2004, the change in running back usage this year is more dramatic. So, is it a temporary anomaly, or a trend?

Here are some of my thoughts:

RUNNING BACK WORKLOADS AND COMMITTEES

Clearly, the biggest reason for this change is that the NFL coaches are avid readers of the PFR blog, and the league has read the earlier posts on running back overuse and injuries.

Of course, I'm joking. But there has been increased focused on the issue of running back workloads in recent seasons, including pieces published in Pro Football Prospectus 2006 and 2007, and on Football Outsiders.com. Several teams in recent years have employed more of a shared carries scenario. Up until this year, however, it seemed that it was a subculture, and the workhorse running back philosophy was still predominant. Is it simply a matter of a copy cat league, following the footsteps of teams that reached the championship game last year? At this point, I think it is debateable whether it is a philosophical shift among a majority of the league, or whether the change so far in 2007 is out of necessity.

INJURIES

Three things in life are certain: death, taxes, and fantasy football players talking about how this year is a crazy year for injuries. But are early injuries to blame for this decline in high carry games, and is the injury rate that different? The injury rate has been approximately 25%, for injuries in the first 6 games of the following season, for backs who did not miss a game from weeks 12-17 the previous year, and averaged 15.0 or more attempts (7% suffered a season ending injury in the first 6 games).

So far this year, 7 of the 17 running backs (41.2%) who met those criteria have missed at least one game with injury--one is a season-ending injury (McAllister), and another has that potential (Jackson) depending on whether the Rams want to risk bringing Jackson back later this year on a currently 0-6 team. The injury rate for this year's group is a little higher, but with the sample size, not enough to make any definite statements that 2007 is substantially different than previous years.

In fact, if we were to expand the criteria to all backs that were healthy at the end of 2006, and averaged at least 10.0 attempts per game, we bring in 9 more running backs, all of whom have played in every game--several of whom barely missed the previous cutoff. I have little reason to believe that the injury rates among the 10.0 to 14.9 group is substantially different from the 15.0 to 16.9 group or the 17.0 to 18.9 groups. 7 out of 26 otherwise healthy backs at the end of 2006 having an injury causing missed game(s) so far in 2007 would be roughly in line with recent history.

However, I do think injuries have played a role, just in a different way. Which brings me to . . .

THE LACK OF HEALTHY STAR RUNNING BACKS OF PEAK AGE

This chart shows the chronological age of the Top 12 Running Backs (by Standard Fantasy Points) for each season going back to 1990:

Year      21   22   23   24   25   26   27   28   29   30+
=========================================================
2006       1    0    3    0    0    1    6    0    0    1
2005       0    1    0    1    0    3    3    1    0    3
2004       0    0    2    1    3    1    1    0    1    3   
2003       0    1    0    2    3    3    1    0    1    1
2002       1    0    2    2    2    1    1    0    2    1
2001       0    2    0    3    0    0    2    3    1    1
2000       0    1    1    1    0    1    4    2    0    2 
1999       1    0    0    1    2    3    1    1    1    2
1998       0    1    0    1    3    3    1    0    2    1
1997       0    0    2    3    4    0    1    1    1    0
1996       0    1    2    3    0    1    2    2    0    1
1995       0    2    1    0    1    3    4    1    0    0
1994       1    1    0    0    3    4    1    2    0    0
1993       1    0    1    6    1    0    1    0    0    2
1992       0    0    3    4    0    3    0    0    0    2
1991       0    2    1    0    1    1    3    0    3    1
1990       1    1    1    5    1    1    0    1    0    1   
TOTAL      5   13   19   33   24   29   32   14   12   22 
=========================================================

Between 1990-2005, 47% of the Top 12 Running Backs were between ages 24-26. 2006 was the only year during this period where there were no 24 or 25 year old backs in the Top 12, and there was only one 26 year old (Willie Parker). And it's not as if the cutoff at the Top 12 is excluding some near misses. Kevin Jones was the top 24 year old last year (by fantasy points) at 23rd. Ronnie Brown was the top 25 year old, at 25th. Willie Parker was the only 26 year old in the top 30 in running back fantasy points.

Why do I think this is significant? Because the running backs between ages 24 and 26 in one season become the 25 to 27 year old running backs the next season (don't try that kind of advanced math at home). Here are the injury rates of the running backs from my previous study with 138 or more (23.0 per game) attempts through the first 6 games of a season, sorted by age. The number given is the number who missed 5 or more games of the final 10, after playing in the first 6.

24 years old or under: 3 of 11
25-27 years old: 1 of 17
28-30 years old: 0 of 3
31 years or older: 2 of 4

And here are the running backs from the 126-137 attempt group through first six games, sorted by age, and number who missed at least 5 of the remaining games:

24 years old or under: 2 of 12
25-27 years old: 0 of 17
28-30 years old: 1 of 12
31 years or older: 0 of 1

If you look at the high workload group (138 or more attempts), the 25-27 year olds make up almost half of the running backs starting the season with a high workload from 1995-2006. The number of 28 to 30 year olds starting a season with high workloads drops dramatically.

It is that 25 to 27 year old group that is in large part absent in 2007. Some of this is due to what were considered relatively weaker draft classes about four to five years ago at the running back position. But I think some of it also has to do with recent usage patterns of young running backs.

Ronnie Brown is emerging this year as a top back at age 26, and has not seen high workload games early in his career. Willie Parker, who emerged as a full-time back last year at age 26, is likely to be the only back joining the high workload group (121 attempts through first five games). Most other backs in this age group have not made it to 2007 without significant injury.

Clinton Portis would be the standard bearer for this group, but missed 8 games last year, and is sharing the workload in Washington. Kevin Jones, Julius Jones, and Cadillac Williams all looked very promising their rookie seasons, but also had periods of high rushing attempt games. Were they just lucky their rookie years, or did they get physically worse due to the early career usage pattern? Another back who would be part of this age group, Domanick (Davis) Williams, now 27, was a promising young running back for the Texans, but as noted in my earlier post, did not survive high workload games in 2004, and probably will never play a meaningful role as a starter in the NFL again. We will have to see if Steven Jackson, at age 24, is the latest victim, as he had a three game stretch of high workload games at the end of 2006, and suffered the groin injury in a game with 30 attempts earlier this year.

I'm not sure you can answer with certainty whether early workloads ruined these backs or not, but the fact remains that none of them did make it to 2007 healthy enough to carry a full workload, when they should be in their prime. In my opinion, this has contributed to a perfect storm in 2007, leading to the drop in high carry games. So, trend or anomaly?

It largely depends on what teams do with the current group of young running backs over the next few years. Until now, Joseph Addai, Laurence Maroney, Maurice Jones-Drew, Adrian Peterson, Marshawn Lynch, and Marion Barber III have not approached 30 carry games, and most have been part of platoons to this point in their careers. It will be interesting to see if the majority of these players make it to their mid-20's fairly healthy or not, and how teams choose to use them over the next few years.

In conclusion, I will note that high workload games of 27 attempts or more show a positive correlation with the week of the season, and increase as the year progresses. The correlation coefficient between the high workload games and numerical week of the season is +0.61. So, I expect that the high workload games will increase by total number, but it will be interesting to see if they increase to the level of the previous decade or not, over the remainder of 2007.

This entry was posted on Wednesday, October 17th, 2007 at 5:59 am and is filed under General. You can follow any responses to this entry through the RSS 2.0 feed. Both comments and pings are currently closed.