clock menu more-arrow no yes

Filed under:

Statistical Review of the 2009 49ers: VI. Running Backs

New, comments

Welcome back for the sixth installment of my 2009 Niner stat review. If you missed Parts 1-5, you can read them here, here, here, here, and here.

Today in Part 6, I'll take a look at the RB position. And by "the RB position," I mean Frank Gore. Sure, the 49ers technically have Glen Coffee, Moran Norris, and Michael Robinson listed as RBs on the 53-man roster; but, as was evident during Gore's 3-game absence in 2009, the 49ers' run OFF screeches to a halt without him.

The fact that Gore is so integral to the Niners' OFF definitely bucks the current NFL trend, which increasingly reflects an affinity for the proverbial RB-by-committee (RBBC). And the irony, of course, is that Gore began his career in 2005 as an RBBC member alongside Kevan Barlow. Since then, the number of teams using a RBBC has increased from 10 to 14, while the number of RBs with 320 or more carries has decreased from 8 to 3.

Not only are true #1 RBs like Gore becoming rarer; so are #1 RBs who produce with Gore's consistency and longevity. As we all know by now, Gore set a franchise record in 2009 with his fourth 1,000-yard season, all of which he achieved consecutively (another record). Not only does this achievement put him in rarified San Francisco air; it puts him in very select company among his contemporaries. Namely, here is the list of RBs who've had over 1,000 carries and 4,000 yards with a single team since 2006:

That's it. 3 guys. And Gore is 1 of them. A 49er is one of them! Actually, not only is he 1 of them; he's the best of them in terms of yards per carry (4.76).

I know this is premature, but much of a player's consideration for the NFL Hall of Fame - for any HOF really - is his standing relative to his era. If that's the case, then thus far in Gore's era (2006-2009), he's a Top-3 RB. If he keeps it up, he might be the next 49er HOF'er, especially if the team actually makes the playoffs one of these seasons. Have you ever contemplated that (Gore in the HOF, not 49ers in the playoffs)? I know I haven't.

So, to sum up the intro, Gore's single-handedly the Niner run OFF; and it's a good thing considering how consistently productive he's been. Therefore, the question today is, "for how much longer can he be this good?"

After the jump, I'll detail why the question is important, and then attempt to answer it...

RB EFFICIENCY

Below are the rushing and receiving efficiency stats for Niner RBs in 2009, alongside those of the average playoff #1 RB. I've omitted the Saints' RBs because they're one of those RBBC teams (#1-#8 performance in bold; #25-or-below performance in italics):

Team

RB Run DVOA

Rk

RB Rec DVOA

Rk

Frank Gore

4.4%

20

5.7%

26

Glen Coffee

-35.1%

NA

-48.0%

NA

Playoffs

3.8%

23.0

-2.4%

32.8

(Housekeeping: NAs mean Coffee didn't have the 100 carries or 25 targets required to qualify for rankings.)

To me, 3 conclusions emerge from this table. First, Gore was nearly 40% more efficient than his primary backup. Good thing that injury only kept him out 3 games! Second, Gore's Run DVOA compares favorably to that of the average playoff #1 RB. That's a good thing, of course. Well, actually, maybe not, because...Third, there's no indication that having a feature RB with a top-20 DVOA meant anything in terms of making the playoffs this past season. Indeed, if you look at the 19 RBs ahead of Gore in Run DVOA, only Ray Rice (#6), Ryan Grant (#10), Marion Barber (#15), and Joseph Addai (#17) could be considered feature RBs on a playoff team. And, in the cases of Barber and Addai, they're within a stone's throw of RBBC territory.

Also, I shouldn't move on without mentioning that #1 RB Receiving DVOA is even less relevant than #1 RB Run DVOA when it comes to playoff teams. The highest-ranked feature RB among playoff teams was Adrian Peterson at #15 with a 15.6% RB Receiving DVOA. Granted he was targeted in the passing game (57) far less frequently than he carried the ball (314), but - believe it or not - Peterson was actually a more efficient receiver than runner in 2009.

RB RUSHING YARDAGE

So Gore was more efficient in 2009 than the average NFL RB, as well as the average playoff #1 RB; and that goes for both running and receiving efficiency. But how productive was he in terms of yardage per run and yardage per target? And just how unproductive was Coffee? Here's the rushing table:

Team

DYAR/R

Rk

Yds/R

Rk

EYds/R

Rk

Frank Gore

0.51

22

4.90

13

4.23

28

Glen Coffee

-1.02

NA

2.72

NA

1.66

NA

Playoffs

0.52

22.9

4.19

28.8

4.48

21.5

As was the case with his DVOA, Gore's DYAR-per-carry was once again very comparable to playoff #1 RBs. Where things get interesting, however, is the flip-flop that occurs when moving from yards-per-carry to EYds-per-carry. Specifically, whereas Gore was nearly a full yard-per-carry better than the average playoff #1 RB, he was worse than the playoff average in EYds-per-carry. Essentially, this means that opponents and game situations gave playoff #1s a 0.29-yards disadvantage on their average carries, but gave Gore a 0.67-yard advantage on his average carry. Or, if you prefer English, the bottom line is that Gore's 4.90 yards-per-carry stat is somewhat of a mirage insofar as he benefitted greatly from softer run DEFs and easier game situations.

As for Coffee, he was abysmally abysmal, which, of course, is the inverse of "tremendously tremendous" (hat-tip: Fooch). Just to make sure you put the stats in their proper perspective, Coffee's DYAR-per-carry of -1.02 means that, if Coffee was the 49ers' starter, and they had to replace him for whatever reason, an average backup would gain 1.02 more yards per carry than he did. However, being that Coffee is, in fact, a backup, his negative DYAR/R really just means that the Niners might want to find a new backup. OK, OK, he's just a rookie. Fine, I'll rephrase. The Niners should not assume they've found their Gore backup.

RB RECEIVING YARDAGE

Here's the receiving table:

Team

DYAR/T

Rk

Yds/T

Rk

EYds/T

Rk

Frank Gore

1.08

27

5.41

31

6.20

27

Glen Coffee

-1.89

NA

4.22

NA

0.06

NA

Playoffs

0.64

32.6

6.03

25.5

5.38

30.8

Interestingly enough, the pattern was reversed for Gore (and playoff #1s) in the passing game. Specifically, rather than helped by schedule and situation, Gore's productivity as a receiver was hurt by facing stingier pass DEFs and tougher game situations. And again, this contrasted with playoff #1s, who benefitted 0.65 yards-per-target from softer pass DEFs and easier game situations (i.e., 6.03 minus 5.38). Of course, as has been the case repeatedly so far, playoff participation didn't exactly require a team's #1 RB to be highly ranked in DYAR/T, Yds/T, or EYds/T.

As for Coffee, if he were actually an average backup, he would have averaged 6.11 Yds/T (i.e., 4.22 minus -1.89). In addition, nearly all of his actual Yds/T (i.e., 4.16 of 4.22) were attributable to facing soft pass DEFs and easy game situations.

THE (RED AND) GOLDEN GOOSE

OK, so far I think it's pretty clear that the 49ers would be doomed if Frank Gore were to suffer a major injury that forced him to miss significant playing time. But what are the odds of that happening? Or, using more optimistic language, what are the odds of that not happening? Well, for this discussion, let's return (finally) to the question I posed at the beginning of the post: For how much longer can Gore be this good?

It turns out that there are several theories about RB longevity in general that may or may not apply to Gore's specific case. The first one is based on research done by Doug Drinen at Pro Football Reference back in 2000. Simply looking at whether or not RBs improve from Age N to Age N+1 (e.g., from Age 24 to 25), he found that the peak occurs somewhere around Age 27. In discussing his finding, Drinen speculated that physical wear-and-tear was not a major factor in the age-related decline for RBs.

Three later theories about RB decline speculated in the opposite direction. The most notorious of the 3 is Football Outsider's Curse of 370, which was developed in 2004 and states that, "a RB with 370 or more carries during the regular season will usually suffer either a major injury or loss of effectiveness the following year, unless he is named Eric Dickerson." The basic idea underlying FO's theory is that, rather than any particular age signaling the likely onset of RB decline, it's actually 370 carries at any age that sounds the death knell. Later studies by Brian Burke at Advanced NFL Stats and Maurile Tremblay at Football Guys both found that the Curse of 370 was more aptly named the Myth of 370, i.e., they by-and-large debunked it as too general insofar as (a) 370 is not a magic number, and (b) football stat analysis could never come up with a magic number to begin with.

Almost simultaneous with FO's unveiling of the Curse, Football Docs published basically the same exact analysis (coincidence?), except they added another component. Namely, they proposed that, in addition to 370 carries being informative vis-à-vis a RB decline, so was a 33% year-to-year increase in carries. In other words, if a RB carries the ball 33% more in Year N than Year N-1, then he's likely to experience a decline in rushing yards the following year (i.e., Year N+1). Football Docs work seems far more anecdotal than statistical, but the theory does seem to be as defensible as FO's (See debunking caveats above).

The final wear-and-tear-based theory was forwarded in 2009 by Dave Richard at CBS Sports. Another anecdote-intensive effort, Richard's theory states that a RB is likely to decline when he reaches 8 full seasons and/or 2,500 career carries. Unfortunately, the jury's still out given that (a) the theory is relatively new, (b) there don't seem to be any hardcore stat analyses to back it up yet, and (c) Richard doesn't go into great detail about how he came up with 8 & 2,500 as the benchmarks.

It should seem obvious from this discussion that each of the theories are flawed to various extents. Nevertheless, let's just assume - for the sake of evaluating Gore - that all of them are right. If that's the case, is Gore likely to decline any time soon according to any of the theories? Well, here are the relevant stats:

RB Characteristic

N

Gore

Decline?

Age > N

27

26

No

Carries in Previous Season > N

370

229

No

% Change in Carries > N

+33%

-4.6%

No

# of Full Seasons > N

8

4.5

No

Career Carries > N

2,500

1,168

No

According to Drinen's age-based theory, Gore has 1 or 2 good seasons left in him. According to the Curse of 370, Gore wasn't even close to 370 carries in 2009, and wasn't even that close in 2007, when he was coming off of a 312-carry 2006 season. Ditto for Football Docs' increased workload theory: Gore actually had fewer carries in 2009 than he did in 2008. Of course, looking back at Gore's underachieving 2007 season, he may have indeed suffered from a 145% increase in workload from 2005 to 2006. Finally, according to Richard's theory, Gore seems to have about half-a-career's-worth of good seasons left.

So, given that our conclusion from Drinen's theory seems to be out of step with the other three, and given that his theory was based on the least-recent data set (i.e., ends in 2000), I'm inclined to lean towards the, "Frank Gore has plenty of good seasons left," conclusion clearly suggested by FO's and Richard's theories, and mildly supported by Football Docs' theory.

With the esteemed theorists accounted for, I'll finish up with what I think is a more-contextual look at Gore's potential career trajectory. And before I discuss it, I want to say that I never realized just how good Gore has been over the past 4 years in relation to his peers, or just how good he has the potential to be from a historical perspective. Regarding what I'm about to show you, I honestly had no idea what I was going to find when I started doing the research. Needless to say, I was pleasantly surprised, and have come to feel that Gore is just about the most underrated and underappreciated RB over the past 4 seasons.

OK, so for my contextualized analysis, I'm going to return to what I said at the beginning of the post. Namely, that Gore's yardage and carry totals over the past 4 seasons put him in a select, elite class of NFL RBs during his era. What I neglected to tell you until now is that his penchant for 1,000-yard seasons - at such a young age - puts him in an elite class of NFL RBs during the modern passing era.

The first part of the Gore story involves determining the ages in which 1,000-yard seasons tend to occur based on recent NFL history.* I've got a handy graph below:

 2009_49ers_season_recap_--_1k-yd_seasons_by_rb_age__1977-2009__medium

What you see pretty clearly in this graph is that RBs tend to be in their 1,000-yard-season prime between the ages of 24 to 27. Indeed, about 55% of all 1,000-yard seasons over the past 33 seasons have been achieved by RBs in this age range. So, already we see that Gore, whose 1,000-yard seasons came from Ages 23 to 26, is actually one year ahead of the curve. He already has four 1,000-yard seasons, and he's likely got 1-2 years still remaining in his 1,000-yard prime. That's the first clue about his longevity.

Of course, given that Gore became great earlier in his career than most, and that he's already had more 1,000-yard seasons than most, it's less than ideal to compare him to RBs like Cleveland Gary and Olandis Gary or Gaston Green and Harold Green, or former Laker Karim Abdul-Jabbar - all of which only had one 1,000-yard season in their entire careers. So let's compare Gore to the RBs since 1977 who are/were just like him.

Below I've listed the 15 RBs other than Gore and Jackson (both haven't yet reached 27) who had 4 consecutive 1,000-yard seasons from Ages 23-26. As a way to identify the most likely Gore outcomes, I've also included various measures of what happened to these RBs after they turned 26:

 

RBs w/ 4 Consecutive 1k-yd Seasons at Ages 23-26 (1977-2009)

RB

1k-yd Seasons

After Age 26

Age at Final

1k-yd Season

Notes

EMMITT SMITH

6

32

"Drop-off" began at 27

WALTER PAYTON

5

32

Lost 28 for 1982 strike

Curtis Martin

5

31

Best season at 31

TONY DORSETT

4

31

Lost  28 for 1982 strike

BARRY SANDERS

4

30

2k-yd season at 29

THURMAN THOMAS

4

30

"Drop-off" began at 27

ERIC DICKERSON

3

29

Best season at 28

Corey Dillon

3

30

Best season at 30

LaDainian Tomlinson*

3

29

Best season at 27

Eddie George

3

30

Best season at 27

Ahman Green

2

29

Lost 1 for inj at 28

EARL CAMPBELL

1

28

Lost 27 for 1982 strike

Ricky Williams*

1

32

Susp at 27; Retire at 29

Terrell Davis

0

26

Torn ACL, MCL at 27

Rodney Hampton

0

26

Mystery

BOLD CAPS = HOFer

Bold = Retired, but not yet eligible for HOF

* = Active player

OK, first, WOW, that's some seriously awesome company! Nearly half (7 of the 15) are already in the HOF. Two others (Martin and Tomlinson) are shoe-in HOFers. Three others (Dillon, Green, and Williams) are borderline HOFers who aren't eligible yet. And finally, two others (Davis and George) are borderline HOFers who've only recently become eligible, and who may be voted into the HOF on future ballots. So, among the 15 RBs on the list, only 1 (Hampton) can be seen as having essentially no shot to make the HOF, which isn't altogether shocking given that his career basically ended for no apparent reason after his 4th 1,000-yard season in 1995. At least Davis had the excuse of a major knee injury for his failure to achieve a 5th. In this context, Gore - and Jackson for that matter - will quickly approach HOF worthiness with 2 or 3 more 1,000-yard seasons.

And how likely is that? Well, based on the above table, a RB who has four 1,000-yard seasons by the Age of 26 is a 3-to-2 favorite to have at least 2 more! And if you add Campbell's strike-shortened 1982 season, in which he was easily on pace to have 1,000 yards at Age 27, the likelihood improves to 80%. In other words, although there have been 440 individual 1,000-yard seasons by a RB since 1977, having 4 consecutive from Ages 23 to 26 is what seem to separate the HOFers from the flashes-in-the-pan. Indeed the 17 "4 x 1,000" RBs (which includes Gore and Jackson) account for 119 of the 440 total 1,000-yard seasons since 1977! And given that those 440 seasons were posted by a total of 166 RBs, this basically boils down to the top 10% of RBs accounting for 27% of all 1,000-yard seasons since 1977!

The point I'm trying to make here - if it's not already evident - is that longevity seems to breed longevity among RBs. The more 1,000-yard seasons a RB accumulates early in his career, the more likely he will accumulate additional 1,000-yard seasons later in his career. Furthermore, the table suggests there's an 80% likelihood that a 26-year old "4 x 1,000" RB won't have has final 1,000-yard season until he's at least 29 years old.

So, taking all the evidence together, this is what we can infer about Gore's potential longevity:

  • Gore's had four 1,000-yard seasons by Age 26
  • The prime years for a 1,000-yard RB extend to Age 27-28
  • The likelihood that Gore will have at least 2 more 1,000-yard seasons is 80%
  • The likelihood that Gore's final 1,000-yard season won't come until he's at least 29 years old is 80%

Therefore, I'd say that - barring a major injury - Gore's got a better than 50/50 shot to have 3 more 1,000-yard seasons by the time he turns 30. And, if that's indeed what happens, Gore becomes a virtual certainty to be enshrined in Canton 5 or so years after his career ends.

BOTTOM LINE

As I said previously, I really had no idea that Gore's sustained level of production is as rare as it is. Perhaps that's because the Niners have been so awful from the moment he arrived, and because great RBs on awful teams tend to be underappreciated nationally. But, in my mind, it's precisely these types of RBs - those on awful teams - that are the most impressive because they're typically running behind bad OLs and they don't benefit from the clock-killing, stat-padding 4th-quarter carries enjoyed by RBs on winning teams.

So, based on the stats I've presented in Part 6 of the season review, here are the things the Niners need to do at the RB position in order to seriously contend in 2010:

  1. Do whatever's humanly possible to keep Gore healthy
  2. Continue to search for an efficient Gore backup in partial fulfillment of #1 above

If they can sustain Gore's production for another 3 or 4 seasons, they'll have earned him a bust in Canton alongside a certain LB that I'll talk about on Friday.

*My choice of 1977-2009 was not some arbitrary decision. Rather, it's based on research by Stimel (2009), which showed that the variability in passing statistics exhibited a structural breakpoint in 1977, such that 1977-present represents a singular high-performance, low-variability passing era of the NFL; presumably due to significant rule changes in the mid-to-late 1970s that made life far more difficult for defensive backs. Therefore, 1,000-yard seasons by a RB prior to 1977 are oranges compared to the apples from 1977 to the present.

**DVOA, DYAR, and EYds statistics used to produce this article were provided by Football Outsiders.

Poll

Be as unbiased as possible here: Frank Gore now has four 1,000-yard seasons in his career. How many MORE1,000-yard seasons do you think Frank Gore needs before YOU WOULD VOTE HIM into the NFL Hall of Fame?

This poll is closed

  • 1%
    None. The 4 he already has makes him a HOFer in my mind.
    (8 votes)
  • 31%
    1 or 2 more and I'd consider him a HOFer.
    (144 votes)
  • 64%
    3+ more and I'd consider him a HOFer.
    (295 votes)
  • 1%
    Infinity. I'll never consider him a HOFer.
    (9 votes)
456 votes total Vote Now