Writers of Pro Football Prospectus 2008

Most Recent FO Features

lionspackers.jpg

» Week 17 Game Previews

The FO crew previews three games in the final week of the season with critical playoff implications: Bengals-Steelers, Lions-Packers, and Panthers-Falcons.

30 Jun 2005

A Different Look at the Best Running Backs Ever

by Ryan Wilson

As Jamal Lewis approached Eric Dickerson's all-time single-season rushing record two seasons ago, I got to thinking about how impressive this performance was in both an historic sense, and in the context of the 2003 season. When having the "greatest running back of all time" discussion, is there a difference between a player who dominates (like Dickerson in 1984) and one who leads the league in rushing in a relatively tight race (like Priest Holmes in 2001)? Of course, I didn't get around to answering this question until now, but I figure better late than never.

If you go to NFL.com and check out the rushing statistics, you'll have no trouble finding which players lead the league in categories like total rushing yards and yards per carry. But what you won't find at NFL.com are statistics that tell you how much better a back was than his competitors in a given season (or even over the course of their career). Let me explain what I mean.

In 2001, Priest Holmes led the league in rushing yards with 1,555. Now 1,555 yards falls well short of the single-season record of 2,105, but how much better is 1,555 yards in 2001 when compared to how the other running backs did that year? Likewise, how much better is 2,066 yards in 2003 when compared to how the other running backs did that year? Or in broader terms, what backs, over the course of their careers, were marginally more productive than their peers? The methods I describe below allow you to compare players across both seasons and careers.

Historically, football statistics have focused primarily on averages (average yards per game, average yards per carry) and sums (total rushing yards, total rushing TDs). Who is tenth on the all-time rushing list? Look it up and you'll find that it's Franco Harris (12,120 yards). Who was tenth for the 2003 season? It was Ricky Williams. (he rushed for 1,372 yards)

But here's the question I wanted to answer: How much better is Jamal Lewis than Ricky Williams in 2003, or Jerome Bettis than Thurman Thomas over the course of their careers, when compared to all other running backs? Now this required a little more work than opening the football almanac, but thanks to our good friends at Pro-Football-Reference.com, I was able comb through the data.

(Disclaimer: This article looks only at total rushing yards over seasons and careers; this is not to diminish the importance of receiving yards by running backs. Instead, in this first iteration, I'm interesting in looking solely at pure rushers -- remember, Jamal Lewis's 2,000 yard season was the impetus for this research, and he's not known for his ability to catch passes coming out of the backfield. In the future I'll take a look at how including receiving yards changes the list.)

Looking at all the rushing data from 1957 to 2003, I normalized rushing yards to something called z-scores. What is a z-score? It's a statistical measure that tells how a single observation (Jamal Lewis rushing for 2,066 yards in 2003, for example) compares to all other observations (all running backs for the 2003 season). A z-score says not only if the observation is above or below the average value, but also how unusual it is (and it's no surprise that Lewis' rushing total would be considered unusual using z-scores).

In general, a z-score of 1.0 means that a player is in the top 16% of all rushers, a z-score of 2.0 means that a player is in the top 2.5% of all rushers, and a z-score greater than 3.0 means that a player is in the top 0.15% of all rushers.

In addition to looking at how running backs compared by season, I also looked at how running backs compared over their careers. This lets us answer questions like: "How did Jim Brown, who played nine seasons, compare to Barry Sanders, who played ten seasons?"

According to standard rushing yards, here are the top single-season rushing performances since 1957:

Rank Last Name First Name Year Team Season
Rush Yds.
1 Dickerson Eric 1984 Rams 2,105
2 Lewis Jamal 2003 Ravens 2,066
3 Sanders Barry 1997 Lions 2,053
4 Davis Terrell 1998 Broncos 2,008
5 Simpson O.J. 1973 Bills 2,003
6 Campbell Earl 1980 Oilers 1,934
7 Green Ahman 2003 Packers 1,883
8 Sanders Barry 1994 Lions 1,883
9 Brown Jim 1963 Browns 1,863
10 Williams Ricky 2002 Dolphins 1,853
11 Payton Walter 1977 Bears 1,852
12 Anderson Jamal 1998 Falcons 1,846
13 Dickerson Eric 1986 Rams 1,821
14 Simpson O.J. 1975 Bills 1,817
15 Dickerson Eric 1983 Rams 1,808
16 Smith Emmitt 1995 Cowboys 1,773
17 Allen Marcus 1985 Raiders 1,759
18 Davis Terrell 1997 Broncos 1,750
19 Riggs Gerald 1985 Falcons 1,719
20 Smith Emmitt 1992 Cowboys 1,713
21 James Edgerrin 2000 Colts 1,709
22 Martin Curtis 2004 Jets 1,697
23 Campbell Earl 1979 Oilers 1,697
24 Alexander Shaun 2004 Seahawks 1,696
25 Foster Barry 1992 Steelers 1,690

No surprises here. Now what happens if we revisit the single-season rushing leaders, but instead of measuring success by total yards gained, we instead measure success by how many more yards a player gains when compared to other players for a given year? The top three rushing yardage seasons no longer rank one, two, and three. Looking at the z-score table, Dickerson is now 9th, Lewis drops to 28th, and Sanders is 8th.


Rank Last Name First Name Year Team Season
Rush Yds.
Z-score
Rush Yds.
1 Simpson O.J. 1973 Bills 2,003 4.00
2 Simpson O.J. 1975 Bills 1,817 3.55
3 Payton Walter 1977 Bears 1,852 3.40
4 Campbell Earl 1980 Oilers 1,934 3.30
5 Brown Jim 1963 Browns 1,863 3.18
6 Sanders Barry 1994 Lions 1,883 2.99
7 Brown Jim 1965 Browns 1,544 2.94
8 Sanders Barry 1997 Lions 2,053 2.89
9 Dickerson Eric 1984 Rams 2,105 2.84
10 White Charles 1987 Rams 1,374 2.84
11 Dickerson Eric 1986 Rams 1,821 2.74
12 Armstrong Otis 1974 Broncos 1,407 2.71
13 Kelly Leroy 1968 Browns 1,239 2.68
14 Smith Emmitt 1995 Cowboys 1,773 2.62
15 Davis Terrell 1998 Broncos 2,008 2.62
16 Smith Emmitt 1991 Cowboys 1,563 2.55
17 Nance Jim 1966 Patriots 1,458 2.54
18 Dickerson Eric 1984 Rams 1,808 2.51
19 Sanders Barry 1991 Lions 1,548 2.49
20 Brown Jim 1964 Browns 1,446 2.49
21 Brown Jim 1961 Browns 1,408 2.40
22 Smith Emmitt 1993 Cowboys 1,486 2.39
23 Simpson O.J. 1976 Bills 1,503 2.39
24 Allen Marcus 1985 Raiders 1,759 2.38
25 Campbell Earl 1978 Oilers 1,450 2.37

Yes, O.J. Simpson now has the top rushing season of all time -- and the second-best season as well. How did these seasons which began 5th and 14th end up on top?

Well, remember, z-scores look at how well a player does in relation to other players during a particular season -- not just total yards gained. And while Dickerson holds the all-time rushing record of 2,105, Walter Payton that same year finished second with 1,684 yards and James Wilder was third with 1,544 yards (somewhere Carl Prine is smiling).

On the other hand, when O.J. Simpson rushed for 2,003 yards in 1973 John Brockington (remember him?) was second with 1,144 yards and Calvin Hill was third with 1,142 yards. So the difference in yards gained between Dickerson and Payton (421 yards) was a lot closer than the difference between Simpson and Brockington (859 yards). It's also important to remember that before 1978, the NFL had a 14-game regular season schedule. This means that performances by guys like Jim Brown and O.J. Simpson are even more impressive when compared to backs who had two more games to rack up yards. But that's the beauty of z-scores -- they allow you to compare a player's performance based on the performance of every other player during a particular season, no matter how long that season was.

Compare O.J. Simpson's 1975 performance (1,817 yards) to Jamal Anderson's 1998 performance (1,846 yards). We find that even though both players rushed for over 1,800 yards, Simpson ranks 2nd using z-scores and Anderson ranks 36th; using single-season rushing yards, Simpson ranks only 14th while Anderson is 12th.

(And yes, sitting at #10 on the list is that Charles White -- the guy who dominated the scabs during the strike year. I thought about actually removing him from the list, but when was the last time White got some pub?)

Seeing these lists, I know the question you are asking: "Jamal Lewis ran for 2,066 yards in 2003, and he's not even in the top 20 when using z-scores. What gives?"

Walter Payton, who in 1977 rushed for fewer than 2,000 yards (1,852), ranks 3rd using z-scores but only 11th using single-season rushing yards. That's because Payton was the only runner in 1977 to gain more than 1,300 yards. In comparison, Lewis was one of twelve players to clear 1,300 yards rushing in 2003.

Using this same reasoning, we can see why Terrell Davis fell from 4th on the single-season rushing list to 15th on the z-score list; and why Edgerrin James fell from 21st to 66th -- there were many more players closer to Davis' 2,000 rushing yards in 1998 (and a lot more players closer to James' 1,709 total in 2000) and as a result their respective z-scores are lower (2.62 and 1.84, respectively) than they would have been in 1977.

All of the z-scores are extremely high, but that's to be expected because we're looking at the top of the list. If we were looking at every player who ever ran the football during an NFL game, we'd also see some z-scores around 0.0 as well as some negative z-scores.

Next, let's look at the all-time career rushing leaders (as of February 2005):


Rank Last Name First Name Career
Rush Yds.
Rank Last Name First Name Career
Rush Yds.
1 Smith Emmitt 18,355 11 Thomas Thurman 12,074
2 Payton Walter 16,726 12 Faulk Marshall 11,987
3 Sanders Barry 15,269 13 Riggins John 11,352
4 Martin Curtis 13,366 14 Simpson O.J. 11,236
5 Bettis Jerome 13,294 15 Watters Ricky 10,643
6 Dickerson Eric 13,259 16 George Eddie 10,441
7 Dorsett Tony 12,739 17 Anderson Ottis 10,273
8 Brown Jim 12,312 18 Dillon Corey 9,696
9 Allen Marcus 12,243 19 Campbell Earl 9,407
10 Harris Franco 12,120 20 Allen Terry 8,614

This also looks pretty familiar. But what happens if we use z-scores for career rushing leaders like we did for single-season rushing leaders above? To accomplish this, I summed each player's z-score for every season he averaged at least 10 carries per game (I'll call it the aggregated z-score). The aggregated z-score gives you an idea of how dominant a player was when compared to other players during his career.

Here's what the z-scored table looks like:

Last Name First Name Career
Rush Yds.
Z-score
Rush Yds.
Rush Rank* Z-score Rank
Sanders Barry 15,269 18.88 3 1
Payton Walter 16,726 17.31 2 2
Brown Jim 12,312 16.97 8 3
Smith Emmitt 18,355 14.13 1 4
Simpson O.J. 11,236 11.76 14 5
Dickerson Eric 13,259 10.62 6 6
Martin Curtis 13,366 9.06 4 7
Dorsett Tony 12,739 8.01 7 8
Thomas Thurman 12,074 7.64 11 9
Campbell Earl 9,407 7.33 19 10
Harris Franco 12,120 6.10 10 11
Davis Terrell 7,114 6.05 26 12
Kelly Leroy 6,734 6.02 28 13
Anderson Ottis 10,441 5.74 17 14
Dillon Corey 9,696 5.46 18 15
George Eddie 10,441 4.95 16 16
McCutcheon Lawrence 5,523 4.84 50 17
Tomlinson LaDainian 5,899 4.79 41 18
Riggs Gerald 6,241 4.63 36 19
Brown Larry 5,467 4.50 52 20
Andrews William 5,199 4.26 58 21
James Edgerrin 7,720 4.21 20 22
Sayers Gale 4,010 4.20 86 23
Alexander Shaun 5,624 4.16 48 24
Watters Rickey 10,643 4.15 15 25

Note: The rush rank in table 4 may vary slightly from the all-time rushing leaders because in the z-score analysis, seasons where players averaged fewer than 10 carries per game were dropped from the analysis.

What immediately stands out is that players like O.J. Simpson and Earl Campbell rank very high when using z-scores, but only rank 14th and 20th, respectively, when looking at all-time rushing yards. There's a straightforward explanation for these results. Both running backs played roughly a decade (Simpson played 11 seasons, Campbell played nine), and for most of their careers they dominated (both had five All-Pro type seasons). So even though both Simpson and Campbell declined as they approached retirement, the fact that they were so much better than their counterparts boosted their aggregated z-score.

On the other hand, a player like Franco Harris, who ranks 10th all-time in rushing yards, only ranks 11th when aggregating z-scores. Why? Primarily because Harris played for 13 seasons and he was always just above average in terms of yards gained per season. In fact, Harris never gained more than 1,250 yards or fewer than 600 yards in any season where he appeared in at least 12 games.

An even more glaring example is Marcus Allen. You'll notice that he's 9th on the all-time rushing list, but he's missing from the aggregated z-score list. Actually, his aggregated z-score is an abysmal 314th! In fact, his aggregated z-score over a 16-year career is -2.07 (yep, that's a negative sign preceding the 2.07). This is such an extreme case that it almost defies logic. Allen made a very long career out of being one of the most consistent running backs ever, but when it came to running the ball he was consistently just below average (Allen is hurt, of course, because we're not counting receiving here). In 1985, Allen rushed for 1,759 yards -- far and away his best season. The other fifteen seasons he averaged just under 700 yards per season. In Allen's case, his longevity actually worked against him in terms of his aggregated z-score. Because his rushing totals were below average almost every season, his z-scores were largely negative.

Obviously, this is one weakness of using z-scores -- at least as a measure of career success. When people mention Marcus Allen, it usually conjures images of that unbelievable run he had against the Redskins in the Super Bowl, not images of his final years in Kansas City. And that's the point -- new statistics are good for clarifying strategies and uncovering the underlying value in measures like total rushing yards. Still, statistics shouldn't obscure mind-boggling feats or great careers; they should supplement these experiences instead of replacing them.

That said, there is a fix. As Mike Tanier pointed out to me, "The career rankings suffer from Linear Weights syndrome: hang around and be slightly below average for 4 years, and suddenly your standing drops. You might consider using just the z-scores of the RBs six best seasons, for example." Yet another option might be to weight a running back's z-scores such that the early and late seasons in their career receive less weight than the middle, more productive seasons. This will give value to seasons where the running back was most effective and lessen the impact mediocre performances have on aggregated z-scores. This is something to be addressed in version 2.0.

Anyway, using z-scores allows us to now compare players not only to their peers, but also to players across different seasons. One of the benefits of this method is that it dispels the myth that milestones like 2,000 rushing yards in a season are the ultimate measure of dominance. Hopefully I've shown here that it also depends on the competition. Maybe run defenses are particularly weak in a season when running backs are particularly strong. Is having five running backs break 1,500 rushing yards in a season less impressive than having one do it? I think so, and that's one of the good things about using z-scores.

As always, there are sure to be better measures of running back success. Michael David Smith has done a lot of work on the subject. Last spring, Michael wrote an article (Introducing Leader Ratio) that compared running backs based on the difference in yards gained between the leading rusher and the runner-up. He followed that up with his Similarity Scores piece that used a classic baseball analysis tool to draw comparisons between running backs with similar numbers.

This article is by no means definitive, but hopefully, it sheds some light on how a running back's rushing performance shouldn't simply be taken as a raw number, but instead should be considered relative to how other running backs performed during a given season (or a career).

Posted by: P. Ryan Wilson on 30 Jun 2005

165 comments, Last at 21 Dec 2005, 8:26pm by felix adams

Comments

1
by Will Allen (not verified) :: Thu, 06/30/2005 - 12:54pm

Was there a significant difference in propensity to fumble, if it was analyzed at all? Aaron has informed me that Pro Football Prospectus has a fumble analyisis, and I intend to get the book, but I've long suspected, without proof, that propensity to fumble is often overlooked in ranking some of the best running backs.

2
by B (not verified) :: Thu, 06/30/2005 - 12:54pm

I like any tool that lists Walter Payton above Emmit Smith, but I think the best way would be to list the top 6 seasons instead of weighting the early and later years less than the middle. We don't want to punish an rb for having a good rookie year.

3
by MDS (not verified) :: Thu, 06/30/2005 - 1:02pm

Will, sorry to echo Aaron, but there are definitely differences in propensity to fumble among the top all-time running backs, and you'll learn about those differences when we discuss Curtis Martin in the book.

I love this article, and I wish I had had Ryan's z-scores when we were having our debate about the relative merits of Earl Campbell and Jerome Bettis. (Click my initials and skip down to comment No. 21 if you missed it.)

4
by Pat (not verified) :: Thu, 06/30/2005 - 1:03pm

Yet another option might be to weight a running back’s z-scores such that the early and late seasons in their career receive less weight than the middle, more productive seasons.

The other option is to have a correction factor per year.

As an example: take every RB who's played 6 years, and average their z-scores for each year (i.e. average all 1st year z-scores, all 2nd year z-scores, etc.). That's the "nominal performance curve". Then treat all z-scores for each player as a deviation from that "nominal performance curve".

So you might get something like: year 1: -0.5, year 2: 0.1, year 3: 0.5, year 4: 0.1, year 5: -0.5, year 6: -1.0 for the nominal performance curve. Then if you actually have an RB who scored "year 1: 1.0, year 2: 3.0, year 3: 3.0, year 4: 2.0, year 5: 0.0, year 6: -0.5" their total score would be 0.5+2.9+2.5+1.9+0.5+0.5=8.8 if my arithmetic is right.

I prefer this solution, because it doesn't minimize late-year performances. It just lowers the expectation for them.

You'd probably only want to derive the nominal performance curve from a set of large statistics, and then try to extend it outwards using some functional form. The number of RBs that actually exist with huge numbers of seasons may bias you, because if a RB starts to suck, he plays less, or retires. So you may get to a situation where a great RBs late-career performance looks average, because only great RBs play that long, and that's where you built up your nominal performance curve from.

5
by mawbrew (not verified) :: Thu, 06/30/2005 - 1:41pm

I also love this analysis. Really interesting. I do have a question though. Is every running back in the league used to determine the z score (which I think the article implies) or a more select group (top rusher for each team?)? The reason I ask is the reference to Marcus Allen. 700 yards/season may be below average for a team's leading rusher but I would think it has to be above average when all the second and third team guys are included.

6
by RyanW (not verified) :: Thu, 06/30/2005 - 1:48pm

Pat,

Great idea. I was thinking about running the regression: zscores = yrs in league + (yrs in league)^2, and using the results as weights, but I like your suggestion better.

mawbrew,

Only RBs who averaged more than 10 carries a game for a season (140 carries through 1977; 160 carries from 1978-2004) were included in the analysis, so that's why Allen's z-score wasn't as high as it would've been if all RBs were included.

7
by Scott de B. (not verified) :: Thu, 06/30/2005 - 1:49pm

To really do this right you should correct for quality of opposition. Just figure out how many yards the teams the back faced gave up compared to the average team (not counting their games against the back) and divide by the total.

So if Runnig Back J. Smith gains 1800 yards against teams that combined allowed 110% more running yards than the average team, correct his total to 1636 yards.

8
by MDS (not verified) :: Thu, 06/30/2005 - 1:57pm

Scott de B., I've begun working on that. In fact, Ryan mentioned in the original draft of this article that I was working on it, but I asked Aaron to take it out because I have a tendency to start projects and then take my sweet time about finishing them.

9
by Richie (not verified) :: Thu, 06/30/2005 - 2:18pm

I love these kinds of articles.

What is the exact formula that is used to calculate a z-score?

I am skeptical of any RB ranking that doesn't put Jim Brown at the top of the career list. (And probably even single-season list.)

10
by senser81 (not verified) :: Thu, 06/30/2005 - 2:39pm

Nice to see a list that upgrades Simpson and Campbell while downgrading Marcus Allen. I've always thought that in the past, statistical analysis would lessen Simpson's and Campbell's accomplishments while making Marcus Allen look like a superstar (which he wasn't after 1985).

For the career rating, its interesting to see guys like McCutcheon and Larry Brown on the list. I am interested in how the career rating was calculated.

11
by Pat (not verified) :: Thu, 06/30/2005 - 2:54pm

Z-scores (standardized scores) are described here. Essentially you're asking "how many standard deviations is this person away from the mean?"

12
by Ben (not verified) :: Thu, 06/30/2005 - 3:09pm

Good analysis... one other idea, though, is to compare a running back's yards in a season to his team's passing yards in the same season. It drives me nuts to hear Emmitt Smith compared positively to Barry Sanders, when Sanders played for so many terrible teams and Smith played for teams with powerful lines and a great passing game.

13
by Johonny (not verified) :: Thu, 06/30/2005 - 3:23pm

Isn't sort of hard to compare Jamal Lewis to OJ given that the player usage was so different in the 2 eras. The chart tells me that prior to 1980 it was unusual for a team to get just 1 running back a large number of carries. Which isn’t that surprising given the difference in player usage in the eras.

14
by Parker (not verified) :: Thu, 06/30/2005 - 3:27pm

It's quibbling, but here it goes: I would prefer if, when doing this type of analysis, titles like "Best Rushing Performances" were used instead of "Best RB Ever".

I think if you are going to remove things like receptions, fumbles, ability to block and such that you should narrow the scope of what you are describing/defining. Just a pet peeve of mine.

Other than the title, I love it.

This is my favorite place on the whole wide web. Thank you, Al Gore!!

15
by Buddha (not verified) :: Thu, 06/30/2005 - 3:36pm

#12: I agree with you that Barry was better than Emmitt and that Emmitt had much better offensive lines throughout his career, but Barry did play on some pretty good passing teams.

Those Moore-Morton-Perryman squads were pretty good (until the playoffs...).

16
by Playit (not verified) :: Thu, 06/30/2005 - 3:41pm

#12

Man not the Emmitt vs Barry line again... I take it we should also factor in rush performance in the 4th quarter, fumbles, ability to block, short yardage conversion, % of runs for positive yards, and any other factors that show a strong coorelation between strong reliable runners and super bowl rings?

17
by B (not verified) :: Thu, 06/30/2005 - 3:53pm

Yea, but Emmit's team won more games.
Yea, but Barry's team had worse defenses.

18
by Will Allen (not verified) :: Thu, 06/30/2005 - 3:54pm

This is another example of how much more complex football analyisis is than baseball analysis. I'm a fan of Walter Payton and Emmit Smith, but it seems to me that trying to proclaim one better than the other without factoring the difference in offensive lines that blocked for them, or the QBs and receivers (Bob Avellini vs. Troy Aikman, anyone?) that kept opposing safetys honest, is problematic at best.

19
by Don Shula (not verified) :: Thu, 06/30/2005 - 4:17pm

Man!
Can you imagine Barry Sanders with Dallas' offensive line! -Scary!

20
by Vince (not verified) :: Thu, 06/30/2005 - 4:24pm

1) I think Marcus Allen gets a low score due in large part to splitting carries with Bo Jackson for part of his career. It seems to me that some guys are being punished for playing in RBBC teams, and that's not really they're fault. My solution would be to look only at those years with positive Z-Scores, and total those.

2) I'd love to see a similar analysis for wide receivers.

21
by Pat on the Back (not verified) :: Thu, 06/30/2005 - 4:26pm

One way to adjust for back usage (marginally) would be to use team rushing totals, but compare the back to the other teams. Personally, I'd be interested to see what teams had the best rushing seasons in history. I had actually started to do something on this but succumbed to the same problem MDS describes above.

22
by mawbrew (not verified) :: Thu, 06/30/2005 - 4:27pm

I'd almost forgotten how good Leroy Kelly was until I saw this list. Wonder if he would be interested in suiting up this year?

23
by Led (not verified) :: Thu, 06/30/2005 - 4:31pm

Great article.

This study tends to show that total career rushing yards is a far more accurate measure of RB quality than I would've thought. Seven of the top 10 in rush yards are in the top 10 in z-score, although the order is somewhat different. Also, the z-score top 10 pretty much matches up with CW on the best RB's ever (again, with the precise order being up for debate). Most people, I think, would put O.J. and Earl Campbell in the top 10 and drop Bettis and Harris out. Marcus Allen's rank is a surprise, though.

I'm also happy to see that Curtis Martin holds up so well in this analysis. I suspected he might be more of a compiler, but not so according to z-scores.

24
by TomC (not verified) :: Thu, 06/30/2005 - 4:36pm

What you'd ideally like to do is marginalize performance over all the "nuisance parameters" such as offensive line play, quality of opponent, etc. One way to do this is to replace the RB in question with a "test particle" back and see how that back does in the exact same situations. Since this isn't a video game, we can't do that particular experiment, but we can at least look at how other backs on the same team the same year performed in similar situations. Now we're stuck with the problem of independently evaluating the backup RB's, because they're not perfect test particles, and you don't want to punish the RB you're trying to evaluate for having a really good backup. If we're lucky, we might have other data on the backup RB's (years spent with other teams, maybe even college if we're desperate).

Anyway, it's a complex, highly non-linear problem that I sometimes daydream about somebody giving me actual money to solve.

25
by mawbrew (not verified) :: Thu, 06/30/2005 - 4:36pm

Ryan -

Thanks for the earlier clarification. One more question, have you looked at the same sort of analysis using yards/carry rather than total yards?

26
by MDS (not verified) :: Thu, 06/30/2005 - 6:30pm

TomC, I tried something similar to what you're discussing. Click on my initials.

Vince, Marcus Allen sharing carries with Bo is no doubt part of it, but he actually lost more carries because of his feud with Al Davis, which led Davis to tell Art Shell to bench him. Remember, a completely healthy Marcus Allen hardly figured in the Raiders' offense at all in 1992, when Eric Dickerson was their leading rusher and Bo was long gone. Davis was telling everyone that Allen was washed up, but Allen actually had five more productive years with the Chiefs.

27
by Trogdor (not verified) :: Thu, 06/30/2005 - 7:16pm

1) Loved it. Good stuff. Want more.

2) If Sayers is in the Hall, Davis should be in. I know this doesn't count ST and receiving, but his career rating shows just how spectaculalry dominant Davis was.

3) I'm glad to see Eddie George scores so well for his career.

4) Does number of games per season matter for this analysis? I guess it doesn't really matter for the deviations, although it would affect which backs are included each season. For example, during Brown's career three or four seasons were only 12 games, so backs with 120 carries should be included. If only those with 140+ were, it could boost his rating for those years a little, which might boost his average per year from 1.885 up above Sanders' 1.888. Not that that's the first thing I checked or anything, of course.

28
by Carl (not verified) :: Thu, 06/30/2005 - 7:22pm

"I had actually started to do something on this but succumbed to the same problem MDS describes above."

I did it. It's a bear of a (interlinked) database, but it includes all offensive numbers on trackable datasets from 1960 to 2004.

I would suggest two things:

1. Receptions be included and that you break out a total "touches" per season category, otherwise you end up punishing outstanding receivers (Faulk, James) and favoring guys like Harris.

2. It's important to consider the very different eras these players toiled in. I like to divide it between the pre- and post-1978 reforms, which would include the coming of free agency and the 1993 CBA, which radically changed the way RBs were used.

In a previous thread, I discussed the work I've done in conjunction with many of the guys on this list (comes out in the fall). It is designed to take into account a certain "punishment" factor that comes from being a premier back in the post-1978 NFL.

And, MDS, I still believe Bettis should be a first-ballot HOFer even if he starts getting mean with the press
(It should be noted that he burned some bridges among LA and St. Louis writers years ago).

As I've said elsewhere, you must account differently for players like Harris (who only had one season wherein he tallied more than half of his team's total "touches" among RBs) and James Wilder, who are asked to shoulder immense offensive loads on their own.

29
by B (not verified) :: Thu, 06/30/2005 - 7:36pm

I don't think Bettis is a first-ballot HOFer (And he won't be if he retires the same year as Favre and Rice, and Deion) but he deserves to be in the hall because of he's been able to stay healthy and productive for so long.

30
by senser81 (not verified) :: Thu, 06/30/2005 - 8:30pm

re: post #28

I fail to see how anyone can claim over the course of a career that James Wilder shouldered more of a load than Franco Harris.

Wilder had two seasons where he had a lot of 'touches', and that is pretty much it when it comes to describing Wilder's NFL career. The Bucs were a combined 8-24 those two years, so perhaps Wilder was used nearly as much out of necessity as opposed to effectiveness. Wilder's 1985 season resembled Ricky Williams' last season in 2003; both had a lot of carries (365 for Wilder, 392 for Williams), both had a low average (3.6-3.5), both caught some passes (53-50), both had a low yards per catch (6.4-7.0), and both scored a decent 10 TDs. I fail to see how that is extraordinary, except for the fact that both the 85 Bucs and 03 Fins had extraordinarily vanilla offenses.

When Franco Harris retired, I believe he had the most rushing attempts in NFL history, which is made even more remarkable when you consider how many postseason carries he had during his career, too. To knock Harris for not shouldering the load during his career while rewarding Wilder for playing on a horrible team for two years is strange.

31
by David E. (not verified) :: Fri, 07/01/2005 - 7:31am

A few comments:

1) Quoting from the article:

"Is having five running backs break 1,500 rushing yards in a season less impressive than having one do it? I think so, and that’s one of the good things about using z-scores."

This is merely stated as fact, but I don't understand why this must be true. What troubles me is that it seems that the entire z-number system is built upon this initial assumption. That is not to say that I do not appreciate the analysis - I just fail to understand why it necessary follows that one back who breaks 1,500 yards is more impressive than if five backs had accomplished the same feat.

Even beyond the traditional reason, quality of defensive opponents, there are many other circumstances that may account for large discrepencies between the #1 and #2 rushers in a given season that may have nothing to do with the disparity in ability between the two players.

Perhaps I'm just completely missing something, but I'm not sure I see the point of the z-number unless that assumption can be proven to be true for all seasons.

2) I agree with the comment above that Jim Brown is the best RB of all time, and this is why - generally speaking - I remain a touch skeptical about using statistics as the end all and be all to measure the worth of a football player.

3) I'm new here, but I noticed some reference to a discussion about the merits of Barry Sanders v. Emmitt Smith...sorry to annoy those who have had enough and repeat an argument that has probably been written 50 times elsewhere but I cannot resist!
Barry was a great RB, but if you're building a team to win a Super Bowl and you have to decide between one of the two to be your TB you are taking Emmitt Smith, hands down. Not so much because of the physical talent, but because of his character. Emmitt loves football and would never quit on his team. Barry was a fun talent to watch on the field - much like Michael Vick is fun to watch - but when it comes to winning games I will take the tough, disciplined, determined back who knows how to score in the red zone and finish off his opponents. Just my two cents - no flames please!

4) This is my first post on the site, thus I would like to conclude by congratulating the entire team of Football Outsiders for putting together a wonderful site. I look forward to receiving your 2005 Prospectus and providing more unsolicited feedback ;)

32
by B (not verified) :: Fri, 07/01/2005 - 10:11am

David: z scores don't compare the top runner to the guy in second place or even the rest of the top five, they compare it to the avearage running back that year. So if you had a year when 5 rbs broke 1500 yards, but the average is still low, all five of them would have impressive z scores. On on the other hand, in a year when five guys break 1500 yards, but the average is 1400 yards, it's much less impressive.

As for Emmit v. Barry, it's easier to find the end zone when yo ustart on the 5 yard line then it is when you start on the 30.

33
by ElJefe (not verified) :: Fri, 07/01/2005 - 10:16am

Not exactly a flame ...

Re. #31:
I too would choose Emmitt over Barry Sanders. Provided he came with Aikman, Irvin, and Novacek. Oh, and Tuinei, the NORML twins (Newton and Stepnoski), Larry Allen and Erik Williams. (Would you still pick Emmitt if he had played his whole career with the Cardinals?)

As mentioned earlier in this thread, this is the problem of applying statistical analysis to football. Baseball is a team sport because more than one player is on the field at the same time. But it could be better described as a "group of individuals" sport. The quality of a pitcher has extremely little to do with how good the 2nd baseman is, and the 2nd baseman will hit the way he does almost regardless of the left fielder. It is very difficult to judge any football player as an individual.

There is no disputing that Emmitt Smith was a terrific running back and worthy of the Hall of Fame. But he never had a successful season without Troy Aikman, Michael Irvin, et cetera. Primarily because by the time that even became possible he was already in the declining stage of his career. Do we punish Emmitt for having good teammates? Do we give him too much credit for having good teammates? [The adjusted top-10 list for individual rushing seasons is littered with bad quarterbacking, with the exception of Jim Browns two seasons. Hmmm ... ] In the end it is in the eye of the beholder.

But this argument is just about where in the top 10-15 running backs who ever played the game Emmitt belongs. In my opinion (without actually trying to make a list) Emmitt belongs around 6th-8th. I find him most comparable to Cal Ripken. What made Emmit special was his ability to be on the field week-in and week-out, his toughness. But I'm not convinced that toughness and greatness are the same.

The odd thing is that Barry Sanders could have very easily been a Cowboy. When Jimmy Johnson was still coaching at the U. of Miami (1985), they played a game against a team led by another future Super Bowl-winning coach, Barry Switzer ( :lol: ). In that game, the Miami defense broke the leg of the Oklahoma QB ... Troy Aikman. This led to Jamelle Holieway taking over for OU, the Sooners going back to a wishbone attack, and OU winning back-to-back national championships. The change of offenses spurred Troy Aikman to transfer to UCLA and as a result of the transfer he sat out a year and entered the NFL draft in 1989, not 1988. Without Aikman in the 1989 draft, the 1-15 Cowboys would very likely have taken Sanders at #1 and (still) taken Steve Walsh in the supplemental draft.

It would have been interesting to see if Walsh/Sanders/Irvin/Novacek could have won a Super Bowl or not.

I also wonder how well Troy would have done in Atlanta (Aundray Bruce???). :)

34
by Pat on the Back (not verified) :: Fri, 07/01/2005 - 10:55am

Of course, the teammate argument is compounded in another way, too, beyond defenses teeing off on the RB if there is no talent around:
Teams with good defenses and/or good offenses also provide more opportunities for the backs. A pourous defense that can't get the other team's O off the field allows:

1) more time to be eaten off the clock by the other team, providing fewer opportunities for the back to rush
2) more close games (when the other team matches them score for score) and/or defecits (when the other team blows them out), requiring less rushing and more passing because the team doesn't want to kill the clock.

Really, on teams that have bad defenses, it is the running back's stats that suffer first on offense, and not just because defenses can tee off on them. So Barry, on the bad teams, was hit with the double whammy of being the only offensive threat and also a restriction on usage.

35
by Who's Your Daddy? (not verified) :: Fri, 07/01/2005 - 10:56am

I think if Emmitt Smith spent his career in Detroit we wouldn't even know who he is.

36
by Short Term Memory Guy (not verified) :: Fri, 07/01/2005 - 11:34am

You mean that guy that played for the Cards last year? 937 yards, a 3.5 average and 9 TDs? He isn't very good.

37
by MRH (not verified) :: Fri, 07/01/2005 - 11:58am

Very good article.

When I looked at Marcus Allen's career stats I immediately thought of the earlier discussion Carl brought up.

Look at his 1985 season: 380 carries (and 67 rec). After that, his highest number of carries was 223 (1988). For most of his career he got around 200 carries per season. He was good enough to be 6th all-time on the carries list. He just spread them out over many years.

He had about a 4 ypa over his 5 years in KC, nothing special, but that was depressed somewhat because he got so many 3rd and short and goalline carries. Marty did not overuse him any year, and sent him out for the tough yards.

A few more seasons of 300 carries in the late '80s and a shorter career might have placed him higher on this list, but would that have made him a better rusher?

38
by TomC (not verified) :: Fri, 07/01/2005 - 11:58am

re: 26

MDS: Nice work (as usual).

39
by Carl (not verified) :: Fri, 07/01/2005 - 12:23pm

Senser,

I meant a Wilder-like season in the sense that he accounted for more than 80-some percent of his team's total rushes and a third of their passing in one season.

I did not mean to imply that he kept this up for very long. In fact, I've made a pretty persuasive argument elsewhere that only a very few players (E. Smith, W. Payton, J. Bettis) have been able to sustain such punishment, toting a majority of the team's "touches" for very many years.

I also am not disputing the inclusion of Harris in the HOF. I believe players should be rewarded for long, successful careers.

But it also should be noted that Harris came at the end of an era for running backs, the "pre-78" guys I talked about earlier, who shared touches in a rotating backfield.

Rocky Bleier was an outstanding blocking back who could carry the ball very well (and, in fact, crossed the 250+ touches barrier one year while Harris struggled with injuries), and I've always felt he should be noted whenever anyone mentions Harris.

As I mentioned earlier, Harris was able to reach a plateau of so many total career touches because he was in a rotating backfield that shared the ball and extended his career.

Had Harris played today, he likely would play a role not so different from Ricky Williams, impressed into 425+ touches per season, against bigger, faster and stronger defenders, usually in lone set-back plays without a nice blocker like Rocky to give up his arm and head for the first fat d-lineman in your way.

By saying this, again, I do NOT want to denigrate what Harris did. I just think we should be aware that the amount of punishment post-1978 backs in a time of free agency, salary caps and pass-happy formations is very different from earlier years.

The methodology we built in data was based on the insight from a number of RBs who worked with us, both post-78 and pre-78, including Jim Brown, Larry Csonka, Jerome Bettis, Rocky Bleier and Ernest Byner, among others.

40
by Who's Your Daddy? (not verified) :: Fri, 07/01/2005 - 12:28pm

Would not 1,000 yards rushing and under 10 touchdowns categorize you with a whole lot of other running backs in the NFL? -Seems more like a minimum requirement.

In my mind, a 1,000 yard rushing season is no longer extraordinary.

41
by Carl (not verified) :: Fri, 07/01/2005 - 12:28pm

Senser,

I also wanted to mention, again, James Wilder, who I believe is forgotten by most reporters but who is a pioneer in the league and should be rewarded as such.

If you're not nice, I'll bring up Herschell Walker again...

42
by Aaron (not verified) :: Fri, 07/01/2005 - 12:33pm

Next year marks the 20th anniversary of Walker's arrival in the NFL, and I've already decided that I have to create a USFL-to-NFL translation system next summer to stick in Pro Football Prospectus 2006. The essay will address the question: "Does Herschel Walker Belong in the Hall of Fame" and be dedicated, of course, to Carl.

43
by MCS (not verified) :: Fri, 07/01/2005 - 12:40pm

Even more tools for my Rice vs. Hudson comparison request.

I'd do the work myself but I can barely calculate an average.

44
by Carl (not verified) :: Fri, 07/01/2005 - 12:44pm

Dedicate it to Senser. That'll get him.

45
by senser81 (not verified) :: Fri, 07/01/2005 - 12:48pm

You seem to have a revisionist history when remembering Franco Harris. When he retired, Harris was the 3rd leading rusher in history and the leading postseason ground gainer and scorer, and was a first ballot HOFer, so I hope you wouldn't dispute his enshrinement.

As for Bleier, he was a special teamer in 1972 when Harris was ROTY, set the NFL record for most consecutive 100-yard games, and gained 1000+ yards with a 5.6 average. So its not as if Harris' success was solely due to Bleier. And after Bleier retired in 1980, Harris had 3 more productive seasons.

The year Bleier 'crossed 250 touches' in 1976 had nothing to do with Franco's "injuries"; it had everything to do with Terry Bradshaw's injuries. Franco in 1976 had 1128 yards rushing and led the NFL with 14 TDs. This might show a flaw with your 'touches' theory, because Bleier's touches probably had more to do with circumstances beyond his control than with Bleier's individual greatness. Bleier's 1976 rushing total (1036 yards) is more than 25% of his 11-year career total, so its an exaggeration to believe that Bleier's 1976 season was representative of his NFL career.

I still fail to see how 'touches' is a gauge for RB greatness...IMO, it appears to be a gauge for bad offenses and RB overusage. James Wilder has my pity, but I would stop short of calling him a great RB. I agree that the way RBs are used has changed, but Franco Harris is a great RB regardless of era.

46
by Carl (not verified) :: Fri, 07/01/2005 - 12:49pm

Just so you know, Aaaron, I'm doing that injury analysis for you now. I'm about a third done.

47
by Carl (not verified) :: Fri, 07/01/2005 - 12:56pm

Your memory of the 1976 season and Bleier's don't seem to jibe. But I was a child at the time, and a fan of the Bears.

Again, I would never dispute the fact that Harris was a RB worthy of the HOF. I'm simply suggesting that the era he played in was very different from those of today.

There are a very many great and small RBs today who easily pick up more touches in a given season, and account for a greater share of their team's offense, than Harris did. That's no knock on Harris, simply a reality of the game post-78, post-Cap, post-free agency.

As for Bleier, had he and Harris entered the league in 2005 and not the early 1970s, both likely would have played on different teams, certainly once Bleier had proved he had rushing skills beyond that of a blocker.

But that's a "what if" game I don't want to play. They played in the era they did, and should be remembered as such.

And Walker belongs in the Hall. His USFL numbers should count.

48
by senser81 (not verified) :: Fri, 07/01/2005 - 1:04pm

re: post #42

Sounds interesting. Maybe you can do a similar translation system for baseball, which proves that Crash Davis of the Durham Bulls is the 2nd greatest hitting catcher of the modern era behind Mike Piazza.

re: post #41

Still on the Wilder love train, I see. What exactly did Wilder "pioneer"? How to be a mediocre RB on a bad team to inflate your stats? Did Wilder pioneer the NFL careers of Adrian Murrell and Raymont Harris?

49
by Carl (not verified) :: Fri, 07/01/2005 - 1:10pm

Wilder (following Harris, actually) was one of the first backs to crank out that many touches and still walk. That he was grossly underpaid for his services, physically broken forever after his two-year run (by his own admission, despite gamely continuing in the league) and largely forgotten today I see as precursors for the vast majority of "feature backs" since him.

I see Wilder and Payton as pioneers. They proved that one man could take an immense amount of punishment in an era of free agency. They gave teams the notion that they could rely on one back for the vast majority of their careers, using their dollars to acquire other positions.

Payton, of course, was a freak of nature who could withstand the brutality and become, in my humble opinion, the greatest RB I ever saw.

Wilder couldn't.

And Wilder is a heck of a lot nicer to talk to than Dickerson.

50
by J (not verified) :: Fri, 07/01/2005 - 1:20pm

Carl,

I was interested in hearing your opinion on optimal RB planning in the current NFL.

Due to high injury risk, it seems paying a feature back big money is a bit risky - as I believe you have posted on other threads. Keeping in mind the salary cap represents a fixed amount to spend each year, by paying one feature back big money would logically leave less to pay for adequate back ups.

Considering this, do you think it is best for a team to have two above average, relatively moderate priced, rotating backs?

If not, what do you think is the best backfield plan for current NFL teams?

51
by senser81 (not verified) :: Fri, 07/01/2005 - 1:25pm

Wilder and Payton played about 10 years before the era of free agency.

I enjoy a James Wilder conversation as much as the next guy, but I still don't understand what he 'pioneered'. Chuck Foreman had a season 10 years before Wilder's big 1984 campaign in which he was 3rd in the NFL in rushing attempts, 1st in receptions, and scored 22 TDs in 14 games. I would say that Foreman was a pioneer before Wilder.

52
by Carl (not verified) :: Fri, 07/01/2005 - 1:28pm

You made me open up the damned database. Curse you, Senser!

One thing I like to look out are the number of "hits" per game (carries + receptions) a RB took over the course of his career or from year-to-year.

Here are a few that stick out, HPG/Yr (up to 2004-05 season):

Wilder 30.75 (1984)
Williams 29.2 (2000)
James 29.16666667 (2001)
George 28.3125 (2000)
Tomlinson 28.1875 (2002)
James 28.125 (2000)
Allen 27.9375 (1985)
Smith 27.86666667 (1994)
James 27.69230769 (2003)
Williams 27.625 (2003)

Career (more than five seasons):

James 26.98461538
Williams 25.97142857
Martin 23.89285714
George 23.375
Payton 22.78947368
J. Lewis 22.53061224
T. Davis 22.51851852
B. Sanders 22.31372549
Faulk 22.25342466
J. Brown 22.21186441

By the way, Tomlinson would be in the second position if I didn't discount for his few years in the league (don't wear him out, SD).

Bear in mind, also, that these guys play 2-4 more regular season games than Brown did during his run.

To me, there should be a premium placed on guys who can shoulder these immense demands. You mention that Williams has a lower YPC ratio than others. But I would suggest that it's partly because he was asked to rush so many times.

Let me get my own prejudices out now: I believe Faulk and James are the best running backs since the salary cap was instituted. They are outstanding receivers and very good rushers. The ability to catch passes from outstanding QBs makes them value added commodities and they should be rewarded not only for their punishing RB play, but their tempting of fate by receiving balls usually over the middle, in the fly, with linebackers bearing down on them.

53
by MCS (not verified) :: Fri, 07/01/2005 - 1:31pm

Hudson = Hutson.

Sorry, Friday b-4 long weekend.

Happy Independence Day to all the Yankees out there.

54
by B (not verified) :: Fri, 07/01/2005 - 1:33pm

The Wilder vs Payton argument feeds nicely into my "longevity is a skill" theory. Payton was able to have a long, successful carreer despite carrying the ball an awful lot of times because he had the skill to avoid big hits. By shucking, jiving, bobbing and weaving, he was able to avoid taking the full front of a lineman/linebacker bearing down on him. Wilder didn't have this skill, and he broke down rather quickly.

55
by Carl (not verified) :: Fri, 07/01/2005 - 1:36pm

Nutty me. I thought the veterans' free agency rights were guaranteed after the Sept. 20, 1982 strike (or, four years before Wilder got his great season and Payton got to a Super Bowl).

The 1982 CBA ran through the 1986 season, and the free-agency system was supposed to go unchanged until 1992.

Along the way, of course, came 1989, when the management committee's Plan B stipulation allowed 229 unconditional free agents to change teams.

This open era of free agency was largely left untouched by the 1993 CBA.

Unless I'm missing something...

56
by Carl (not verified) :: Fri, 07/01/2005 - 1:42pm

I think, Senser, what you're saying is that the wide availability of free agency -- which came after settling the 1987 NFLPA lawsuit five years later -- is the "true" beginning of the practice, although it certainly happened in a different form before then.

I've always placed the beginning of free agency at an earlier date, although for most players this didn't occur until after 1992, in conjunction with salary caps, which further spurred their movement.

It's important to remember, however, that certain veterans had free agency rights before 1987, and they exercised them, although it wasn't nearly as widespread as it is today.

57
by Carl (not verified) :: Fri, 07/01/2005 - 1:50pm

Without boring every poster here, Mackey v. NFL, which challenged the free agency "Rozelle exclusion" clause protected certain players, although most lost their free agency rights under the 1977 CBA.

There is always some debate over how much latitude key veterans had from 1977-1992 in the NFL, although there were free agent rights extended to certain players, often even encapsulated in contracts wherein teams ceded their rights under the CBAs (although this usually accrued to management's benefit).

By now, I've lost the attention of everyone here and caused Aaron to shout, "You're killing my web traffic!" Shut up about Mackey v. NFL!"

58
by senser81 (not verified) :: Fri, 07/01/2005 - 1:55pm

Carl,

Yeah, I don't really view the "Plan B Free Agency" as the era of free agency. I don't think Herman Fontenot (probably one of your favorite RBs) going from the Browns to the Packers had that big of an effect on the NFL landscape. But I guess 'technically' you are correct. Plan B was funny, because the 37th-42nd worst players on a team's roster were eligible to be free agents, so if you were the 36th worst player, you were out of luck.

Anyway, I do find it very interesting how teams began to rely on one RB rather than a few RBs. I think Emmitt and Barry were the first RBs to continually have 90-95% of their team's rushing attempts. Neither of them were very big, so perhaps other coaches thought if a 5 foot 9 inch RB can carry such a load, then my RB should be able to do the same thing. I would like to know what you think of that...as long as you don't mention James Wilder!!

59
by Carl (not verified) :: Fri, 07/01/2005 - 2:26pm

Reporters usually make the mistake when they say "free agency began in 1992." What they should say is "unrestricted free agency began in 1992."

Certainly, restricted free agency had been around for quite awhile, and was used. Mackey opened the door before 1977 by forcing the NFL to scrap the "Rozelle" clause, but the union renegotiated a CBA that continued the restricted trade off for veterans leaving teams.

Wilder actually left his team, if I recall, as a Plan B guy (he ended up on the Eagles, I think). By then, of course, the beatings had turned him into a role player, not a stud.

Anyway, here's a snapshot of life in the NFL. I pulled up the 2003 numbers because it's always good to bring up Ricky Williams around Aaron.

You have your RB followed by his percentage of rushes and receptions on the team.

LNAME Rush % REC %
Williams 81.16% 19.46%
Martin 78.97% 13.46%
McAllister 78.70% 21.97%
Henry 77.70% 9.56%
Tomlinson 75.06% 33.67%
Barber 72.02% 20.18%
Alexander 71.96% 13.25%
Holmes 71.75% 21.83%
Taylor 71.73% 14.50%
Green 70.02% 16.13%

Wait, I thought King believes a Travis Henry trade deserves a first-round draft pick? With those hands? No way.

Extending the metaphor, here's the sort of punishment they took:

RB Touches Touches
per Yr per Gm

Williams 442.00 27.63
Martin 365.00 22.81
McAllister 420.00 26.25
Henry 359.00 23.93
Tomlinson 413.00 25.81
Barber 347.00 21.69
Alexander 368.00 23.00
Holmes 394.00 24.63
Taylor 393.00 24.56
Green 405.00 25.31

Here's the way the game was played in 1960:

Brown 56.28% 11.88% 234 19.5
Taylor 49.89% 10.95% 245 20.42
Tracy 46.72% 17.27% 216 18
Haynes 45.88% 35.03% 211 15.0
Casares 42.90% 5.56% 168 14
Smith 42.54% 20.69% 210 17.5
Rolle 42.21% 9.01% 151 10.79
Pietrosante
41.18% 7.83% 174 14.5
Crow 38.53% 19.84% 208 17.3
Dupre 38.10% 14.00% 125 11.36

And here's when the changes began (please note that it took until the mid-1980s for the NFL to finally sport more WRs on rosters than RBs, despite the success of teams such as SF):

Wilder 84.44% 26.23% 492 30.75
Payton 58.98% 20.83% 426 26.63
Dickerson
70.06% 12.00% 400 25.00
Riggs 72.93% 14.63% 395 26.33
Anderson
59.71% 20.65% 359 23.93
Dorsett 64.53% 16.09% 353 22.06
Winder 58.85% 17.12% 340 21.25
Allen 53.50% 24.06% 339 21.19
Jackson 64.91% 9.77% 335 20.94
Riggins 55.90% 2.46% 334 23.86

You really begin to see certain players (Payton, Wilder, Dickerson) notching some high-touch games that wouldn't be out of place in today's game.

As everyone knows, Payton's share of the rushing game in previous years certainly was higher than this snapshot (1984).

What this proved to teams is that they could rely on certain players (first with Harris, later with others) to account for well more than 325+ touches every season, a number that took greater precedence as the offensive game began to shift toward the pass, and away from the run.

If you need to stockpile WRs, TEs and pass blockers rather than HBs, TBs, FBs or any other combination of RB you previously had, why not just give the ball to one guy?

For those in here who don't grab the names, that's Ottis Anderson and Earnest Jackson (both of whom were very good players).

60
by Carl (not verified) :: Fri, 07/01/2005 - 2:37pm

At this time, I also would like to mention the worst receiving feature back in the history of professional football, Christian Okoye.

In 1989, playing for the Chiefs, he accounted for 372 touches, all but 2 of which were rushes.

I can imagine the shock that came over the DB's face when Okoye's hands o' meatloaf managed to hang on to the ball.

I get the feeling the defenses didn't exactly scheme for the safety valve pass.

61
by senser81 (not verified) :: Fri, 07/01/2005 - 2:56pm

Interesting stuff, especially the 1960 part. In a way, you could argue that if Jim Brown got 70% of his team's carries, his numbers would have been even more astounding. But thats just conjecture.

Also, I remember everyone hating Earnest Jackson. Buddy Ryan called him the worst 1000-yard RB in league history. He almost rushed for 1000 yards 3 years straight on 3 different teams (Chargers in 84, Eagles in 85, Steelers in 86). He had like 950 yards with the Steelers. I remember watching him and being very nonplussed at how he had 1000-yard seasons.

62
by princeton73 (not verified) :: Fri, 07/01/2005 - 3:03pm

sorry--any rating system that doesn't end up with Jim Brown as #1 is wrong--by definition

(and, yes, I'm prejudiced)

63
by Carl (not verified) :: Fri, 07/01/2005 - 3:04pm

I remember his as a terrible receiver, an OK runner and a guy who blocked like a demon.

I remember him picking up a blitz once and taking out two defenders at the knees.

But that's what Coryell wanted. I'm sure Fouts appreciated it.

By the way, I had forgotten until I looked up that year that Chuck Muncie played on the Chargers that year, too.

I kind of thought that had Muncie caught on at any team other than New Orleans or San Diego he might have been somebody.

Of course, I thought Abdul-Jabbar was the perfect RB for Miami, so take it for what it's worth.

64
by Richie (not verified) :: Fri, 07/01/2005 - 3:09pm

One thing I just found the other day, that gives me some pause in my ranking of Jim Brown as easily the best RB ever, is that Leroy Kelly had immediate success with Cleveland upon Brown's retirement. His numbers weren't quite as good as Jim's, but stellar nonetheless. It's still amazing that Jim Brown had ~50% more career yards when he retired than any other RB. (Jerry Rice has about 50% more than Tim Brown, the all-time #2 receiver.)

Maybe Dickerson needs to be downgraded due to the success of Greg Bell, etc.

65
by Carl (not verified) :: Fri, 07/01/2005 - 3:24pm

In 1965, Jim Brown accounted for 289 rushes, 1,544 yds, 17 rushing TDs, 34 receptions for 328 yds, 4 rec tds.

That made up 61 percent of his team's rushes and about 22 percent of their receptions.

His 323 touches averaged out to 23 per game.

He was the best RB in the NFL.

In 1966, Leroy Kelly played 14 games, rushing 209 times for 1,141 yds, 15 rushing tds, catching the ball 32 times for 366 yds and 1 td.

He made up 50.36% of the team's rushes and 15.24% of Cleveland's receptions. He had 241 touches, or 17.21 per game.

I'd say Kelly was the seventh best RB in the league that year, far behind Nance of New England.

Kelly's teammate, Ernie Green, also had a great year -- 144 rushes, 750 yards, 3 rushing TDs, 45 receptions, 445 yds, 6 rec tds. He made up 34.70% of the Browns' rushes and 21.43% of their receptions. That added up to 189 touches, or 13.50 per game.

What I'm suggesting is that to replace Brown, Cleveland relied on two very good players, who both had swell years, but they weren't Mr. Brown.

66
by Richie (not verified) :: Fri, 07/01/2005 - 3:28pm

Carl - would you mind taking those averages out another decimal place, please. :)

#51 - Chuck Foreman does not equal James Wilder because Foreman was 3rd in the league in carries. Foreman had 280 carries on a team that ran the ball 535 times. Wilder had 407 carries on a team that ran the ball 448 times. Surely you see a difference!

67
by Parker (not verified) :: Fri, 07/01/2005 - 3:31pm

Carl, that % of touches stuff is very interesting, from era to era. Thank you.

I am still not at all convinced that Wilder's ability to take punishment makes him an all time great back or makes any individual season of his one of the best ever. An impressive performance? You bet. Something he should be remembered for? Gotcha. A fairly unique talent set? I'll give it to you. But one of the best RBs? Ever? No. I just can't go there.

68
by Carl (not verified) :: Fri, 07/01/2005 - 3:35pm

Parker,

I don't believe Wilder is one of the greatest RBs of all time. But I think the two-year stretch 1984-85 was the most punishing run ever put together by a tailback.

I take into account the division he played in, of course, especially the fact that he had to face the Bears two times per annum.

That had to hurt.

69
by Carl (not verified) :: Fri, 07/01/2005 - 3:40pm

And, for Senser, the mandatory Walker aside.

HW accounted for a quarter of the Cowboys' receptions in 1986, despite starting only nine games.

He certainly made the case that year that he should have been starting instead of Dorsett.

70
by Parker (not verified) :: Fri, 07/01/2005 - 3:42pm

I'll not dispute that. I think you have used different words to describe those seasons in another thread, but no matter.

I'm glad you are here and enjoy the insight.

71
by Carl (not verified) :: Fri, 07/01/2005 - 3:47pm

Parker,

I believe Wilder turned in the single greatest rushing season of all time in 1985 because of his p-score (punishment).

I still can't believe he lived to run in 1985.

72
by Parker (not verified) :: Fri, 07/01/2005 - 3:56pm

We'll agree to disagree. But if you'll change 'greatest' to 'most impressive' we won't even need to do that.

73
by senser81 (not verified) :: Fri, 07/01/2005 - 3:59pm

re: post #66

Richie, my point was that guys like Foreman and Lydell Mitchell were the first RBs to pull extensive rushing/receiving double-duty for their teams like Wilder did. I still don't see what Wilder 'pioneered', if anything he pioneered running back burnout. He didn't prove that a RB could have a huge amount of carries and receptions...if anything, he proved the opposite because his career was shot after 1985.

As for the statistical difference between Wilder and Foreman, surely I saw the difference, but I didn't know huge statistics were what made a pioneer. I never claimed that Foreman was Wilder's statistical equal. That would be like you saying that Jerry Rice was more of a receiving pioneer than Mac Speedie because Rice had much better statistics...surely you see the how illogical that is.

74
by Richie (not verified) :: Fri, 07/01/2005 - 4:22pm

Here's the list of all RB's who had at least 90% of his team's rushes in a single season (since 1960):

Name Year Team Carries TeamCarries Percentage

James Edgerrin 2000 clt 387 392 98.7%
Sanders Barry 1992 det 312 321 97.2%
James Edgerrin 1999 clt 369 382 96.6%
Wilder James 1985 tam 365 383 95.3%
Sanders Barry 1995 det 314 333 94.3%
Sanders Barry 1990 det 255 271 94.1%
Williams Ricky 2001 nor 313 334 93.7%
Sanders Barry 1996 det 307 328 93.6%
White Lorenzo 1992 oti 265 285 93.0%
Sanders Barry 1994 det 331 359 92.2%
Johnson Rudi 2004 cin 361 392 92.1%
George Eddie 1998 oti 348 378 92.1%
Tomlinson LaDainian 2002 sdg 372 406 91.6%
Staley Duce 1999 phi 325 355 91.5%
Rhett Errict 1995 tam 332 363 91.5%
Tomlinson LaDainian 2001 sdg 339 371 91.4%
Henry Travis 2002 buf 325 356 91.3%
McAllister Deuce 2002 nor 325 356 91.3%
Tomlinson LaDainian 2003 sdg 313 343 91.3%
McAllister Deuce 2003 nor 351 386 90.9%
Wilder James 1984 tam 407 448 90.8%
Bettis Jerome 1994 ram 319 352 90.6%

James Wilder is definitely the outlier in this list. He's on it twice. Nobody did it before him, and the first person to do it after him was Barry Sanders in 1990.

Before Wilder, the record was 87.4% by Jim Brown in 1961. Brown's performance now ranks 39th all-time.

75
by Carl (not verified) :: Fri, 07/01/2005 - 4:27pm

%$**$#&&%&&%&!!!!

I put away the database and there Senser goes and makes an argument I can refute! Well, I'm putting the thing away for the weekend.

I definitely would agree with him that Mitchell and Foreman presaged the era of backs who routinely made more than 350 touches per season, or 25 per game.

But I see guys like Payton routinely accounting for nearly 400 touches annually in the late 1970s and early 1980s and surviving from season to season.

I mention Wilder in this just because I like to talk about Wilder's 1985 season.

If I were being intellectually honest, I would say that Payton is the better example of a guy who begins to turn in season after season of consistently high touches, numbers no one in the 1970s would have believed prudent to do.

Without reviving another argument from another thread, I would suggest that this has a lot to do with the changing economics of the game and the shakeout from the post-1978 reforms and the growing use of (restricted) free agency.

The strongest case for a rotating backfield came from owners, themselves, before the late 1970s. They believed it was suicide for them to waste their top offensive talent (rushers in a running league), "burning" them out or exposing them to increased risk of career-ending injury by forcing them to account for even half of their backfield's output.

That's why Franco Harris only once made up more than half of his team's touches in any given season. A GM invested a great deal of money on the runners (both for the top rushers and the line of blockers that protected them) and wanted to preserve that investment for as long as possible.

That's why you saw certain teams (Steelers, Miami) extended the rotation system well into the late 1970s. It bothers when I see people scoff at Csonka's numbers, for example, without remembering he was the fullback focus in a three-headed attack that showcased other great, albeit different, runners.

In a way, Senser, I think Earl Campbell perhaps had more in common with Wilder than Payton, who was a freak of nature. Had Wilder been able to run a few more years at his 1985-86 pace, he likely would have been in the HOF, like Campbell.

Campbell, of course, played for better teams.

76
by PerlStalker (not verified) :: Fri, 07/01/2005 - 4:30pm

Okoye was all the rage for a while because he just flattened people. I remember the first Chiefs/Broncos game durring Steve Atwater's rookie year. Atwater repeatedly knocked Okoye back 2-3 yards. The first time he did it, Okoye got up looking a bit surprised. It was like it had never happened to him before. Ah, the memories.

77
by Carl (not verified) :: Fri, 07/01/2005 - 4:30pm

Actually, Richie, my math is different than yours. Maybe our stats are different, too. I get mine from Elias.

I have James at 89 percent in 2000, for example.

78
by Carl (not verified) :: Fri, 07/01/2005 - 4:31pm

Total team carries, according to Elias, for the Colts in 2000 was 433, not 392.

79
by Carl (not verified) :: Fri, 07/01/2005 - 4:34pm

And James had a few more rushes, too.

Hmmmm...

80
by senser81 (not verified) :: Fri, 07/01/2005 - 4:39pm

Please, no more comparisons of Earl Campbell and James Wilder! I think everyone played for a better team than Wilder did. Those mid-80's Bucs teams were horrendous. That still doesn't make it right to compare Campbell to Wilder.

In post #74, the chart shows that Wilder had 365 of Tampa's 383 rushing attempts...but my information shows that Tampa had 434 rushing attempts in 1985. What happened to those other 51 carries? By my calculation, Wilder only had 84% of his team's rushing attempts in 1985 (365/434).

81
by Richie (not verified) :: Fri, 07/01/2005 - 4:42pm

For my next trick:

I took the leading rusher (in terms of carries) for each team since 1960. I then totalled up the carries for the team leader for each season. Divided that by the total number of rushes for each season. The goal is to see what percentage of carries the leading rushers for each team average (by season).

In 2002, the leading ball carriers for each team averaged handling the ball on 69.6% of it's team's carries. That is the all-time high. The goal is to see how the trend has changed over time. The lowest is 41.8% in 1970.

1981 was the first season above 50%. 1994 was the first season above 60% (hmmm...right when free agency was beginning).

1960 43.2%
1961 48.2%
1962 48.5%
1963 46.8%
1964 47.1%
1965 46.1%
1966 46.6%
1967 44.2%
1968 43.6%
1969 44.8%
1970 41.8%
1971 42.9%
1972 46.5%
1973 46.6%
1974 41.9%
1975 43.4%
1976 42.7%
1977 42.8%
1978 42.2%
1979 45.8%
1980 45.1%
1981 51.0%
1982 52.8%
1983 53.6%
1984 53.7%
1985 55.0%
1986 51.3%
1987 47.5%
1988 50.1%
1989 54.8%
1990 52.3%
1991 52.6%
1992 59.0%
1993 56.5%
1994 61.9%
1995 65.2%
1996 61.0%
1997 62.4%
1998 65.1%
1999 62.7%
2000 69.0%
2001 65.7%
2002 69.6%
2003 64.8%
2004 64.2%

82
by Richie (not verified) :: Fri, 07/01/2005 - 4:45pm


Actually, Richie, my math is different than yours. Maybe our stats are different, too. I get mine from Elias.

I have James at 89 percent in 2000, for example.

Hmm...I'm using pro-football-reference.com. I don't know how accurate that site is. Apparently not very.

83
by Carl (not verified) :: Fri, 07/01/2005 - 4:46pm

I think the database he's using is missing a few rushes, most probably from QBs and WRs.

That's more important in the earlier eras when QBs were expected to run more.

84
by Richie (not verified) :: Fri, 07/01/2005 - 4:47pm

I see part of the problem. I am only comparing these RB's to other RB's on the team. (Carries by QB's, WR's do not count in the team totals.)

85
by Carl (not verified) :: Fri, 07/01/2005 - 4:47pm

Richie, you're right about Wilder's numbers that year. They link with Elias' count.

I can't vouch for every year you have, though, and I trust Elias.

86
by Carl (not verified) :: Fri, 07/01/2005 - 4:49pm

I love how I now have Senser and everyone else talking about James Wilder. I should go over to the daily Manning Sucks, Brady Rules board and interject a little bit about Wilder.

87
by Richie (not verified) :: Fri, 07/01/2005 - 4:50pm

I think the database he’s using is missing a few rushes, most probably from QBs and WRs.

That’s more important in the earlier eras when QBs were expected to run more.

The goal of my charts here was to see the progression of RB overuse - the "feature RB" effect. I would say that in most offenses the number of carries by non-RB's mainly just noise in the data (and largely cancels out). An exception may be Atlanta who gives plenty of carries to Michael Vick.

Anyway, in 1992 the "feature back" effect took a big jump from previous seasons.

88
by Carl (not verified) :: Fri, 07/01/2005 - 4:51pm

R,

You might want to note the strike years.

89
by Richie (not verified) :: Fri, 07/01/2005 - 4:55pm

The 1981 strike didn't seem to have much of an affect, but the 1987 one did since they used scabs.

90
by Carl (not verified) :: Fri, 07/01/2005 - 4:57pm

The reason why you want to count WRs and QBs, R, is because of the earlier years (and Vick now).

In 1960, for example, you had RB/WR combos like Bobby Mitchell and Johnny Morris out there, not to mention guys like Al Dorow, Ralph Gugliemi and Jack Kemp and Johnny U who could put up some serious rushing numbers.

91
by Carl (not verified) :: Fri, 07/01/2005 - 5:00pm

By the way, in 2002 Vick accounted for 21.46% of his team's rushes.

If Al Dorow were playing today, would he make it onto the cover of Madden06?

92
by Carl (not verified) :: Fri, 07/01/2005 - 5:02pm

Click on name.

Eat your heart out, Michael Vick!

93
by Richie (not verified) :: Fri, 07/01/2005 - 5:03pm

The little spike in 1972 and 1973 is interesting. In 1971 only 6 RB's had over 50% of it's team's carries (OJ's first heavy season). Then it was like in 1972 the league copied Buffalo and 9 players were over 50%. Then in 1973 11 players were over 50%. But then in 1974, it dropped down to just 5 players again. Ron Johnson was a heavy lifter for the Giants, but in 1974 he was no longer their #1 back. So that probably had a big effect on the 1974 drop. (The Giants' leader went from 59% in 1973 to 36% in 1974.)

94
by MCS (not verified) :: Fri, 07/01/2005 - 5:04pm

So, anyway. . .

Z-score.

Pretty cool stuff. Good report Ryan.

95
by MCS (not verified) :: Fri, 07/01/2005 - 5:05pm

What I really wanna know is. . . How does Wilder stack up against Bettis? :-)

96
by Carl (not verified) :: Fri, 07/01/2005 - 5:10pm

My only knock on OJ was that he wasn't a very good receiver.

Everybody talks about seasons like 1973, but he only caught six passes that year. I know if you're running like that a QB isn't so tempted to throw it to you, but another reason for OJ was that he was very good at catching the ball.

At this point Senser is going to freak out, but here it goes: I might have been tempted, if I were a GM, to goor Boobie Clark TODAY over a back like OJ.

That's today, not in the early 1970s.

97
by Carl (not verified) :: Fri, 07/01/2005 - 5:11pm

That should be "wasn't" good at catching the ball.

98
by senser81 (not verified) :: Fri, 07/01/2005 - 5:14pm

re: post #87

Another exception may be the 1985 Bucs with James Wilder, when the Bucs had a QB named Steve Young. Doh!

re: post #96

Boobie Clark over OJ Simpson? Do you say things just to upset people?

99
by Carl (not verified) :: Fri, 07/01/2005 - 5:18pm

From 1978-2004, James Wilder had the GREATEST (or, most impressive) season (1984). Bettis had the 43rd (1997).

Bettis, however, is the greater of the two rushers and belongs in the HOF, alongside Herschell Walker, Bill Polian and Ray Guy.

100
by Carl (not verified) :: Fri, 07/01/2005 - 5:20pm

Yes, Senser, I do.

I could make an honest argument for Calvin Hill being a great player today. I just threw out Boobie Clark to watch you flip out.

101
by Parker (not verified) :: Fri, 07/01/2005 - 5:21pm

(I agree)

102
by senser81 (not verified) :: Fri, 07/01/2005 - 5:21pm

I can understand Calvin Hill being a great player today...but that is completely different from saying you'd rather have Boobie Clark instead of OJ Simpson. I guess the big question is, would you take Boobie Clark instead of James Wilder?

103
by Carl (not verified) :: Fri, 07/01/2005 - 5:22pm

Senser, Young only played in five games that season. That said, he accounted for about 9 percent of the team's rushes. Shades of things to come, on a better team.

Too bad they couldn't have hooked up in SF.

104
by Carl (not verified) :: Fri, 07/01/2005 - 5:25pm

If I could have the James Wilder of 1984-85, I'd play him every year. The problem is that if you do that, you'll get the hobbled Wilder of 1987.

If I could have any running back today it would Faulk or James (in their prime), but that's just me.

I've watched teams mortgage their future for Ricky Williams, so what do I know?

105
by senser81 (not verified) :: Fri, 07/01/2005 - 5:25pm

Even in Wilder's greatest season of 1984, he still trailed Eric Dickerson in yards from scrimmage that year, despite having 92 more touches.

106
by Carl (not verified) :: Fri, 07/01/2005 - 5:28pm

And had Wilder played for the Rams instead of Dickerson, we wouldn't be having this discussion.

107
by senser81 (not verified) :: Fri, 07/01/2005 - 5:36pm

And if Wilder had played for the Colts in 1987, he still would have had less than 500 yards rushing like he did in real life. Lemme guess, if Wilder had played on the 1980 Houston Oilers, he would have ran for 1934 yards just like Campbell did.

I would love to hear you debate about baseball...I bet you'd argue that Wilbur Wood was a better pitcher than Sandy Koufax.

108
by Carl (not verified) :: Fri, 07/01/2005 - 5:42pm

By the way, Wilder played against Dickerson in 1984.

The Rams were a playoff bound team. The Bucs were 4-9 entering the game. The Rams edged them 34-33.

The Bucs led 26-17 going into the fourth quarter.

And we won't mention, Senser, that Young came from the USFL.

109
by Carl (not verified) :: Fri, 07/01/2005 - 5:47pm

By the way, here's a spin on my "versatility" argument about the best backs by a guy who is a Rams fan.

Please notice that Wilder scores big while Dickerson is nowhere to be found, and he's Rams' fan!

As for Koufax, if Wood went 25-4 AND hit a fourth of his teams RBIs AND stolen bases AND fielding average, I'd say he was a better pitcher.

Dickerson couldn't catch very well.

And Wilder threw a TD pass.

110
by carl (not verified) :: Fri, 07/01/2005 - 5:48pm

Ooops. Click on name.

111
by senser81 (not verified) :: Fri, 07/01/2005 - 5:48pm

I remember that 1984 game. Dickerson had 191 yards rushing and 3 TDs and single-handedly won the game for LA.

Nor will we mention that Steve Young's USFL career ensured his Canton enshrinement...much like Larry Csonka's WFL career ensured his HOF status.

More on Wilder:

'John McKay's last game, played in front of 43,813 Buccaneer faithful, did take on a sense of the bizarre. James Wilder had the best season ever for a Bucs running back. He rushed for a team record 1,544 yards on an NFL record 407 carries, and also caught 85 passes for 685 yards giving him 2,229 total combined net yards for the season. When a Wilder TD run gave Tampa Bay 41-14 lead with 1:21 left in the contest, McKay went for an onside kick to allow Wilder to get the neccessary yardage to break O.J. Simpson's record of 2,243 combined yards in a season. Because of Buccaneers penalties, Obed Ariri made three onside kick attempts before the Jets recovered. In McKay's last game as head coach he ordered his defense to let Jets running back Johnny Hector score, so Wilder would have another shot at the record. While Wilder fell 15 yards short of the record, the "Buc Flop" generated plenty of controversy. Jets players spewed obscenities at McKay after the game. New York head coach Joe Walton was not pleased. "The way it ended was a total embarrassment to the NFL. It set it back twenty years. It was completely uncalled for." McKay was not apologetic. "I'm sorry we didn't get the record for Jimmy, but we tried, and the Jets almost attacked me." On his decision to retire McKay said he had "No second thoughts." He closed his post game press conference with "Adios, gentlemen. God bless you." The league did end up fining McKay the largest amount ever levied against a coach.'

112
by Carl (not verified) :: Fri, 07/01/2005 - 5:53pm

You can't blame Wilder for that!

113
by Carl (not verified) :: Fri, 07/01/2005 - 5:59pm

I would humbly suggest that had James Wilder run behind Kent Hill and Doug Smith, or had Henry Ellard on his wing to catch a few passes, the Buc would have turned in a better season than Dickerson, that year.

I would like to note that Steve Courson played on the Bucs, too.

114
by Ron Mexico (not verified) :: Fri, 07/01/2005 - 7:41pm

In honor of the Outsiders, here's one of the only Brown U cards in existence, in re the NFL.

115
by Carl (not verified) :: Fri, 07/01/2005 - 7:42pm

Check out the Syracuse collection. Not a shabby crew.

116
by CTF (not verified) :: Mon, 07/04/2005 - 2:54pm

Great article.

Interesting that Walter had the 3rd best Z total for a year in 77 before Earl got there. In 78, with Walter in the league, Earl had the 20th best "Z" year, and in '80 he had the 4th best. Walter might not be on that list had Earl graduated in 76 vs. 77 (or if Juniors could declare back then).

117
by another Carl (not verified) :: Mon, 07/04/2005 - 10:37pm

re 60, 76...

ah, Christian Okoye. Thanks to Tecmo Super Bowl, the 1991 NFL rosters and Christian Okoye will forever be enshrined in my memory. He was pretty much unstoppable. I used to take KC and play entire seasons rushing Okoye on every single down - no pass plays, and Barry Word was relegated to kick returns. Okoye would rack up 4-500 yards per game, with 100% of the touches - match that, James Wilder.

sorry for the aside...

118
by zlionsfan (not verified) :: Tue, 07/05/2005 - 12:30pm

Re Barry/Emmitt trading places: I actually tried this once with Front Page Sports: Football Pro in the mid '90s when both were at their peak. I simply traded the players straight up and went to sim the season ... but after I simmed the season a couple of times (don't remember which season), I realized that it wasn't as simple as swapping the players.

What else would have changed if, somehow, they'd traded places? Would Dallas have kept their huge OL or looked for more mobile linemen? Would Detroit have stuck with a run-and-shoot-style offense or would they have changed to be more conventional? I don't know how much you'd have to change to make the scenario realistic, and how close you'd be to simply swapping entire offenses rather than individual players.

Even with a complex football sim, as we've all seen on this site, it's very difficult to tell how much of a player's success is because (or in spite) of the system in which he played. It's not like baseball, where you can pretty much swap two RFs, for example, and see how their teams would do with the new players.

119
by Dan (not verified) :: Tue, 07/05/2005 - 4:22pm

A quick note from a stats head. Z-scores may give some useful insight, but they are not a magic solution to the problem of determining who was the best running back. Z-scores are only directly comparable if the distributions from which they are derived are the same, or at least have very similar shapes. This is especially true when you are working with relatively small samples (e.g., less than 100), as would be the case with running back performances from any given year. For example, the standard deviation may be unduly influenced in one year by a small number of outliers, thereby reducing the apparent difference between the top one or two yardage totals (in z units) and the central tendency of the distribution (i.e., rest of the pack). This is just one of may ways in which z-scores may mislead, in absence of careful consideration of the shapes of the distributions from which they were derived. There are a number of methods that may be used as points of comparison for z-scores, with simple ones including the absolute deviation from the median or third quartile. As a place to start you might want to see how the standard deviation varies from year to year, which is an interesting piece of information to consider in an of itself.

120
by Nelphonious of Pennefielde (not verified) :: Wed, 07/06/2005 - 12:26pm

Z score also benefits comparisons of individual seasons where( ieJIM BROWN)) career spent in 12 game seasons vS. Dickerson in both 14 and 16 game seasons vs. Sanders and Smith all 16 game seasons.(RE:#40)pROPORTIONALLY,the ole 1,000 yd. season benchmark is now 1,350 yds.Further Z score refinement,fumbling frequency,yds. per total touches (run + receptions),option thrown tds, points per total touches,etc lead to a Jamesian T.R.B.V. index.

121
by senser81 (not verified) :: Wed, 07/06/2005 - 2:27pm

Is 'yards per total touches' even a stat? If so, my money says that Brian Mitchell is the greatest RB of all time. Option thrown TDs is relative, too? How about including kicking statistics as well, so Paul Hornung and Gene Mingo can climb up a few spots.

122
by Carl (not verified) :: Wed, 07/06/2005 - 2:34pm

As I mentioned earlier, any assessment of a running back should include receptions and his share of the team's rushing/receiving stats.

It likely would boost some outstanding backs -- Faulk, James, Calvin Hill, Walker, Payton, et al -- and bring some others down a notch (O.J., Dickerson, Riggins).

The notion of football is to move the ball forward. A RB should be rewarded if he does that with a carry or a reception.

123
by Dan (not verified) :: Wed, 07/06/2005 - 2:39pm

That would depend on whether you were interested in the best 'pure runner' (i.e., considering only attempts where the ball is received via a handoff) or the best 'running back'. It all depends on what you want to look at.

124
by senser81 (not verified) :: Wed, 07/06/2005 - 3:07pm

All RBs share the same responsibility of running the football. IMO, it appears that things such as receptions and 'percentage of touches' are dependent on outside factors as opposed to being reflective of the RBs skill. Rushing yards is still the single most important stat for a RB, and rushing yards equates with greatness moreso than rushing attempts. Eric Dickerson's 1984 season was superior to James Wilder's no matter how you slice it. Dickerson had more rushing yards, more TDs, more total yards, a higher average...the only thing Wilder had more of is touches. I don't understand how you fail to grasp that Wilder's insane amount of touches was a result of the Bucs being so bad; not the result of Wilder being so good. Is it at all surprising that when Fred Willis became the first RB to lead his conference in pass receptions, he played for the 1-13 Oilers? Does that mean Willis was an exceptional player, or does it mean that the Oilers were so bad that a RB was their leading receiver?

125
by B (not verified) :: Wed, 07/06/2005 - 3:11pm

If we're going to consider how many yards a RB accumulates in the passing game, we should also consider the percentage of passes thrown his way that he catches.

126
by Carl (not verified) :: Wed, 07/06/2005 - 3:27pm

And the fact that in 1985 James Wilder threw one completion for a TD!

Take that Dickerson!

127
by Carl (not verified) :: Wed, 07/06/2005 - 3:39pm

Again, Senser, you miss my points.

The reality is that the contemporary running back is NOT there to generate rushing yards. This was true in 1965. It's not so much in 2005.

Indeed, it would be fair to say that the running back was the single most important position in 1965, and by now he's fallen to the point of becoming nearly marginal to the game of proffesional football.

This is reflected not only in linear regression analysis that tracks the relative value of premier backs to winning in the league, but is also mirrored in the median salaries paid to rushers, a process the artificial market of salary caps and free agency has articulated better than coaches.

I would agree that Wilder's 1985 outlier season was an anomaly based more on the horrible offense perpetuated by the Bucs. Fair enough.

But his herculean ordeal presaged a reality that was already a trend -- the advent of the multi-purpose, high-touches "feature" back. By being able to withstand such punishment, for little financial reward, he predicted what we now accept as conventional.

A high-touches back saves a team money because the relative cost of a RB is less than other position players. A high-touches back who can catch the ball is a very great value added commodity for a team because the post-1978 rules are rigged in favor of the passing game, and those who can catch balls are more important than the backs who can't (which is why I'm so biased for Faulk, James, Hill, et al).

You forget that Wilder's season was more impressive than Dickerson's in another respect: He was a much better receiver. By catching nearly a third of his team's passes, he also saved them the cost of acquiring an elite WR.

One likely wouldn't make that argument with Faulk or James, but certainly Hill or Walker would fit for Dallas during those years.

128
by Parker (not verified) :: Wed, 07/06/2005 - 4:28pm

Didn't Dickerson have 50 receptions in 1983? Is his lack of receptions in 84 supposed to indicate that he suddenly lost his ability to catch the ball or is it more likely that the Rams just decided to not throw it to him as much that year?

Frankly, I don't care if Wilder was a better reciever that year. 2105 yards and 5.6 per carry can make up for any deficiency in the passing game in my book.

Maybe that isn't 'FO Correct' thinking, but there it is.

129
by Jim A (not verified) :: Wed, 07/06/2005 - 4:34pm

I'm not sure it's really fair to count RB reception yards on the same scale as rushing yards because passing plays inherently gain more yards than running plays. Yards/touch stats would favor 3rd down receiving specialists such as Larry Centers, Amp Lee, Dave Meggett, or even Keith Byers, all of whom gained more receiving yards than rushing yards at about 6 yards per touch overall, yet were mediocre rushers. Even Eric Dickerson averaged 7.6 yards per reception. RBs also tend to catch a high percentage of passes thrown their way because the distance is usually short and the coverage is often soft. Should ED's teams have thrown him the ball more, or is there some other benefit of 4.5 yards per carry that we're not considering?

130
by Carl (not verified) :: Wed, 07/06/2005 - 4:59pm

Jim,

I only care about stats that make Wilder look better than Dickerson, at least for one year.

Quit bringing up inconvient arguments like that.

131
by senser81 (not verified) :: Wed, 07/06/2005 - 5:13pm

re: post #127

I guess we are just going to have to disagree on the cause/effect of Wilder's 1984 numbers. You say "By catching nearly a third of his team’s passes, he also saved them the cost of acquiring an elite WR", as if its BECAUSE of Wilder, the Bucs do not need an elite WR. I would argue the opposite and say "Because the Bucs had poor WRs, Wilder caught nearly a third of his team's passes" and say that the EFFECT of Tampa not having good WRs was Wilder getting a ton of receptions (and scoring a whopping 0 TDs on 85 receptions).

132
by dryheat (not verified) :: Thu, 07/07/2005 - 10:52am

In 2003, The Dainian Tomlinson caught 100 passes. Is this because he is such a great pass catcher, or is it more likely that San Diego's WR corps "featured" Reche Caldwell, Tim Dwight, Eric Parker, and Kasim Osgood?

I'm betting on the latter. I think Wilder falls into the same category.

133
by Carl (not verified) :: Thu, 07/07/2005 - 11:12am

Senser, we don't disagree about Wilder. In 1984 and 1985, he certainly was a very good player (I would say outstanding) who "lucked" into a situation wherein he was the best RB, best WR and best kick/punt returner on a horrible, horrible team.

His efforts made a truly monstrous team somewhat mediocre, which was a herculean effort that ended up breaking him as a player.

But what he did became the model for elite "feature" or "premier" or "whatever" backs in the coming years, especially after free agency and the salary cap took over the finances of the game.

That's why I would give extra value to RBs who can catch the ball alongside their rushing duties. The game has evolved to make these value added players some dominance over their peers (anybody can be a Droughns because they're fungible; very few can be a James or Faulk or Tomlinson).

Also, the primary benefit (from a GM's standpoint) is NOT whether a back puts up OJ-type numbers, although he wouldn't dislike that. Rather, he prefers the sort of durability, high-touch load from guys like Bettis, Martin, Faulk, et al, who have really become the model of the successful back.

134
by senser81 (not verified) :: Thu, 07/07/2005 - 12:27pm

James Wilder had 0 kick return yards in 1984 and 1985. In addition, he had 0 punt return yards in 1984 and 1985. For his entire 10 year career, Wilder had 61 total yards in kick returns and punt returns.

In 1985, the Bucs were 2-14. That is not somewhat mediocre.

I fail to see how Wilder is a model for anything. The teams he played for accomplished nothing. If anything, Wilder was a cautionary tale to coaches showing what overuse can do. Wilder was already burned out by 1986.

Also, you are confusing cause-and-effect. You make it sound as if 'touches' are all that is necessary to be an elite RB, as if only a select few great RBs can 'touch' the ball a lot in a given season. Production is still what separates RBs. Again, that is like saying the best pitchers are the ones with the most innings pitched or the best batters have the most at bats...it makes no logical sense to imply that merely showing up equates with greatness.

In the last 10 years only 4 RBs have ever finished among the league leaders (top10) in receptions (Faulk, LT, Centers, Garner). In the 5 years prior to Wilder's "revolutionary" 1984 season, 7 RBs finished among the league leaders in receptions (Washington, Young, Ted Brown, Cooper, Andrews, Springs, and Wilder in 82). Quite a trend-setter.

135
by Carl (not verified) :: Thu, 07/07/2005 - 12:38pm

As for Tomlinson, his 2003 reception numbers were the highest, ever, for a RB.

Ever.

The nearest any back had come to his 2003 tally was Roger Craig's 92 receptions for the 1985 Niners.

Or maybe you want to suggest that the WRs and QBs for the 85 49ers were somehow equivalent to their colleagues on TB or the SD effort of 2003.

Charlie Garner also had 91 receptions for the Raiders in 2002. Another team with a mediocre QB and a trio of pretty crappy receivers, right?

A guy I really admire is Faulk, who had 87 receptions in 1999. I'm going to assume you're not going to make the same argument that he played for a team with a shoddy set of WRs.

So, maybe, just maybe, Tomlinson is a uniquely gifted athlete. Perhaps his numbers were inflated, somewhat, by the fact that the WRs on the Chargers in 2003 weren't that good. But it's equally possible that he was the right guy in the right place, who has very good hands if not breakaway speed on the fly.

136
by Carl (not verified) :: Thu, 07/07/2005 - 12:44pm

Senser, sorry, I meant to make that about Walker, not Wilder. I mixed up my Ws.

137
by senser81 (not verified) :: Thu, 07/07/2005 - 1:01pm

As someone mentioned a few posts ago, look at Tomlinson's career in terms of seasonal receptions. Rookie year he has 59 receptions, then 79, then 100, and last year 53. So do we assume that LT was a decent receiver coming into the league, then became the best receiving back ever, then all of the sudden fell back to the 'decent' category? Or could it be that in 2003 due to circumstances beyond LT's influence allowed him 100 receptions?

Look at Roger Craig's 1985 season. He had 92 receptions, but more importantly he averaged 11 yards per catch and had 1000 yards receiving. He usually had a high number of receptions, but his 1985 receiving yards almost doubles his next highest seasonal total. Why? In 1985, the Niners were transitioning out Solomon and Clark for Rice, and perhaps more importantly Earl Cooper was benched, so the 41 receptions Cooper had in 84 turned into 2 receptions in 85. In sum, there were more opportunities for Craig to catch passes, but that doesn't mean Craig was twice as good as a receiving back in 1985 than at any other point in his career.

If Garner's 91 receptions in 2002 is reflective of his skills, how come that one season is nearly a quarter of his 11-year career total? Did Garner suddenly become a tremendous receiver, or were there other factors involved?

Lionel James' seasonal reception totals read 23, 86 (1985 season), 23, 41, 36. In his 86 reception season, James averaged 12 yards per catch and had 1027 yards receiving. Does that mean James was the best receiving back of all-time in 1985, or could James' abberational total be explained by Kellen Winslow's injuries and 1984 Charger #1 RB Earnest Jackson's departure?

This is my biggest concern with this site in general...instead of just looking at stats, sometimes looking at what actually happened helps, too.

138
by B (not verified) :: Thu, 07/07/2005 - 1:27pm

Didn't Tomlinson in 2003 earn the Duece McAllister Award for for most meaningless (receiving) yards accumulated? It seemed like every catch he had that year was good for 5 yards on 3rd and 8, or 7 yards on 3rd and 10.

139
by Carl (not verified) :: Thu, 07/07/2005 - 1:30pm

Senser,

LT could be a 100-reception back every season, although his value as a WR isn't so great as Faulk's or Craig's was.

You certainly aren't going to make the argument that Craig or Faulk weren't good receivers or weren't tied intimately into the team's passing game, ditto with Garner before his shortcomings as an all-down back were exposed.

This was Craig's career:

48 (1983) 71 92 81 66 76 49 25 17 22 19 (1993)

Obviously, from 1984-1989 he was a premier receiver in the league, albeit one operating out of the RB position. The 92 in 1985 was a high point, but one couldn't say the 81 or 76 in following years were outliers in his career.

He only played in 14 games in 1987, of course, and 1989 really marked his dwindling abilities as a full-time back.

Here are Faulk's numbers, beginning in 1998 when he was with the Colts:

86 87 81 83 80

Again, I'm sure you would agree that his high point wasn't an outlier and that he really was that good of a receiver. It's not like Warner didn't have other options.

I would say Garner's numbers were inflated, but that wasn't because he was a sometime receiver who cashed in that year.

He had been advancing for the previous four years in receptions while at SF and Oakland (56 68 72) before he hit 91. Really, it reflected a changing role for him as the Raiders realized he wasn't going to survive as an all-down back, but would need to have a more tailored game that better played up his ability to catch the ball.

One of my problems with this board is that people make arguments based on stats without ever talking to the people who make decisions for the team. I can assure you that SD believes Tomlinson is an outstanding receiver, ditto for Faulk and Garner with their squads.

I'd grade Tomlinson as a back as capable of 75+ receptions per year, especially if he were playing in a system like those employed in Indianapolis or St. Louis.

I saw Herschell Walker in the same way, a good receiver, a touch below James or Faulk, but very good in his own way.

140
by senser81 (not verified) :: Thu, 07/07/2005 - 2:05pm

As usual, this discussion has turned futile. I mention that Craig's 1016 receiving yards were nearly double his next highest seasonal total, and you mention Craig had 81 receptions in 1986. Huh? And I certainly wouldn't make the argument that Craig and Faulk weren't good receiving backs...I have no idea where you got that from.

Anyway, the point I have been repeating ad nauseum is that a RBs receptions are dependent more on the role of the RB position in general in that particular offense and the surrounding players than a RBs receiving skills. The receiving stats for a RB are misleading moreso than their rushing stats. Take a look at your favorite player, Herschel Walker.

In his last year with the Generals, Walker has 37 receptions in 18 games in 1985. In 1986-1988 with the Cowboys, Walker has 76, 60, and 53 receptions. Whoa, did Walker finally figure out how to catch a football? But then with the Vikings, Walker has 18, 35, and 33 receptions. Looks like Walker forgot how to catch. But wait, when Walker goes to the Eagles he has seasons of 75 and 50 receptions! Herschel can catch again! So I guess we assume that Walker was a poor receiver with the Generals, developed his receiving skills greatly with the Cowboys, forgot everything he learned with the Vikings, then remembered everything with the Eagles.

141
by ZasZ (not verified) :: Thu, 07/07/2005 - 5:04pm

Clearly, senser, Herschel Walker was a great receiving back in the NFC central.

142
by dryheat (not verified) :: Thu, 07/07/2005 - 5:19pm

RE: 135

Off the top of my head, Larry Centers had over 100 catches (101, I think) a few years back, so to say that Tomlinson had the most...catches...ever...by an RB is simply erroneous. At best, he would have been #2.

143
by B (not verified) :: Thu, 07/07/2005 - 5:22pm

Except Larry Centers is a fullback.

144
by Carl (not verified) :: Thu, 07/07/2005 - 6:08pm

That's true. I should have pulled up the FB database, on which he's an outlier.

I just did. He has 101 in 1995 and 99 the following year.

The reason he's not part of this discussion, of course, is because HE NEVER GOT MORE THAN 116 CARRIES.

As a FB, he was no Larry Csonka, Franco Harris or Jim Taylor, to put it mildly.

My analysis was based on RBs who were "feature" backs, thus the mentioning of players such as Tomlinson, Faulk or James.

145
by Carl (not verified) :: Thu, 07/07/2005 - 6:17pm

"So I guess we assume that Walker was a poor receiver with the Generals, developed his receiving skills greatly with the Cowboys, forgot everything he learned with the Vikings, then remembered everything with the Eagles."

That's not what I was saying, and you know it. The various data we have been discussing point to tendencies among the "feature" backs, or, at least, those who mustered more than 225+ carries+receptions at some year in their career (what NFL statisticians call the "James Brown Line," in honor of the very great JB's 1960 season).

All I'm saying is that when evaluating the merits of a "feature" back, one should include receptions.

Nobody would say that this isn't determined, in part, by the offensive schemes employed by the teams, and that going from one squad to another can affect rushing or receiving totals.

I agree fully that one must take statistics into context when detailing a player.

That said, however, I firmly believe that when a RB like Tomlinson picks up 100 receptions along with 313 rushes for 1,645 yards (and 13 rushing TDs), it means something. It gives him added value.

You and I have both mentioned in the past that Craig probably should get some consideration for the HOF (although I wouldn't vote him in). If you were honest, you would probably say that it's his reception totals that marked him off from many of his peers.

I see the same thing with Walker, and also believe he should get some credit for his USFL numbers (which came during the time when a RB is very productive in terms of age) and the duty he pulled on returns.

You disagree. Fair enough. All I'm saying is that there should be some minimum standards that come into play when evaluating a RB (mininum number of carries, for example), ypc, etc., but also std dev for all other rushers, reception tallies, etc.).

This is so common among NFL front offices, I can't see why you won't accept it.

146
by senser81 (not verified) :: Thu, 07/07/2005 - 6:19pm

Where does Eric Metcalf fit into this mess? He had a season with 28 carries and 104 receptions for 1189 yards in 1995. Would we say he was a better RB that year than Terrell Davis?

Also, does James Wilder still hold the record for most receptions in a season (85) without a TD? I know Keyshawn had a year with over 100 receptions and just 1 TD, but 1 is still more than none.

147
by B (not verified) :: Thu, 07/07/2005 - 6:23pm

I wonder if Larry Centers' talents have been under-appreciated. If a WR or a TE caught the ball 100 times a season, he'd be in the pro-bowl, but cause he's a fullback, and fullbacks are supposed to run and block, and Larry doesn't do do those things well, he gets forgotten about. On the other hand, catching the ball 100 times doesn't do you much good if you're getting 6 yards per reception.

148
by senser81 (not verified) :: Thu, 07/07/2005 - 6:38pm

I agree that receptions should somehow be taken into account, but it seems I disagree as to how important a RBs receiving stats are. I also find it hard to believe that anyone in an NFL front office would take James Wilder over Eric Dickerson, or would say that James Wilder was more productive than Dickerson in 1984. Saying that Wilder was better than Dickerson because Wilder had more receptions is like me saying Danny White was a better QB than Roger Staubach because White was the better punter.

I see what you are saying, but when guys like Herschel Walker and Tomlinson have such a variance in their receiving stats on a seasonal basis, it makes me wonder how much of a RBs receiving stats are due to circumstance.

149
by Jerry (not verified) :: Thu, 07/07/2005 - 6:51pm

One of my problems with this board is that people make arguments based on stats without ever talking to the people who make decisions for the team.

I guess that's why it's Football OUTSIDERS. (And I'm not holding my breath waiting for those decision makers to take my call.)

Having said that, I'm sure I'm not the only person here who appreciates that "inside" information when those of you who have it post it.

And Carl, best of luck in Iraq.

150
by Carl (not verified) :: Thu, 07/07/2005 - 6:52pm

Actually, Senser, he does hold that dubious distinction.

Perhaps that's not as bad as Mario Bates, who went his entire career without a receiving td (65 receptions for 419 yards but 0 in the endzone). Gary Olandis holds the same distinction (59 415).

So does Reuben Mayes, George Rogers, Anthony Thomas, Lewis Tillman (remember Tillman???), Troy Hambrick, Christian Okoye (my favorite), Dominic Rhodes (must have a crappy QB on that team), Mark Higgs, Reggie Brooks, Terry Miller, Joe Delaney, Rudi Johnson and William Green.

So says Elias (1978-2003, analyzing all "premier" backs over their careers)...

I just noticed that Mike Rozier had 90 receptions in his career, and only one rec td, which also is something of a low point.

In 2003, Priest Holmes caught 74 passes for 690 but, alas, no TDs. I guess you tend not to notice that when you've scored 27 TDs on the ground.

Some others:

1. John Williams, Seattle (1990), 73 699 0

2. T Barber, NYG (2001) 72 577 0
3. T Barber, NYG (2002) 69 598 0
4. Deuce McAllister, NO (2003) 69 516 0

5. E Smith, Cowboys (1995) 62 375
0 (but he had 25 on the ground)
6. Wilbert Montgomery, Philly (1984) 60 501 0

There are a lot of others, including Curtis Martin in 2001 and 2003, Marcus Allen in 1994. Walter Payton in 1982 and Michael Pittman (name your year).

151
by Carl (not verified) :: Thu, 07/07/2005 - 7:01pm

Actually, in 1995 Davis only played in 14 games, so had I been a GM with the ability to predict the future, I probably would have passed on him for (in order):

E Smith
Sanders
Curtis
Warren
Allen
Watters
Rhett
Hampton

All of whom played all 16 games and gained more yards from scrimmage.

Looking at the stats, I'd say Edgar Bennett had a better year than Davis.

But, again, Davis missed a couple.

152
by B (not verified) :: Fri, 07/08/2005 - 10:50am

Is a yard on the ground worth a yard in the air? I always thought that passing yards were inheriantly more risky than rushing yards. An incompletion or a sack is worse than getting stuffed, and there's also the chance of an interception. Of course a RB could fumble, but that can happen after a pass, too. The advantage of a pass is the potential to get more yards, but passes to backs are typically short passes. That's more of a function of a system than an indication of the back's talents though. I think the most important stat when looking at a back's receiving yards is how many yards after the catch he earns.

153
by Pat (not verified) :: Fri, 07/08/2005 - 1:41pm

Is a yard on the ground worth a yard in the air?

Nope. It's easy to show why, too, even ignoring turnovers. Take the distribution of yards that a RB gains (or loses) running and histogram it. You'll get some nice Gaussian with a spread of about 2 yards. So a RB might gain, say, 3+/-2 yards on each rush.

Now calculate the yardage that a RB gains receiving. Remember to add zero on an incomplete. Receiving is generally 80% successful for an RB, and you'll probably get a nice Gaussian of about 3-4 yards, centered at 6-8 yards. But you'll also have this spike of 20% of the integral of the others at 0. That pulls the average noticeably down.

Of course, this is all complicated due to the question of whether or not the ball was catchable anyway.

154
by B (not verified) :: Fri, 07/08/2005 - 2:40pm

I think what Pat said proves my theory that a RB's value in receiving yardage is best measured by his YAC.

155
by Pat (not verified) :: Sat, 07/09/2005 - 10:22am

B:

And completion percentage as well, of course. But I agree. All the other stats almost completely depend on the the quarterback's decisions and the coaches' choices, not the running back's skills.

156
by Carl (not verified) :: Sat, 07/09/2005 - 2:09pm

"I think what Pat said proves my theory that a RB’s value in receiving yardage is best measured by his YAC."

Actually, that's not what he said. He is providing a statistical tool that sometimes will reveal a RB adept at catching a ball and sometimes not.

For example, I can envision a scheme that would put the RB fairly far downfield, usually in the slot, so that the RB's ability to get open (usually against covering LBs, which creates a nice mismatch) is a skill not so different from that employed by WRs.

Faulk and James are particularly adept at that, which is a very different sort of play than a screen pass.

The YAC, for these backs (I'm thinking Craig and Walker here, too) is as important as where on the field they catch the ball.

Pat is right, of course, that ultimately the RB's skills are either helped or hurt by the ability of a QB to get the guy the ball.

As I mentioned earlier, D Rhodes has replaced E James many times during the star RB's injuries. Passes to the RB dwindle to almost nothing, and he's not even considered a redzone reception threat.

That suggests to me that it's not so much that the QB isn't very good or that the team doesn't know how to employ RBs over the middle in the passing game. Rather, it suggests that Rhodes isn't that great at either (1) getting open in the flat, even against middle linebackers; or (2) that he's not a very good receiver.

Part of the problem for statisticians, of course, is that RBs typically are used "over the middle" because sprinting WRs like to highlight their speed by going near- or farside to exploit DBs. WRs also don't much like going into the center of play because it exposes them to larger DBs (usually safeties) at very great velocity hitting them often midair.

Hines Ward is a guy who likes to go over the middle. Plaxico, not so much, which became evident in a lot of the play calling you saw.

157
by Carl (not verified) :: Sat, 07/09/2005 - 4:07pm

OK, B, let's say you want to take a look at a bell curve of all RB receivers from the last, say, 27 years.

What Pat is saying is that there should be a standard distribution of receptions (he's plotting for yards) among the various RBs, which, when put on the page, will resemble a curve.

Some players will be outliers -- such as James, Craig or Faulk or, for a year or so, Wilder -- others will fall into a normal pattern of receptions/yards, and still others (Okoye, Rhodes, et al) will be outliers in a different sort of way (they can't catch a ball).

When Pat says a RB-cum-receiver is good for, say, 5 yds per play in extra yard catching rather than rushing, he's certainly not saying this is true of every RB on every play, nor that it's true of every RB who might average, say, 30 receptions per season, mostly short-yardage screens or dump-offs.

Some players might consistently get 6 receptions per game for pretty good yards (such as Faulk), whereas others might get 8 receptions one game, then 3 the next, with yardage that varied, too (such as Craig).

But the problem for statisticians is that this must be correlated to the guy tossing him the ball. There is a formula (probalistic) that will correct for this (variance, covariance, SQRT, etc., which you can plug into Excel very easily or, better, SPSS or SAS), but you still must try to define your predictors as carefully as possible.

Pat, has anyone ever actually done this? I don't recall ever reading a study about it, although I imagine some teams have paid for them before drafting (James) or trading (Faulk) and, yes, I'm implying that would be the Colts.

Maybe we should try it?

158
by Pat (not verified) :: Sun, 07/10/2005 - 11:38pm

Some players will be outliers – such as James, Craig or Faulk or, for a year or so, Wilder – others will fall into a normal pattern of receptions/yards, and still others (Okoye, Rhodes, et al) will be outliers in a different sort of way (they can’t catch a ball).

Bingo. The problem mainly is that the distribution for rushing yardage is easy. It's a nice, easy-to-characterize distribution. I looked at the rushing distribution for RBs a while ago, and it pretty much just is a Gaussian with a long tail in positive yardage (because you can't rush for -30 yards, but you can rush for 30 yards).

The distribution for passing yardage is much different, though, especially because a completion becomes harder the farther you go out. So you want to punish an RB for incompletions, but not punish him for ones he wasn't likely to catch. (You need a "normalized reception percentage" curve somewhere. That is, the chance for an average receiver with an average QB to catch a ball as a function of distance thrown).

If you want to add in "receiving yardage" to "rushing yardage", that's really, really hard to do fairly. It makes more sense to characterize rushing and receiving separately. But if you really wanted to characterize receiving properly, you'd do it in the way I was mentioning - with a characteristic curve, in the same way you'd do rushing. Except now you also have to take into account the incompletions in a sane way.

Pat, has anyone ever actually done this? I don’t recall ever reading a study about it, although I imagine some teams have paid for them before drafting (James) or trading (Faulk) and, yes, I’m implying that would be the Colts.

Got me. I haven't seen it here, but then again, we're missing yards-after-completion information.

If I could ask the NFL to add one thing to the play-by-play stats, that'd be it: point of completion. Psst, Carl, think you can mention that to anyone? I'm sure Aaron, and everyone else, would be thrilled to finally get that friggin' stat. Without that one piece of information, there's so little you can actually correct for.

159
by CTF (not verified) :: Tue, 07/12/2005 - 11:17am

You draft a RB cause he can run it. If he can catch it, that's a bonus. That may give one guy the edge if they're fairly even otherwise.

Cedric Benson was not nearly the accomplished receiver Williams or Brown were and it didn't stop anyone from ranking him right there w/them both.

Of course it's situational as some west coast offense guys might rank receiving skills at more of a premium, but first and foremost a RB needs to run it.

Why not throw blocking ability into the equation? I bet miost coaches put a bigger premium on that at the running back position than their receiving skills.

It's all about the running imo.

160
by Dan (not verified) :: Thu, 07/14/2005 - 2:45am

I'd like to see how this new measure compares with VOA and PAR for the seasons for which VOA and PAR stats are available (it probably makes sense to leave out defensive adjustments). You could calculate correlations between z-scores, VOA, and PAR both for individual years and for cumulative totals, averaging or adding together all of the years for which there is data. High correlations would be a good sign for your stat. My guess is that z-scores will be more closely correlated with PAR than with VOA, since both are based on totals rather than averages.

You could also make an alternative z-score statistic that is based on RB's yards per carry for each season, rather than their total rushing yards for the season (for those with at least a minimum number of carries). My guess is that this ypc z-score would be more closely associated with VOA than with PAR, since both deal with averages rather than totals.

I think that there is no way of resolving whether totals or averages are a better measure of a RB's performance, since they are influenced in different ways by the rest of the team and the coaching strategy, so it's good to look at both. Even FO does a version of each with its super-complicated stats (DVOA and DPAR).

161
by Matt (not verified) :: Tue, 09/13/2005 - 8:28pm

Just found this page. I really like this form of analysis. I've been using a less exacting variant of this for a while just based on the premise that a player should get no credit on an "all-time great" comparison for years where he is simply an average player for his position.

You definitely still have some work to do to perfect it, but it seems like you're fully aware of how to go about doing it.

162
by Eddie (not verified) :: Thu, 09/29/2005 - 11:14pm

There are opinions and there are facts. The RB's job is to help move the offense from one side of the field to the other. Emmitt Smith has done that better than any RB in the game according to his stats. Then you must ask if he can put it in the end zone. E. Smith has done that better than any RB in the history of the game and he only ranks second all-time behind J. Rice. When you do these things at the highest level, then you get awarded with MVP's and rushing titles. Again, Emmitt has accomplished that. How about the post-season? E. Smiths toughness and stats in the post season too are at the top! Now, the biggest game of the year, the SUPER BOWL. Certainly a great RB can carry his team to a SUPER BOWL. And this too Emmitt has done well enough to capture a SUPER BOWL MVP. But still there is a knock on Emmitt Smith because his critics say he had a great offensive line. Only if Barry Sanders or Walter Payton had that O-line. So What! Because he had that great line and great teammates he should have the great stats. None of his back-ups had great games and they had the oppurtunity to run behind that line. Nobody would dare argue that Jerry Rice is the greatest WR of all-time. Yet Jerry Rice had two Hall of Fame QB's throw to him. I'm sorry, but if you claim that Barry Sanders or Walter Payton are greater RB's than Smith...then Andre Reed and Art Monk are greater receivers than Jerry Rice. Sure they are! Fact..Emmitt Smith is the greatest RB of all-time.

163
by Matt (not verified) :: Mon, 10/03/2005 - 7:50pm

Eddie, when you say "The RB’s job is to help move the offense from one side of the field to the other. Emmitt Smith has done that better than any RB in the game according to his stats.", what do you mean? Since you mention TD's separately, I assume you must be talking about Smith being the all-time rushing leader. If that's the case I strongly object to your statement. It's just silly to think that the yards Smith rushed for in his last few years turned him into the greatest at moving the football ever. Those years were meaningless stat padding.

164
by Eddie (not verified) :: Wed, 10/12/2005 - 7:05pm

Matt, you must agree that a running backs job is to help his team (offense) move across the field no matter the field position to the ultimate goal, which is to score. I claim that Emmit Smith has done that better than any RB in history because he has the individual accolades as well as the team accolades. Sure there have been great running backs that have great individual stats along with team accolades (Super Bowls), but better than Emmit? NO! This is why Emmit Smith is the best RB of all-time. I've heard the arguments for Barry Sanders, Walter Payton, Gale Sayers and Jim Brown, but they don't hold water. For example, Barry Sanders quit! Just quit! For whatever reason, only he really knows, he quit! Yes he was graceful and more eloquent to watch but his desire just fizzled out. He had no playoff success and no Championships. Are you willing to admit that a RB with exceptional talent with less yards and no championships is greater than Emmit? If you say yes, then that's opinion and not fact. Walter Payton was an awesome RB with both speed and strength and team success in his latter years. He was the all-time rushing leader for quite awhile. Guess what? One with greater stats and championships passed him up! Emmit Smith. The Jim Brown debate is difficult because the era's are different. Many claim if Jim Brown had played more years the record would be untouchable. Well, he didn't play on and found other interest during his prime. What does that say for his desire for the game. To say that Emmit Smith piled on to his stats is just not right! He still had a competetive fire and was the starter for the Cardinals. Unlike Jerry Rice, considered the best WR of all-time, who latched on to three teams and was not even the second option in any of them. If an NFL player is atop of the stats all-time at his position and has gloried in the ultimate team accomplishment, what can be better and more importantly who can be better at his position?

165
by felix adams (not verified) :: Wed, 12/21/2005 - 8:26pm

steve smith of the 2005 carolina panthers is the best player of all-time