Exactly.

[pbmax]

Yards per carry are a terrible way to measure the contribution of a running back.

I have been saying this all along, haven't I?

http://www.footballperspective.com/t...rds-per-carry/

Chase Stuart

One way to look at this is to say that carries are king; like routes run, they represent opportunities given to a player, which is a strong indicator of talent. Another interpretation is that YPC is a fickle mistress. It is extremely sensitive to outliers, and two or three carries can completely skew a player’s production. I’m not telling you anything you didn’t already know, but the data here seems overwhelming to me.

[MadScientist]

The article does a good job of complaining about the stat, but not a good job about saying would would be a good stat. There is a difference between Lacy and the collection of stiffs from the previous couple of years. Perhaps an explosive run rate stat, ie attempts per explosive run (10+ yards). I don't have time to compile that stat, but it might prove interesting.

[texaspackerbacker]

Yeah, exactly, but what IS a good stat or correlation of a star RB or an adequate RB or a shitty RB? One answer is you know them when you see them. There are so many factors, though, it's really hard to say: quality of the O Line, type of offense, other star or quality players to share attention of the D, etc. The thing about Lacy that is impressive to me is that he still looked good even when Rodgers was hurt and defenses could really focus on him. The thing that does not impress me about him is he can't seem to break loose for a big play - something which yards per carry is an indicator of. We're better off with Lacy playing most of the downs than any of the previous RBs including Starks, but he's no Jim Brown either or Adrian Peterson in his prime either.

[Guiness]

The article does a good job of complaining about the stat, but not a good job about saying would would be a good stat. There is a difference between Lacy and the collection of stiffs from the previous couple of years. Perhaps an explosive run rate stat, ie attempts per explosive run (10+ yards). I don't have time to compile that stat, but it might prove interesting.

The comment pbmax quoted is also suspect - carries are king? Carries are given by the coach based on his evaluation of the talent, which is subjective and could, obviously, be wrong!

[hoosier]

YPC is only a meaningful metric when it is put in some kind of context. But the same goes for other metrics including rushing attempts. Perfect example is Chris Johnson, who gets a ton of carries regardless of how he's doing simply because his team's offense has no other real threats.

Explosive runs might be a good metric with some teams, but for other offenses the more important variable might ability to pick up consistent positive yardage on 1st and 2nd downs. Even when he was good Chris Johnson would lose yardage on a high percentage of his runs. For an offense like Green Bay's that depends on maintaining manageable down-and-distance, Johnson would not be nearly as valuable as a guy like Lacy who consistently moves forward.

[3irty1]

YPC is about the best option you've got for comparing backs in the cases where there are two good RB's on one team or zero good RB's on one team. I understand using carries so to not be unfair to running backs on bad offenses where they are the only thing opposing defenses have to stop, but carries is unfair to guys who made the most of limited or shared opportunities. If you've got to go by just one stat I think its got to be yards.

[MadScientist]

Explosive runs might be a good metric with some teams, but for other offenses the more important variable might ability to pick up consistent positive yardage on 1st and 2nd downs. Even when he was good Chris Johnson would lose yardage on a high percentage of his runs. For an offense like Green Bay's that depends on maintaining manageable down-and-distance, Johnson would not be nearly as valuable as a guy like Lacy who consistently moves forward.

Not sure about that. How much will a DC scheme to stop a guy who gets 4 +/- 2 yards a carry vs a guy who has a fair number of -2 to +3 yard runs but will go 20-60 yards if the defense every 10th run, especially if the defense is napping?

[Just Jeff]

Liars figure and figures lie, but the absence of numbers is just people shooting thier mouths off, like bunch of opinionated fans on a message board. I'll go with the stats, like YPC. Carries are good too. I love stats.

[bobblehead]

Yards per carry are a terrible way to measure the contribution of a running back.

I have been saying this all along, haven't I?

http://www.footballperspective.com/t...rds-per-carry/

Chase Stuart

I'm not fully disagreeing. I have always said that the purpose of effectively running the ball is 1) Create favorable down and distance and 2) Keep a defense honest. BUT...I hate any argument that says "if you don't count these 3 really successful attempts, then Eric Dickerson sucks" They happened. They are part of the sample size. They count. You can't not count them.

[Cheesehead Craig]

Pro football focus does a nice job of rating players taking a lot of stuff into context. I think Lacy is the #5 RB in the NFL with their research. I put way more stock in their analysis than the overly simplistic ypc stat.

[hoosier]

Not sure about that. How much will a DC scheme to stop a guy who gets 4 +/- 2 yards a carry vs a guy who has a fair number of -2 to +3 yard runs but will go 20-60 yards if the defense every 10th run, especially if the defense is napping?

With the Packers I think the second kind of RB is just duplicating a threat that the passing game can carry out more effectively. If you're the DC facing the Packers and you have to choose between letting Rodgers beat you throwing or a Chris Johnson type break off the occasional long run sandwiched around a bunch of negative runs, that is a pretty easy choice to make. Chris Johnson's all-or-nothing is much less scary than Rodgers throwing against an eight man front. So is Lacy's four-to-six, except a defense can't let Lacy do that to it all day long.

[pbmax]

One point not explicitly made by Stuart but brought up by Brian Burke at Advanced Football Stats, is that because the NFL measures everything rounded to an yard, simple accuracy is missing in a lot of running back yardage stats. When you are trying to decipher whether a drop from a 4.5 ypc to 4.1 is a bad omen, it doesn't help that a lot of the fluctuation could be rounding.

[Joemailman]

One point not explicitly made by Stuart but brought up by Brian Burke at Advanced Football Stats, is that because the NFL measures everything rounded to an yard, simple accuracy is missing in a lot of running back yardage stats. When you are trying to decipher whether a drop from a 4.5 ypc to 4.1 is a bad omen, it doesn't help that a lot of the fluctuation could be rounding.

It could be, but isn't it more likely that rounding would even itself out over the course of a season?

[pbmax]

It could be, but isn't it more likely that rounding would even itself out over the course of a season?

On average, , yes.

But there are always exceptions.

The best point the article makes is the even/odd splits. The discrepancy points out that there are WAY too many other factors happening with the YPC stat to use it to judge a RBs performance.

[red]

one thing that i think really skews the YPC is the home runs. if a guy gets 1 or 2 70 or 80 yard td's in a season that really boosts his ypc.

a guy like eddie can just pound away and never break the long one, but his 250 other carries or so will be just as good as the other guys, or better

guy A carries the ball 200 times and gets 1000 yards. 5.0 yards per carry

the home run hitter also carries the ball 200 times, but had the two runs of 70 and 80. 1150 yards. 5.7 yards per carry

huge difference because of 2 carries

[3irty1]

With the Packers I think the second kind of RB is just duplicating a threat that the passing game can carry out more effectively. If you're the DC facing the Packers and you have to choose between letting Rodgers beat you throwing or a Chris Johnson type break off the occasional long run sandwiched around a bunch of negative runs, that is a pretty easy choice to make. Chris Johnson's all-or-nothing is much less scary than Rodgers throwing against an eight man front. So is Lacy's four-to-six, except a defense can't let Lacy do that to it all day long.

Just devils advocate but there is something to be said as well for a guy who can do a ton of damage with fewer snaps. Guys like that can keep play action viable without having to blow all the clock it takes to establish the run with a guy like Lacy. Generally shortening the game increases the likely hood of an upset so on a great passing team it could make sense to desire running the minimum amount of running plays that are required to keep defense honest.

[texaspackerbacker]

For once, I agree with you, 3irty1. I never liked that "shortening the game" concept when you have a guy like Aaron Rodgers. Shortening the game is what a weaker team wants to do to increase its chance of an upset. I also never liked the idea of a run-first offense. I saw that from the Packers for literally decades, some with success back to Lombardi, some not in the decades after. Even in the good years, it was frustrating, though. I like Eddie Lacy - as a change of pace threat and maybe to run clock late in the game with a lead, but I hate the idea of getting away from riding Aaron Rodgers' arm in a pass first offense. Right now, our star runner is a power guy, and our change of pace - Starks is more of a breakaway threat. What I would really like is if that was reversed - we had some big bruiser to use for short yardage, running clock late, etc., and our star would be somebody like Chris Johnson in his prime or A.P. in his prime - a home run threat. Lacy is like having Rod Carew or Ty Cobb - a singles hitter - batting cleanup.

[bobblehead]

one thing that i think really skews the YPC is the home runs. if a guy gets 1 or 2 70 or 80 yard td's in a season that really boosts his ypc.

a guy like eddie can just pound away and never break the long one, but his 250 other carries or so will be just as good as the other guys, or better

guy A carries the ball 200 times and gets 1000 yards. 5.0 yards per carry

the home run hitter also carries the ball 200 times, but had the two runs of 70 and 80. 1150 yards. 5.7 yards per carry

huge difference because of 2 carries

2 game changing carries....yea, we shouldn't count those.

[bobblehead]

For once, I agree with you, 3irty1. I never liked that "shortening the game" concept when you have a guy like Aaron Rodgers. Shortening the game is what a weaker team wants to do to increase its chance of an upset. I also never liked the idea of a run-first offense. I saw that from the Packers for literally decades, some with success back to Lombardi, some not in the decades after. Even in the good years, it was frustrating, though. I like Eddie Lacy - as a change of pace threat and maybe to run clock late in the game with a lead, but I hate the idea of getting away from riding Aaron Rodgers' arm in a pass first offense. Right now, our star runner is a power guy, and our change of pace - Starks is more of a breakaway threat. What I would really like is if that was reversed - we had some big bruiser to use for short yardage, running clock late, etc., and our star would be somebody like Chris Johnson in his prime or A.P. in his prime - a home run threat. Lacy is like having Rod Carew or Ty Cobb - a singles hitter - batting cleanup.

A weaker team, or a team with a great D that wants to limit the time you have to wear down that great D?

[texaspackerbacker]

A weaker team, or a team with a great D that wants to limit the time you have to wear down that great D?

Either way, it's the other guys, not us.