The Good, The Bad, The Ugly, and Down the Line

[NoDak]

@NoDak 's Pollard hate scares people, that hot-linkin sonofabitch!

My blood boils at the very mention of his name.

Apparently.

 

[boozeman]

My blood boils at the very mention of his name.

Apparently.

You are apparently a "hater".

 

[data]

Coach D, we barely knew ye

Coach Garrett got a decade. Coach D got one thread

It ain’t in ya when it just ain’t in ya

 

[shoop]

YPC is a TERRIBLE measure of a RB's effectiveness.

Especially for a backup with a backup's volume of carries.

Troy Hambrick averaged 5.1 YPC one year.

It means jack squat.

Here's a good case in point with Pollard. Look at his start against SF last season.

He had 12 carries for 69 yards. That's over 5 YPC which of course is great.

But one of his runs was for 40 yards. Take that one outlier run out and he's at 11 carries for 29 yards, which is 2 something YPC and dreadful.

You'd see that with Joseph Randle too.

I'm not saying Pollard is bad, I think he's pretty good.

But he's not a starter. He's more of a Kamara type who is at his best in a secondary role where they can use him in optimal situations rather than burning him up grinding out yards.

I was just looking for a comparison point. YPC is about the only fair equivalency unless the eye test works and you can see every game.

 

[Genghis Khan]

I was just looking for a comparison point. YPC is about the only fair equivalency unless the eye test works and you can see every game.

It's not fair at all actually. It's one of the worst and most deceiving stats in sports. The best it can tell you is maybe who is bad.

I don't want to get overly argumentative so I'll just present a couple examples.

The top all time YPC running back is Marion Motley. Number 2 is Jamaal Charles. Number 3 is Jim Brown. But number 4 is Mercury Morris. Number 5 is Joe Perry.

Other than Brown, does that look like anybody in the world's top 5?

The next 5 are gale Sayers, Barry Sanders, Derrick Henry, Napoleon Kaufman, and Paul Lowe.

Other than Sanders and Brown (and hell, Sanders is too low), is that even close to anybody's top 10?

It's a terrible stat.

One more example.

Joseph Randle's career YPC is 4.5.

Emmitt Smith's career YPC is 4.2.

Who was the better player?

It's a terrible stat.

 

[shoop]

It's not fair at all actually. It's one of the worst and most deceiving stats in sports. The best it can tell you is maybe who is bad.

I don't want to get overly argumentative so I'll just present a couple examples.

The top all time YPC running back is Marion Motley. Number 2 is Jamaal Charles. Number 3 is Jim Brown. But number 4 is Mercury Morris. Number 5 is Joe Perry.

Other than Brown, does that look like anybody in the world's top 5?

The next 5 are gale Sayers, Barry Sanders, Derrick Henry, Napoleon Kaufman, and Paul Lowe.

Other than Sanders and Brown (and hell, Sanders is too low), is that even close to anybody's top 10?

It's a terrible stat.

One more example.

Joseph Randle's career YPC is 4.5.

Emmitt Smith's career YPC is 4.2.

Who was the better player?

It's a terrible stat.

I agree about not being argumentative but my question/point/something was if not ypc and you can’t just say eye test, then how do you define great vs good vs only good in situations?
honest question. Longevity, touchdowns and yards all count buteach Can be misleading.

 

[Genghis Khan]

I agree about not being argumentative but my question/point/something was if not ypc and you can’t just say eye test, then how do you define great vs good vs only good in situations?
honest question. Longevity, touchdowns and yards all count buteach Can be misleading.

That's a good question. I'm not sure I've seen one single reliable running back stat.

The most reliable stats if you want a stat that correlates as close to the eye test as possible is probably yards or attempts, which honestly seems ridiculous (keep in mind Parcells said something along the lines of attempts being the most important RB stat so maybe it's not so crazy).

But look at the all time leaders in rushing attempts. Outside of missing Brown for obvious reasons it's a who's who of most of the greatest RBs. There's a few outliers that I don't like (Gore, Martin, and Bettis) but otherwise that's a pretty solid list. Sanders should be higher but again, you can explain it easily.

1	Emmitt Smith+	4,409	1990-2004	2TM
2	Walter Payton+	3,838	1975-1987	chi
3	Frank Gore	3,735	2005-2020	5TM
4	Curtis Martin+	3,518	1995-2005	2TM
5	Jerome Bettis+	3,479	1993-2005	2TM
6	Adrian Peterson	3,192	2007-2020	5TM
7	LaDainian Tomlinson+	3,174	2001-2011	2TM
8	Barry Sanders+	3,062	1989-1998	det
9	Edgerrin James+	3,028	1999-2009	3TM
10	Marcus Allen+	3,022	1982-1997	2TM
11	Eric Dickerson+	2,996	1983-1993	4TM
12	Franco Harris+	2,949	1972-1984	2TM
13	Tony Dorsett+	2,936	1977-1988	2TM
14	John Riggins+	2,916	1971-1985	2TM

Yards makes even more sense than attempts when you look at the all time list. To be clear, I don't think these stats are perfect. But if I'm going to pick one stat to judge a RB it's probably yards, and next it would be attempts.

1	Emmitt Smith+	18,355	1990-2004	2TM
2	Walter Payton+	16,726	1975-1987	chi
3	Frank Gore	16,000	2005-2020	5TM
4	Barry Sanders+	15,269	1989-1998	det
5	Adrian Peterson	14,820	2007-2020	5TM
6	Curtis Martin+	14,101	1995-2005	2TM
7	LaDainian Tomlinson+	13,684	2001-2011	2TM
8	Jerome Bettis+	13,662	1993-2005	2TM
9	Eric Dickerson+	13,259	1983-1993	4TM
10	Tony Dorsett+	12,739	1977-1988	2TM
11	Jim Brown+	12,312	1957-1965	cle
12	Marshall Faulk+	12,279	1994-2005	2TM
13	Edgerrin James+	12,246	1999-2009	3TM
14	Marcus Allen+	12,243	1982-1997	2TM
15	Franco Harris+	12,120	1972-1984	2TM

 

[Cowboysrock55]

I like YPC to some extent but you have to have similar carries for it to mean much. YPC means nothing when one running back is carrying it 20+ times a game and the other is getting 10 carries in passing situations. The back running the ball on 3rd and 1 is going to have a lower YPC than a back carrying the ball on 3rd and 15. Just as an example.

 

[Genghis Khan]

I value consistently good carries in a running back above all else.

Let's say you give running back A the ball 6 straight times. He gets 4 yards every time. It's only 4 YPC but much more importantly it's 2 1st downs and you're driving.

Let's say you then give running back B the ball. He gets 20 yards on his 1st carry. Great! His next three carries go for 1 yard, 1 yard, and 2 yards. It's now 4th and 6 and you're punting.

Running back A has 4 YPC and 24 yards.

Running back B had 6 YPC and 24 yards.

But who did better?

Running back A, because the consistency of his carries was way more important and resulted in 1st downs. Running back B's carries, even though he had that one flashy run, resulted in a punt.

This is why you can't rely on YPC. Running back B had a higher YPC by a lot, but had a negative net impact.

Who had more carries though? Running back A, and it told you a better story about the drive.

I wish there was a stat that measured a running back's consistent production but I'm not aware of one.

 

[data]

I value consistently good carries in a running back above all else.

Let's say you give running back A the ball 6 straight times. He gets 4 yards every time. It's only 4 YPC but much more importantly it's 2 1st downs and you're driving.

Let's say you then give running back B the ball. He gets 20 yards on his 1st carry. Great! His next three carries go for 1 yard, 1 yard, and 2 yards. It's now 4th and 6 and you're punting.

Running back A has 4 YPC and 24 yards.

Running back B had 6 YPC and 24 yards.

But who did better?

Running back A, because the consistency of his carries was way more important and resulted in 1st downs. Running back B's carries, even though he had that one flashy run, resulted in a punt.

This is why you can't rely on YPC. Running back B had a higher YPC by a lot, but had a negative net impact.

Who had more carries though? Running back A, and it told you a better story about the drive.

I wish there was a stat that measured a running back's consistent production but I'm not aware of one.

What you're discussing is essentially the premise that launched Football Outsiders with their DVOA metric (which incorporates game/score situation, down & distance (success %), strength of defense faced). Since then, their DYAR metric has added quantity of rushes.

To better ground this DYAR metric, here's an article by Football Outsiders to rank their top RBs of the 2010 decade.

Quick Reads Decade in Review: RB Totals

Our look back at the best and worst running backs of the 2010s includes a lot of LeSean McCoy, Marshawn Lynch, and a gaggle of assorted Chargers.

www.footballoutsiders.com

 

[Genghis Khan]

What you're discussing is essentially the premise that launched Football Outsiders with their DVOA metric (which incorporates game/score situation, down & distance (success %), strength of defense faced). Since then, their DYAR metric has added quantity of rushes.

To better ground this DYAR metric, here's an article by Football Outsiders to rank their top RBs of the 2010 decade.

Quick Reads Decade in Review: RB Totals

Our look back at the best and worst running backs of the 2010s includes a lot of LeSean McCoy, Marshawn Lynch, and a gaggle of assorted Chargers.

www.footballoutsiders.com

I generally like Football Outsiders.

This looks good and I think it's probably reliable.

Even then though, there is a bunch of subjectiveness written in. At first glance I don't love factoring in the situation and strength of defense faced.

For example, the Eagles only ran for like 60 yards against us the other night, which makes it look like good run defense. But we know in context they only tried to run a ridiculously low amount of times, and their running back was successful in both of his attempts. How do you rate our run defense in that situation? There are a lot of rabbit holes to go down.

 

[data]

I generally like Football Outsiders.

This looks good and I think it's probably reliable.

Even then though, there is a bunch of subjectiveness written in. At first glance I don't love factoring in the situation and strength of defense faced.

For example, the Eagles only ran for like 60 yards against us the other night, which makes it look like good run defense. But we know in context they only tried to run a ridiculously low amount of times, and their running back was successful in both of his attempts. How do you rate our run defense in that situation? There are a lot of rabbit holes to go down.

I’m confident their metrics take into account there were only 3 RB rushes.

Again, their whole premise is a 4 yard gain is more valuable on 1st and 10 then a 10 yard gain on 4th and 20.

Their evaluation of defenses take into account garbage time, I’m guessing if a team is winning by 21 points in the fourth quarter, they probably downgrade the influence of stats racked up into that DVOA metric and/or compare how other defenses performed when winning by 21 points.

We don’t know the algorithms, but I like game situation for sure. Completing a 4th qtr 20-yard pass in a 41-14 vs 21-23 game makes a huge difference

Same thing with field position as value for a 4-yard run from your own 20 vs red zone differs.

 

[ravidubey]

I generally like Football Outsiders.

This looks good and I think it's probably reliable.

Even then though, there is a bunch of subjectiveness written in. At first glance I don't love factoring in the situation and strength of defense faced.

For example, the Eagles only ran for like 60 yards against us the other night, which makes it look like good run defense. But we know in context they only tried to run a ridiculously low amount of times, and their running back was successful in both of his attempts. How do you rate our run defense in that situation? There are a lot of rabbit holes to go down.

I'm a fan of DYAR/Game

 

[p1_]

I generally like Football Outsiders.

This looks good and I think it's probably reliable.

Even then though, there is a bunch of subjectiveness written in. At first glance I don't love factoring in the situation and strength of defense faced.

For example, the Eagles only ran for like 60 yards against us the other night, which makes it look like good run defense. But we know in context they only tried to run a ridiculously low amount of times, and their running back was successful in both of his attempts. How do you rate our run defense in that situation? There are a lot of rabbit holes to go down.

You can’t use that game to judge our run D. You must use the whole of 3 games. And that is pretty indicative of a good run D.

 

[Genghis Khan]

You can’t use that game to judge our run D. You must use the whole of 3 games. And that is pretty indicative of a good run D.

Maybe, but I'm not convinced we've been really challenged yet.

 

[Rev]

Maybe, but I'm not convinced we've been really challenged yet.

We haven't. Philly with 3 runs. Tampa didnt have to. Chargers had a little success but I think they ran only to break up the consecutive passes. They didn't seem to focus on running the ball.

 

[boozeman]

Maybe, but I'm not convinced we've been really challenged yet.

If Carolina means what they say, they won't be running under ten times on Sunday.

 

[Cowboysrock55]

We haven't. Philly with 3 runs. Tampa didnt have to. Chargers had a little success but I think they ran only to break up the consecutive passes. They didn't seem to focus on running the ball.

With Parsons at LBer I feel great about the run defense.

With Jaylon? Not so much.

 

[p1_]

were the least tested run defense with just 45 attempts against. We don't know what we have just yet, but it hasnt been at all like last season.

 

[shoop]

thanks. That makes a lot more sense. I agree that a stat like football outsiders is the closest we have. I would be interested in the info on a conversational basis but not interested in paying for a subscription fee. Good to know it is there if I ever get curious enough.