Cowboys Stats - Rushing does help the Cowboys Win

[Nightman]

This is the formula I was referring to..
[net yards - (50 * int)) / drop backs]

Dallas 295 / 34 = 8.67

Pittsburgh 400 / 47 = 8.51

but if Zeke doesn't bust those 2 TDs runs Prescott's and Ben's passing numbers don't change but Pitt wins and Dal loses

Dak also gets credit for the 83 yd TD 'pass' to Zeke in the first half

 

[JD_KaPow]

I just showed you 100% correlation. It does not get better than that.

Repeat after me: correlation is not causation. Correlation is not causation.

(Actually, in this case, there is causation, but it goes the other way. You run a lot because you're winning, not the other way around. You pass a lot because you're losing, not the other way around).

 

[Doomsay]

But they won the passing efficiency formula that Adam uses. Even though Ben passed for more yards, Dak scored higher on the passing efficiency formula

He has the 2nd highest play action pass efficiency in the league - how is that factored in to the inherent collinearity of the passing efficiency stat?

 

[xwalker]

Repeat after me: correlation is not causation. Correlation is not causation.

(Actually, in this case, there is causation, but it goes the other way. You run a lot because you're winning, not the other way around. You pass a lot because you're losing, not the other way around).

There is no proof that the Passing Efficiency correlates to winning is cause or effect. Does Passing Efficiently cause winning or does Winning cause Passing Efficiently. Offenses and defenses have been know to vary what they do when they are up or down on the scoreboard. Sometimes even when they are just up by 1 score. It is not just a 4th quarter issue. It can happen in the 1st quarter. Once example is that defenses will go more pass rush and less run defense when their own team is up on the scoreboard. There a numerous examples in something as complex as trying to define winning in a football game.

 

[percyhoward]

Dak also gets credit for the 83 yd TD 'pass' to Zeke in the first half

So does Zeke.

Most games, 150+ yards from scrimmage, rookie season
6 Billy Sims 1980
5 Ezekiel Elliott 2016*
5 Curtis Martin 1995
5 Barry Sanders 1989
5 Eric Dickerson 1983
5 Ottis Anderson 1979

 

[JD_KaPow]

There is no proof that the Passing Efficiency correlates to winning is cause or effect. Does Passing Efficiently cause winning or does Winning cause Passing Efficiently. Offenses and defenses have been know to vary what they do when they are up or down on the scoreboard. Sometimes even when they are just up by 1 score. It is not just a 4th quarter issue. It can happen in the 1st quarter. Once example is that defenses will go more pass rush and less run defense when their own team is up on the scoreboard. There a numerous examples in something as complex as trying to define winning in a football game.

Actually, this is also true. Teams that are behind have to adopt riskier strategies that are less efficient. I don't know how much of an effect this has. But I do know for sure that pointing to rushing vs. passing yardage volume as a cause of winning or losing is incorrect.

 

[Nightman]

So does Zeke.

Most games, 150+ yards from scrimmage, rookie season
6 Billy Sims 1980
5 Ezekiel Elliott 2016*
5 Curtis Martin 1995
5 Barry Sanders 1989
5 Eric Dickerson 1983
5 Ottis Anderson 1979

Air yards to ground yards on that play- 1 to 83

Zeke could use them as rush yards to catch Dickerson

 

[gmoney112]

This is the formula I was referring to..
[net yards - (50 * int)) / drop backs]

Dallas 295 / 34 = 8.67

Pittsburgh 400 / 47 = 8.51

Nice. I'm going to start using that more. And yeah, haha I've definitely seen the discussions, I mean, it really is a fact that passer rating differential and TO differential are statistically strong indicators of winners, that's not really debatable because it has such a strong precedent. Was looking for the Aikman formula, in particular, so thanks and it was cool of you to work it out.

I did some research last year maybe on the success of Dallas' run game in 2014, when I think we led the league in first down yardage, leading to easier to convert(higher % scenarios on subsequent downs) in large part to our run game. There was a correlation in subsequent increased efficiency and YPA when working with higher percentage scenarios.

I don't think anyone with eyes can tell you that the run game has not been a huge boon in our wins, whether that's from teams not being able to play purely coverage schemes leading to better chances of completions, or whether it's more favorable down and distance scenarios leading to an expansion of the playbook, and thereby more unpredictably for defenses to cover, leading to more success.

It's hard to pinpoint with this team because there's been no OL like this really in the modern era, not with this kind of quality QB and RB. It's a complete outlier. Statistical precedent can still correlate, but as far as causation, it genuinely lends to the notion that the total is greater than the sum of the parts.

 

[AdamJT13]

AdamJT13 disagrees with you.

No, I don't.

Of course winning teams often have more rushing yards, and losing teams often have more passing yards. That's because teams that are already losing usually pass to catch up, while teams that are already winning don't need to pass as much and can run more in the second half. It's simply confusing cause and effect.

Yards per play is a much different story.

 

[Hoov]

but if Zeke doesn't bust those 2 TDs runs Prescott's and Ben's passing numbers don't change but Pitt wins and Dal loses

Dak also gets credit for the 83 yd TD 'pass' to Zeke in the first half

that is possible. Or Dak might have passed and scored on the play following the first run which would have changed Daks numbers but it would have been a positive change - and on the second its very likely Bailey would have made that FG if zeke didn't bust the second run so at that point cowboys still would have won as they only needed a FG.

But, I was sure glad I didn't have to sit through a nailbiting FG attempt thanks to Zeke and the line.

 

[Hoov]

He has the 2nd highest play action pass efficiency in the league - how is that factored in to the inherent collinearity of the passing efficiency stat?

I think your question is what impact does the play action success have on the passing efficiency ? I would have to say I don't know the answer to that.

 

[Hoov]

Nice. I'm going to start using that more. And yeah, haha I've definitely seen the discussions, I mean, it really is a fact that passer rating differential and TO differential are statistically strong indicators of winners, that's not really debatable because it has such a strong precedent. Was looking for the Aikman formula, in particular, so thanks and it was cool of you to work it out.

I did some research last year maybe on the success of Dallas' run game in 2014, when I think we led the league in first down yardage, leading to easier to convert(higher % scenarios on subsequent downs) in large part to our run game. There was a correlation in subsequent increased efficiency and YPA when working with higher percentage scenarios.

I don't think anyone with eyes can tell you that the run game has not been a huge boon in our wins, whether that's from teams not being able to play purely coverage schemes leading to better chances of completions, or whether it's more favorable down and distance scenarios leading to an expansion of the playbook, and thereby more unpredictably for defenses to cover, leading to more success.

It's hard to pinpoint with this team because there's been no OL like this really in the modern era, not with this kind of quality QB and RB. It's a complete outlier. Statistical precedent can still correlate, but as far as causation, it genuinely lends to the notion that the total is greater than the sum of the parts.

Just to be clear though - I dont want you to credit me for that - I didn't do much but post the formula that Adam posted here before, after Percy gave it to me today. I asked him for it because I thought based on watching the game and how good Ben was that steelers would have had the higher efficiency but they did not.

 

[gmoney112]

Just to be clear though - I dont want you to credit me for that - I didn't do much but post the formula that Adam posted here before, after Percy gave it to me today. I asked him for it because I thought based on watching the game and how good Ben was that steelers would have had the higher efficiency but they did not.

I know man. All good. Just giving you some props for actually using it for that game. Quite a few of us do appreciate that kind of stuff, and I've been way too busy to do it.

 

[AdamJT13]

but if Zeke doesn't bust those 2 TDs runs Prescott's and Ben's passing numbers don't change but Pitt wins and Dal loses

That's quite an assumption (nothing would have changed except those two plays?), and it's probably not correct. We were down by one point both times, and we were within field goal range both times. Unless you think Bailey would have shanked about a 29-yard field goal if not for Elliott's 14-yard TD run, we would have at least taken the lead, probably with less than a minute to play and the Steelers out of timeouts. And then it would have been up to our pass defense, which is the half of the equation that everyone conveniently forgets.

 

[AdamJT13]

Actually, this is also true. Teams that are behind have to adopt riskier strategies that are less efficient. I don't know how much of an effect this has.

Actually, adopting a riskier strategy tends to increase ANYPA (interception rate goes up, but so does yards per attempt), while taking a conservative passing strategy after you're ahead tends to reduce ANYPA (interception rate goes down, but so does yards per attempt).

 

[DallasInDC]

So does Zeke.

Most games, 150+ yards from scrimmage, rookie season
6 Billy Sims 1980
5 Ezekiel Elliott 2016*
5 Curtis Martin 1995
5 Barry Sanders 1989
5 Eric Dickerson 1983
5 Ottis Anderson 1979

That's mighty good company Zeke's keeping...I have a feeling we will see his name next to the best RBs in history for the rest of his career.

 

[TwoDeep3]

35-30. Here is a stat as long as Dallas has 35, they win every time.

 

[Doomsay]

I think your question is what impact does the play action success have on the passing efficiency ? I would have to say I don't know the answer to that.

Partially. It's more that what people are using as independent variables are not, in fact, independent, especially on an individual team basis. Play action passing efficiency is just a good illustration of that. Dak was 2nd in the league in PA efficiency and his completion % differential was inordinately higher than the league average after a previous play rushing gain of 5 yards or more. Ergo, selective rushing success, creates passing success which in turn, generates scoring efficiency per drive.

The head-to-head rushing / passing stats are patently simplistic yet available, that's what this argument has been based on for the past few years. Not saying that you were proffering that theory though.

 

[JD_KaPow]

Actually, adopting a riskier strategy tends to increase ANYPA (interception rate goes up, but so does yards per attempt), while taking a conservative passing strategy after you're ahead tends to reduce ANYPA (interception rate goes down, but so does yards per attempt).

Hmm. If we believe that ANYPA is a good proxy for offensive efficiency, this would suggest that risk-neutral strategy is much more aggressive than teams currently use (which I can believe, given that NFL teams are extraordinarily risk-averse). But this has to break down and go the other way at some point (I can't believe that throwing a Hail Mary on every play would have a higher ANYPA: the minuscule completion rate, not to mention the sacks, would have to outweigh the higher yards per completion).

Ah, I think sacks are not included in ANYPA? That would explain some of this seeming discrepancy.

 

[percyhoward]

.Ah, I think sacks are not included in ANYPA?

They are. That's the "net" in Adjusted Net Yards Per Attempt. It's really more "per dropback" than "per attempt."