Outliers, YPC, and the Cowboys running game

[Hoofbite]

I disagree...............basically you are saying the RB doesn't matter, it is all about the blocking.

Top backs in this league have proven they are more valuable than just an average back. If you want proof, look no further than the Dallas Cowboys. They offered Murray $6 million a season and offered McFadden a happy meal to sign. The team obviously didn't think last year's rushing totals was mainly due to blocking, otherwise you don't offer a RB $6 million a season unless you think he personally is responsible for a lot of the yards himself.

Interesting thought.

 

[Idgit]

Devil's advocate, has anyone actually posted stats about the passing efficiency of a team like the Packers, or Patriots, in years they were able to run the ball more efficiently?

I'd assume that the efficiency of the passing game is affected positively by the *threat* of the run game in the teams with better QB's. I haven't really done any legwork on that assumption.

The only meaningful correlation between rushing effectiveness and wins that I've ever seen was related to short yardage and goal line rushing. If the Pats or the Packers were able to pass the ball effectively, they've been doing it irrespective of their ability to run the football.
The same obviously goes for the ToP argument for defense. If it helped teams to win games consistently, it'd show up in the form of more victories. It doesn't, because you can play good defense by stopping the pass effectively without dominating time of possession.

But that's not really what the thread is about. I took burm's bait because it's so tasty and hard to resist.

 

[Brooksey]

We're pretty clearly going to keep up the relatively high rushing percentage.

Agreed. I believe it will be the same or increased slightly due to the huge success we had last year "Passing Less". Last year we overworked Murray but this year we may be looking at 30-35 runs a game by committee.

I'd also expect us to continue to run on 1st down about the same with any change being more play action.

 

[MichaelWinicki]

This year you can expect to see the same or slightly more runs then last year but it will be more spread around. Combine that with a better defense and it should get us over the hump. Anyone who thinks we are going back to pass happy is mistaken, and anyone "Hoping" will be disappointed.

Oh I don't think they're going to be pass-happy like in 2012 & 2013.

I just wouldn't be surprised if the rushing attempt/passing attempt was closer to 50//50 than last season.

 

[ThreeandOut]

I disagree...............basically you are saying the RB doesn't matter, it is all about the blocking.

Top backs in this league have proven they are more valuable than just an average back. If you want proof, look no further than the Dallas Cowboys. They offered Murray $6 million a season and offered McFadden a happy meal to sign. The team obviously didn't think last year's rushing totals was mainly due to blocking, otherwise you don't offer a RB $6 million a season unless you think he personally is responsible for a lot of the yards himself.

Murray wasn't offered $6 million per year for the 7 long runs per year. Randle can probably give you those runs. Murray was offered $6 million per year because he can carry 25+ times per game and move the chains.

 

[AzorAhai]

Oh I don't think they're going to be pass-happy like in 2012 & 2013.

I just wouldn't be surprised if the rushing attempt/passing attempt was closer to 50//50 than last season.

Even if the run to pass ratio was similar to last season, I would like to see something closer to 50/50 on 1st downs. I fully believe that the stubbornness to run 75% on 1st downs cost them points last year.

What's your take on the 1st half struggles last year? To me it always felt like they couldn't get into a good rythm early in games and I think the 1st down play calling was a big factor.

 

[ABQcowboyJR]

Standard deviation would be interesting to look at as well.

 

[YosemiteSam]

Standard deviation would be interesting to look at as well.

You get to work on creating that bell curve. (or probability distribution if you prefer) I would be interested in seeing it.

 

[AbeBeta]

The expected points went from 15.2 to 34.6 after his top 7 runs, including one TD. Almost 2.8 points per run.

But we should take those runs out when calculating his value sez the OP

 

[Doomsday101]

You would think after 3 seasons of Dallas passing a ton and losing only to turn around and see Dallas actually run the ball more than pass it and win, that people would finally admit the importance of the run game but sadly no. As for Murray it is not just about yards per carry it is about blocking being able to pick up the blitzer who comes free as defense sends more rushers than you have blockers. It is about being able to grind out runs when defense knows full well you are going to run and it is about goal line an area this team has struggled with in the past but have not as soon as we started running more often. The rules over the years have changed but the run game for some teams is still a critical part of the offense. Murray alone was 40% of the offense that does not take into account his role as a blocker for Romo.

 

[AbeBeta]

Standard deviation would be interesting to look at as well.

No, it wouldn't. You've got a distribution that is non-normal and will always be so as runs of 5+ yards are fairly common but the loss of 5 or more yards is rare.

All looking at SD would tell us is whether someone failed to understand what they were taught in Stat 101

 

[MichaelWinicki]

Even if the run to pass ratio was similar to last season, I would like to see something closer to 50/50 on 1st downs. I fully believe that the stubbornness to run 75% on 1st downs cost them points last year.

What's your take on the 1st half struggles last year? To me it always felt like they couldn't get into a good rythm early in games and I think the 1st down play calling was a big factor.

I definitely think they ran too often on first down.

I'm not of the ilk that believes "You score too fast and you leave the defense vulnerable".

To me it's all about points. Luv GB's offense in the first half of games, light-up the defense of the other team and then run Lacey at them hard later on.

The Cowboys will not follow that model of heavy passing and I get that, but a little more passing especially on first down would be welcome in my eyes.

 

[Doomsday101]

I definitely think they ran too often on first down.

I'm not of the ilk that believes "You score too fast and you leave the defense vulnerable".

To me it's all about points. Luv GB's offense in the first half of games, light-up the defense of the other team and then run Lacey at them hard later on.

The Cowboys will not follow that model of heavy passing and I get that, but a little more passing especially on first down would be welcome in my eyes.

I would have liked to have seen a bit more play action on 1st down myself. I still would have used the run through out the game but at times felt the Cowboys did run too often on 1st.

 

[Deep_South]

I'm wondering what the league average would if all of the good runs are subtracted. Right now the comparison is being made between Murray and an average that includes his long runs, along with those of all of the other backs. Presumably the league average would go down if those were not included, although I have no idea by how much.

 

[MichaelWinicki]

I would have liked to have seen a bit more play action on 1st down myself. I still would have used the run through out the game but at times felt the Cowboys did run too often on 1st.

Yeah, I'm not advocating going back to how it was in '12 and '13, but another couple pass attempts per game on first down would potentially stop some of that run-blitz/8 men in a box stuff that was so prevalent over the last 1/4 of the season.

 

[Rockport]

We've talked a lot this offseason about the effects of removing Demarco Murray from the lineup and what it's likely going to do to change the sorts of defenses we'll have expected to run against, and the sort of production we're likely to have as a result. I was doing a little bit of research a few days ago on some other RBs who might be available for us later in the offseason here, and came across this article which I thought was pretty interesting. The basic gist of it is that, while a relatively high YPC over a long period of time is a very good barometer of the best backs in league history, statistically, that's maybe a bit misleading because the very highest YPC runs are so very few and far between. As a result, it's a very small number of running plays spread over the course of a long season that separates a productive runner for one who is average or below average.

A case in point from the article, using Demarco Murray as the example:




7 out of 393 plays. That's not very many. That's the number of plays we'd need over league average performance from our RBs to effectively match the sort of production we got from Murray last season. Whether that's coming from the guys on our roster, or instead we're getting it from the passing game where last season we might have run it. The vast majority of the runs we got from him in that fantastic season last season were not outliers, and the pattern more likely correlated with what we might see from other runners.

Now, this doesn't address the dirty runs argument that we've seen so much of. But then those dirty carries are rolled up in the other 386 carries that also happen to be fairly dependent on our ability to execute our blocking scheme effectively. In fact, it might even be the case that we see a higher percentage of these outlier YPC runs from the other backs, both McFadden and Randle have show the ability to that that more often than Demarco did.

What do you guys think? I was surprised how few of an exceptional RBs carries were actually unusual by statistical standards, and how much that affected the estimation of RB performance, overall. This also goes a ways towards explaining why rushing effectiveness might not be all that significant a factor in winning football games. You can easily see the net effect of Murray's 7 outlier runs from last season getting swamped in what would essentially be less than .5 incremental big passing plays/week. With QBs like Romo and Rodgers going up against QBs like RGIII and Foles, that half a big play isn't really that much to make up at all.

As far as the effectiveness of the running game you're ignoring the mental aspect of the offensive line controlling the line of scrimmage. When your line is physically dominating the defense it demoralize's them and affects their entire team. And it really doesn't matter how many rushing yards you get.

 

[Doomsday101]

As far as the effectiveness of the running game you're ignoring the mental aspect of the offensive line controlling the line of scrimmage. When your line is physically dominating the defense it demoralize's them and affects their entire team. And it really doesn't matter how many rushing yards you get.

Good point, for OL the run allows them to attack, in pass blocking you are defending. Let the OL impose their will on the defensive fronts and pound on them

 

[Brooksey]

Oh I don't think they're going to be pass-happy like in 2012 & 2013.

I just wouldn't be surprised if the rushing attempt/passing attempt was closer to 50//50 than last season.

Well it was 51% run to 49% pass last year.. if it went to 50-50 with a few more shots on play action (first down) I would not complain. Like I've said previously, the heavy run on first down was by design, they really like the 2nd and 7 or 2nd and 6 play call options. There were a few games where it seemed over-cooked, not just running on first down but running Murray 30 + times (while a rested Randle watched) or playing the starters the whole game with a playoff berth clinched.

I believe that is Jason Garrett making a statement.

 

[Idgit]

As far as the effectiveness of the running game you're ignoring the mental aspect of the offensive line controlling the line of scrimmage. When your line is physically dominating the defense it demoralize's them and affects their entire team. And it really doesn't matter how many rushing yards you get.

Not ignoring it, because I agree it's fun to see. It's just that if all that demoralization doesn't help you win games, I don't consider it a factor beyond making the offense more fun to watch.

Perhaps this season we'll play defense at a level that makes it harder for the good teams to score themselves after being demoralized. That's when we'll really be able to see the offense take it to the other team.

 

[Rockport]

Not ignoring it, because I agree it's fun to see. It's just that if all that demoralization doesn't help you win games, I don't consider it a factor beyond making the offense more fun to watch.

Perhaps this season we'll play defense at a level that makes it harder for the good teams to score themselves after being demoralized. That's when we'll really be able to see the offense take it to the other team.

It's the overriding factor in winning games. Always has been and always will be. Sure you can point out exceptions but the fact remains that if you control the offensive and defensive lines you're going to win a lot of football games.