[washingtonismoney]

In an article about Stanford's ground game, Shaw says:

In two games against flimsy opposition, the Cardinal has mustered just 3.2 yards per rush.

Three. Point. Two.

"We've had too many little things keep us from our four yard average,'' coach David Shaw said.

I mean, four yards per carry is...not good. And well below previous years' standards. Baby steps, I guess, but still.

http://www.mercurynews.com/sports/ci_288...um=twitter

[washingtonismoney]

In an article about Stanford's ground game, Shaw says:

In two games against flimsy opposition, the Cardinal has mustered just 3.2 yards per rush.

Three. Point. Two.

"We've had too many little things keep us from our four yard average,'' coach David Shaw said.

I mean, four yards per carry is...not good. And well below previous years' standards. Baby steps, I guess, but still.

http://www.mercurynews.com/sports/ci_288...um=twitter

[BostonCard]

Year: YPC (RB/QB)
2008: 4.9 (Gerhart/Pritchard)
2009: 5.3 (Gerhart/Luck)
2010: 5.2 (Taylor/Luck)
2011: 5.3 (Taylor/Luck)
2012: 4.4 (Taylor/Hogan)
2013: 5.0 (Gaffney/Hogan)
2014: 4.3 (Wright/Hogan)
2015: 3.2 (McCaffrey/Hogan)

yuck.

BC

[BostonCard]

Year: YPC (RB/QB)
2008: 4.9 (Gerhart/Pritchard)
2009: 5.3 (Gerhart/Luck)
2010: 5.2 (Taylor/Luck)
2011: 5.3 (Taylor/Luck)
2012: 4.4 (Taylor/Hogan)
2013: 5.0 (Gaffney/Hogan)
2014: 4.3 (Wright/Hogan)
2015: 3.2 (McCaffrey/Hogan)

yuck.

BC

[CTcard]

You can learn something a little different by looking at yds/carry from the RBs. That tells us more about the bread and butter running plays and takes out QB scrambles/sacks, end arounds, and all that. It says a little more about whether you can operate the basic run game, OL can get some push, etc.

It's similar to the numbers overall, but a few nuances.
              RB
Year    yds/carry    #carries/gm
2008      5.54            30.3
2009      5.41            33.7
2010      4.97            30.0
2011      5.41            33.3
2012      4.64            29.1
2013      5.11            32.4
2014      4.94            25.0
2015      4.01            26.0
------------------------------------
vs NU      5.00            19.0
vs UCF    3.52            33.0

So, we saw a notable drop in effectiveness when Gerhart left, but not much of a drop in focusing on the run.
2011 was just a loaded team - that was our best offense ever.
2012 we dropped off in effectiveness but kept running the ball.
2013 was getting back on track.
2014 was actually okay in yds/carry, but to some extent we abandoned the run game until the final three games.
2015 is way off so far - but yds/carry were actually okay in the Northwestern game (at least compared to the rest of the offense) but we didn't run much. Against UCF we kept at it despite struggling in yds/carry - which maybe is what opened up the pass game?

[CTcard]

You can learn something a little different by looking at yds/carry from the RBs. That tells us more about the bread and butter running plays and takes out QB scrambles/sacks, end arounds, and all that. It says a little more about whether you can operate the basic run game, OL can get some push, etc.

It's similar to the numbers overall, but a few nuances.
              RB
Year    yds/carry    #carries/gm
2008      5.54            30.3
2009      5.41            33.7
2010      4.97            30.0
2011      5.41            33.3
2012      4.64            29.1
2013      5.11            32.4
2014      4.94            25.0
2015      4.01            26.0
------------------------------------
vs NU      5.00            19.0
vs UCF    3.52            33.0

So, we saw a notable drop in effectiveness when Gerhart left, but not much of a drop in focusing on the run.
2011 was just a loaded team - that was our best offense ever.
2012 we dropped off in effectiveness but kept running the ball.
2013 was getting back on track.
2014 was actually okay in yds/carry, but to some extent we abandoned the run game until the final three games.
2015 is way off so far - but yds/carry were actually okay in the Northwestern game (at least compared to the rest of the offense) but we didn't run much. Against UCF we kept at it despite struggling in yds/carry - which maybe is what opened up the pass game?

[washingtonismoney]

Well, to further decompose running, there are a lot of elements that go into the play -- the skill of the RBs, the effectiveness of the OL, the alignment of the defense (are they stacking the box), among other stuff.

Against NW, the 5.0 ypc is very much due to the quality of the RBs, in my estimation. Take out the McCaffrey 26-yard run (due partially to defensive misalignment and partially to McCaffrey's quality), and the running back yards-per-carry falls to around 3.5.

[washingtonismoney]

Well, to further decompose running, there are a lot of elements that go into the play -- the skill of the RBs, the effectiveness of the OL, the alignment of the defense (are they stacking the box), among other stuff.

Against NW, the 5.0 ypc is very much due to the quality of the RBs, in my estimation. Take out the McCaffrey 26-yard run (due partially to defensive misalignment and partially to McCaffrey's quality), and the running back yards-per-carry falls to around 3.5.

[CTcard]

Against NW, the 5.0 ypc is very much due to the quality of the RBs, in my estimation. Take out the McCaffrey 26-yard run (due partially to defensive misalignment and partially to McCaffrey's quality), and the running back yards-per-carry falls to around 3.5.

Yep. Take out the good runs and the ypc goes down.

That particular play had good blocking by the line, though it wasn't power just straight ahead hit your guy. It was aided by the two NU LBers running to the same gap, and impossible to tell what would have happened without that. McCaffrey had a nice deke at about 12 yards, but he didn't have to do much before 12 yards downfield.

[CTcard]

Against NW, the 5.0 ypc is very much due to the quality of the RBs, in my estimation. Take out the McCaffrey 26-yard run (due partially to defensive misalignment and partially to McCaffrey's quality), and the running back yards-per-carry falls to around 3.5.

Yep. Take out the good runs and the ypc goes down.

That particular play had good blocking by the line, though it wasn't power just straight ahead hit your guy. It was aided by the two NU LBers running to the same gap, and impossible to tell what would have happened without that. McCaffrey had a nice deke at about 12 yards, but he didn't have to do much before 12 yards downfield.

[81alum]

[quote author=washingtonismoney link=topic=12825.msg129046#msg129046 date=1442628703]
Against NW, the 5.0 ypc is very much due to the quality of the RBs, in my estimation. Take out the McCaffrey 26-yard run (due partially to defensive misalignment and partially to McCaffrey's quality), and the running back yards-per-carry falls to around 3.5.

Yep. Take out the good runs and the ypc goes down.

That particular play had good blocking by the line, though it wasn't power just straight ahead hit your guy. It was aided by the two NU LBers running to the same gap, and impossible to tell what would have happened without that. McCaffrey had a nice deke at about 12 yards, but he didn't have to do much before 12 yards downfield.
[/quote]
So wouldn't the median yards per carry be a better measure than the mean yards per carry?  It would do a better mob of measuring consistency in the running game, although of course it would not focus on big running plays.  To the extent that our running game is thought to be part of a ball-control, clock eating strategy, consistency would be more important to measure than big plays.

[81alum]

[quote author=washingtonismoney link=topic=12825.msg129046#msg129046 date=1442628703]
Against NW, the 5.0 ypc is very much due to the quality of the RBs, in my estimation. Take out the McCaffrey 26-yard run (due partially to defensive misalignment and partially to McCaffrey's quality), and the running back yards-per-carry falls to around 3.5.

Yep. Take out the good runs and the ypc goes down.

That particular play had good blocking by the line, though it wasn't power just straight ahead hit your guy. It was aided by the two NU LBers running to the same gap, and impossible to tell what would have happened without that. McCaffrey had a nice deke at about 12 yards, but he didn't have to do much before 12 yards downfield.
[/quote]
So wouldn't the median yards per carry be a better measure than the mean yards per carry?  It would do a better mob of measuring consistency in the running game, although of course it would not focus on big running plays.  To the extent that our running game is thought to be part of a ball-control, clock eating strategy, consistency would be more important to measure than big plays.

[Mick]

Year: YPC (RB/QB)
2008: 4.9 (Gerhart/Pritchard)
2009: 5.3 (Gerhart/Luck)
2010: 5.2 (Taylor/Luck)
2011: 5.3 (Taylor/Luck)
2012: 4.4 (Taylor/Hogan)
2013: 5.0 (Gaffney/Hogan)
2014: 4.3 (Wright/Hogan)
2015: 3.2 (McCaffrey/Hogan)

yuck.

BC

What's the ypc once you take out the portion of the game that the fourth rate opponent played at 1:00 a.m. body clock time?

[Mick]

Year: YPC (RB/QB)
2008: 4.9 (Gerhart/Pritchard)
2009: 5.3 (Gerhart/Luck)
2010: 5.2 (Taylor/Luck)
2011: 5.3 (Taylor/Luck)
2012: 4.4 (Taylor/Hogan)
2013: 5.0 (Gaffney/Hogan)
2014: 4.3 (Wright/Hogan)
2015: 3.2 (McCaffrey/Hogan)

yuck.

BC

What's the ypc once you take out the portion of the game that the fourth rate opponent played at 1:00 a.m. body clock time?

[BostonCard]

[quote author=Boston Card link=topic=12825.msg129026#msg129026 date=1442621718]
Year: YPC (RB/QB)
2008: 4.9 (Gerhart/Pritchard)
2009: 5.3 (Gerhart/Luck)
2010: 5.2 (Taylor/Luck)
2011: 5.3 (Taylor/Luck)
2012: 4.4 (Taylor/Hogan)
2013: 5.0 (Gaffney/Hogan)
2014: 4.3 (Wright/Hogan)
2015: 3.2 (McCaffrey/Hogan)

yuck.

BC

What's the ypc once you take out the portion of the game that the fourth rate opponent played at 1:00 a.m. body clock time?
[/quote]

Actually, it doesn't change.  As CTCard noted, we did not run well against UCF.  Our YPC was 3.2 against Northwestern and 3.2 against UCF.  However, if you look at just the running backs, we actually ran better against NW than UCF.

BC

[BostonCard]

[quote author=Boston Card link=topic=12825.msg129026#msg129026 date=1442621718]
Year: YPC (RB/QB)
2008: 4.9 (Gerhart/Pritchard)
2009: 5.3 (Gerhart/Luck)
2010: 5.2 (Taylor/Luck)
2011: 5.3 (Taylor/Luck)
2012: 4.4 (Taylor/Hogan)
2013: 5.0 (Gaffney/Hogan)
2014: 4.3 (Wright/Hogan)
2015: 3.2 (McCaffrey/Hogan)

yuck.

BC

What's the ypc once you take out the portion of the game that the fourth rate opponent played at 1:00 a.m. body clock time?
[/quote]

Actually, it doesn't change.  As CTCard noted, we did not run well against UCF.  Our YPC was 3.2 against Northwestern and 3.2 against UCF.  However, if you look at just the running backs, we actually ran better against NW than UCF.

BC

[BostonCard]

So wouldn't the median yards per carry be a better measure than the mean yards per carry?  It would do a better mob of measuring consistency in the running game, although of course it would not focus on big running plays.  To the extent that our running game is thought to be part of a ball-control, clock eating strategy, consistency would be more important to measure than big plays.

Yes, the median is less susceptible to outliers than the mean.  However, the mean is readily available; calculating the median would require compiling each individual play and extracting the median from that, since nobody readily calculates it.  Moreover, since runs are discrete, chances are that our median will be 4 yards for most years, so the median would hide a lot of variability.

All summary statistics require a sacrifice of information in order to tell "the story" concisely.  Thus, no single stat can tell the whole story, and all stats require context.  That's why I appreciate the narratives that WIM and CTCard provide to give a fuller meaning to the statistics.

[BostonCard]

So wouldn't the median yards per carry be a better measure than the mean yards per carry?  It would do a better mob of measuring consistency in the running game, although of course it would not focus on big running plays.  To the extent that our running game is thought to be part of a ball-control, clock eating strategy, consistency would be more important to measure than big plays.

Yes, the median is less susceptible to outliers than the mean.  However, the mean is readily available; calculating the median would require compiling each individual play and extracting the median from that, since nobody readily calculates it.  Moreover, since runs are discrete, chances are that our median will be 4 yards for most years, so the median would hide a lot of variability.

All summary statistics require a sacrifice of information in order to tell "the story" concisely.  Thus, no single stat can tell the whole story, and all stats require context.  That's why I appreciate the narratives that WIM and CTCard provide to give a fuller meaning to the statistics.