The CardBoard
ATTN: David Shaw; RE: Run-Run-Pass - Printable Version

+- The CardBoard (https://thecardboard.org/board)
+-- Forum: C-House! (https://thecardboard.org/board/forumdisplay.php?fid=4)
+--- Forum: The CARDboard (https://thecardboard.org/board/forumdisplay.php?fid=5)
+--- Thread: ATTN: David Shaw; RE: Run-Run-Pass (/showthread.php?tid=17151)

Pages: 1 2


ATTN: David Shaw; RE: Run-Run-Pass - BostonCard - 01-11-2019

Someone send this memo to David Shaw.

https://fivethirtyeight.com/features/you-called-a-run-on-first-down-youre-already-screwed/

I'll be curious how often we tried the run-run-pass combo (my guess is a fair bit).

Quote:This result is the exact opposite of what we would expect to find if establishing the run via play sequences like rush-rush-pass were winning strategies. Instead of making a team less predictable, establishing the run on first and second down creates a game state that is often quite predictable for the defense. The opposing team is expecting a pass on third down because the first two plays were unsuccessful.

There are two major problems with this analysis, before we get too far into the triumpalism about how dumb Shaw's play-calling is.  The first, of course, is acknowledged in the piece, which is that in large part, any play sequence that ends in a run is likely to have been more successful than one that ends in a pass, because passes are called most frequently on third and long and runs most frequently on third and short.  Because the second and third down play-calling is in part conditional on the outcome of the first down, you really can't compare the sequences as a whole, unless the play-calling was scripted in advance.  Still, looking at first down plays, the expected points added from a first down pass (+.07) is better than the EPA from a run (-.08).

The second critique is more broad.  Play-calling is not completely independent, even between sets of downs.  That is, the idea of "establishing the run" is in part that running early benefits the offense later in the game.  That may not be true, but the analysis in fivethirtyeight ignores that possibility.  Teams may be willing to sacrifice a little bit of offensive productivity early in the game in order to "tire out the defense" or to make play action passing (and other attacks keyed off the run) more effective later in the game.

For the record, we passed more often than we ran in the first quarter and certainly in the first half of games last year.  However, even though I can rightfully be called a "Shawpologist", even I wished we would pass more on first and second run.


BC


RE: ATTN: David Shaw; RE: Run-Run-Pass - SamuelMcF - 01-11-2019

(01-11-2019, 03:41 PM)BostonCard Wrote:  I'll be curious how often we tried the run-run-pass combo (my guess is a fair bit).

I just went through the play-by-play of our first game vs. SDSU. Out of 13 1st-2nd-3rd down sequences, ZERO were run-run-pass. I repeat, ZERO. (There were 8 run-pass-pass sequences.)

EDIT: I went through the play-by-play on all games. See my post below.


RE: ATTN: David Shaw; RE: Run-Run-Pass - Goose - 01-11-2019

(01-11-2019, 03:56 PM)SamuelMcF Wrote:  
(01-11-2019, 03:41 PM)BostonCard Wrote:  I'll be curious how often we tried the run-run-pass combo (my guess is a fair bit).

I just went through the play-by-play of our first game vs. SDSU. Out of 13 1st-2nd-3rd down sequences, ZERO were run-run-pass. I repeat, ZERO. (There were 8 run-pass-pass sequences.)
That is an important statistic, but it was undoubtedly largely driven by us being second and nine more often than not. We HAD to throw on second down.


RE: ATTN: David Shaw; RE: Run-Run-Pass - d4cohn - 01-11-2019

(01-11-2019, 03:41 PM)BostonCard Wrote:  Someone send this memo to David Shaw.

https://fivethirtyeight.com/features/you-called-a-run-on-first-down-youre-already-screwed/

I'll be curious how often we tried the run-run-pass combo (my guess is a fair bit).

Quote:This result is the exact opposite of what we would expect to find if establishing the run via play sequences like rush-rush-pass were winning strategies. Instead of making a team less predictable, establishing the run on first and second down creates a game state that is often quite predictable for the defense. The opposing team is expecting a pass on third down because the first two plays were unsuccessful.

There are two major problems with this analysis, before we get too far into the triumpalism about how dumb Shaw's play-calling is.  The first, of course, is acknowledged in the piece, which is that in large part, any play sequence that ends in a run is likely to have been more successful than one that ends in a pass, because passes are called most frequently on third and long and runs most frequently on third and short.  Because the second and third down play-calling is in part conditional on the outcome of the first down, you really can't compare the sequences as a whole, unless the play-calling was scripted in advance.  Still, looking at first down plays, the expected points added from a first down pass (+.07) is better than the EPA from a run (-.08).

The second critique is more broad.  Play-calling is not completely independent, even between sets of downs.  That is, the idea of "establishing the run" is in part that running early benefits the offense later in the game.  That may not be true, but the analysis in fivethirtyeight ignores that possibility.  Teams may be willing to sacrifice a little bit of offensive productivity early in the game in order to "tire out the defense" or to make play action passing (and other attacks keyed off the run) more effective later in the game.

For the record, we passed more often than we ran in the first quarter and certainly in the first half of games last year.  However, even though I can rightfully be called a "Shawpologist", even I wished we would pass more on first and second run.


BC

Nice analysis by 538, but would love to see the analysis applied to college football before definitively accepting the translation of this principle to the college game.

Quite simply, the top rushing teams in college football (I am certainly not saying we were that this season, but we have been in the past) average more yards per carry than the top NFL teams (top in the NFL this year was the Carolina Panthers, headed by CMac and company, at 5.1 YPC; I was even surprised the top NFL team was this high. In contrast, Oklahoma averaged nearly 6.6 YPC). 

This very well may not change the overall conclusion, but perhaps the difference is less stark. I certainly know you are not a critic of Shaw's, but for the fans who are more critical of Shawvita, this is a red apple to say a green apple comparison between college football and the NFL.


RE: ATTN: David Shaw; RE: Run-Run-Pass - SamuelMcF - 01-11-2019

I went through the play-by-play of all games in the past season. Drumroll, please...

Out of 171 1st-2nd-3rd down sequences run by our offense, 17 were run-run-pass. 10%. 

Other notables: 
- Only 1 time total in the last 4 games (OSU, UCLA, once against Kal, Pitt).
- In addition to the 3 above games, SDSU and UC Davis were also healthy scratches.
- As you might expect, many of our run-run-pass sequences didn't pick up a first down.
- The 4 run-run-pass sequences to open a game (U$C, ND, Utah, ASU) all resulted in 3-and-outs. EDIT: Whoops, 1st drive vs. U$C was a TD. My bad. Other 3 are correct.


RE: ATTN: David Shaw; RE: Run-Run-Pass - slide - 01-11-2019

(01-11-2019, 03:56 PM)SamuelMcF Wrote:  Out of 171 1st-2nd-3rd down sequences run by our offense, 17 were run-run-pass. 10%.

of those 17, how many were R2P2 (run-run-pass-punt)?

thanks.


RE: ATTN: David Shaw; RE: Run-Run-Pass - SamuelMcF - 01-11-2019

(01-11-2019, 05:08 PM)slide Wrote:  
(01-11-2019, 03:56 PM)SamuelMcF Wrote:  Out of 171 1st-2nd-3rd down sequences run by our offense, 17 were run-run-pass. 10%.

of those 17, how many were R2P2 (run-run-pass-punt)?

thanks.

10; if you count run-run-pass-FG against Kal, it's 11. That's pretty damning, IMO.


RE: ATTN: David Shaw; RE: Run-Run-Pass - 81alum - 01-11-2019

This whole debate is ancient, although it is nice to see 538 discovering what Bill Walsh discovered in the 1970s.  Walsh emphatically thought you had to pass to set up the run--much more than the other way around.  This was true both at Stanford and later at the 49ers.  Pass blocking, he said, exhausts a defensive line more effectively than run blocking.  You pass first in the first half to get a lead, and with the defensive line worn down you run the ball late in the game to eat clock and seal the win.  Having been victimized by the pass for a half, the defense will respect it and be even more vulnerable to the run. 

I'm going to give Shaw credit for this season.  He adjusted to the personnel and the success we were having.  It certainly felt to me that as the season went along he became far less predictable and abandoned the run-first mentality that has so often typified him.


RE: ATTN: David Shaw; RE: Run-Run-Pass - Goose - 01-11-2019

(01-11-2019, 06:04 PM)81alum Wrote:  I'm going to give Shaw credit for this season.  He adjusted to the personnel and the success we were having.  It certainly felt to me that as the season went along he became far less predictable and abandoned the run-first mentality that has so often typified him.
I think most coaches that keep their jobs attempt to do so by featuring what their team does best. Sometimes, it takes a while to figure out what that is. For years, Shaw ran the ball and pounded into the line because it worked. When it stopped working, it took a bit of experimentation to determine we could pass lots better than we could run. Next year, I expect Shaw will try to return to type a bit. I hope we have the OL to do so.


RE: ATTN: David Shaw; RE: Run-Run-Pass - yvonne - 01-12-2019

fansplaining


RE: ATTN: David Shaw; RE: Run-Run-Pass - Hulk01 - 01-12-2019

Among all kinds of problems here: 

Teams that tend to run-run-pass are teams that also tend to (1) run well and (2) pass less well, even poorly. 

Play calling sequences do not exist independently of that consideration and many others including score and opponents.  Heavy run-run opening sequence are more common among teams sitting on a lead, for just one obvious example.

Finally, I've met three critics of Stanford play calling.  None played the game beyond junior high and none show more than a casual fan's understanding of football.  This seems an area where casual fans like them vastly overestimate their knowledge and insight.  This is compounded by the tendency of almost all of us, me included. to think we're pretty damned intelligent, and more than intelligent enough to understand a kid's game. 

It's vivid that this is not true.


RE: ATTN: David Shaw; RE: Run-Run-Pass - OutsiderFan - 01-12-2019

There is a big difference between putting 6 OL and a FB on the field and RRP, and using less predictable formations and personnel groupings to RRP.

I just watched some NFL Films match-up analysis of the Saints and Eagles on ESPN this morning.  One of the plays was a run to Saints RB Mark Ingram.  The play featured two TEs on the same side on the strong side, one down and one up.  It had Alvin Kamara lined up as a slot WR who went in motion with, a fake handoff to him at the snap.  This froze the defense and with proper blocking execution, allowed Ingram an outside path to gain the edge.

You can throw out of the same formation and motion, hand off to Kamara, or run the same play again. Or you could run a different play.  But I would submit not being predictable, based out of formation and personnel groupings makes RRP more effective.

It seem the choice preference for offensive coordinators is:
1. A run D knows is coming, but they can't stop
2. A run D knows is coming, it may or may not stop, but if they do, will yield large plays for overcommitting (with either run or pass)
3. Keeping D off balance, never knowing what to expect on a play to play basis

Stanford got away with being able to take the first approach for the most part for a long time.  But over the years defenses have made doing it difficult.  Shaw seemed to go to the second choice in 2017, where Love could make such big plays it was worth getting stuffed a few times to have the opportunity for big gains. Shaw did move a lot more to the third preference in 2018.

To me, the third option is where the most effective offenses operate.  You never know what is coming.  The offense may pass 10 times in a row, or it may run 10 times in a row. This said, I totally agree with the Walsh philosophy that it's better to pass to set up the run than run to set up the pass, all things being equal. Even if you can mow physically inferior teams down by running, I think in the longer term and to win championships against teams with more physical defenses, it pays off to be balanced in all games and use the pass to setup the run from the start.


RE: ATTN: David Shaw; RE: Run-Run-Pass - ChicagoCard - 01-12-2019

You lost me at "I just watched some NFL Films".


RE: ATTN: David Shaw; RE: Run-Run-Pass - martyup - 01-12-2019

I haven't rewatched our bowl game, but I recall Pitt used the RRP approach when they should have just RRRRRRRRRRRRRRRTD.


RE: ATTN: David Shaw; RE: Run-Run-Pass - Goose - 01-12-2019

(01-12-2019, 07:18 AM)OutsiderFan Wrote:  It seem the choice preference for offensive coordinators is:
1. A run D knows is coming, but they can't stop
2. A run D knows is coming, it may or may not stop, but if they do, will yield large plays for overcommitting (with either run or pass)
3. Keeping D off balance, never knowing what to expect on a play to play basis

Stanford got away with being able to take the first approach for the most part for a long time.  But over the years defenses have made doing it difficult.  Shaw seemed to go to the second choice in 2017, where Love could make such big plays it was worth getting stuffed a few times to have the opportunity for big gains. Shaw did move a lot more to the third preference in 2018.

To me, the third option is where the most effective offenses operate
I can't directly agree with the last statement. The most effective offenses use the first option. If you can reliably overpower your opponent, you do so. You can win a lot of games being one dimensional, and you can win it all that way if you are truly dominant. The problem is that creating this type of dominance is difficult. A few years ago, we could do so. Clearly, we can't reliably do so now, so the third option is the best choice if it is available.
The problem for any offense to make the third option work is that it must be truly "multiple". If you can't run effectively or you can't pass effectively, the defense will just load up against what you can do well. They will take their chances with what you don't do well, expecting that you can't beat them playing to your own weakness. The Sun Bowl was a lot of that. We loaded up against the run. Whether we could have stopped the run even then was another question, but Pitt chose option three and tried to "surprise" us with passes. They couldn't do that well, and so being "multiple" actually hurt them.
In 2018, we almost couldn't run the ball, so we were "multiple" only in what pass routes we could run. That won't get it done. Option three requires that we can run the ball well enough to win.  Against good teams, we couldn't mount an offense that was effective on the ground. That is the difference between 8-4 and 10-2, I think. If we can run the ball well enough that RRR is a viable way to get a first down some of the time, we can be effective with the third option. If not, 8-4 is about what we will get, best case. (I am not counting Bowl Games because they are "outliers" these days. With so many opting out, so many "dud" games,  and with coaches playing for "next year", who wins is often not indicative of anything.)


RE: ATTN: David Shaw; RE: Run-Run-Pass - BostonCard - 01-12-2019

That’s great in theory, but a competent, disciplined defense will not get caught by surprise, because they are  taught not to try to guess the play, and to be prepared to defend whatever the offense throws.  It is only if you have an undisciplined defense, or if the defense is pressing because they are being beaten at the point of attack that they are most vulnerable to surprise.

Our most successive offensive games have been through #1 and #2.  The best example of #1 was the beat down against USC, where the Lb was quoted as saying “If you run power one more time I’m taking myself out”.  Two prominent examples of #2 were the orange bowl game against VA Tech and the success we had with play action over the top, and the Rose Bowl against Iowa, where once we had the Iowa defense on its heals, we brought out the fake fumble play.  But you can’t do that if you don’t have McCaffrey blowing through the line of scrimmage and getting the Iowa linebackers and safeties overcommitting to the run, leaving Michael rector all alone in the end zone.

What you often ascribe to #3 “keeping the d off balance” is just the offense dominating the defense.  The defense seems off balance because it is just being beaten, and changing the play-calling wouldn’t alter things.

BC


RE: ATTN: David Shaw; RE: Run-Run-Pass - Hulk01 - 01-13-2019

I feel like screaming this.

From a defense's point of view, there is no such thing as predictably advantageous play calling. 

As others point out, a defense must as well prepared to defend the less likely play than the more probable one.

RRP, PRR, RRR, PPR--it does not matter.

This argument, endlessly repeated by Stanford fans, drives me nuts.


RE: ATTN: David Shaw; RE: Run-Run-Pass - 81alum - 01-13-2019

(01-13-2019, 09:12 AM)Hulk01 Wrote:  I feel like screaming this.

From a defense's point of view, there is no such thing as predictably advantageous play calling. 

As others point out, a defense must as well prepared to defend the less likely play than the more probable one.

RRP, PRR, RRR, PPR--it does not matter.

This argument, endlessly repeated by Stanford fans, drives me nuts.
But defensive coordinators put in different personnel packages and call different defensive alignments depending on what they expect the offense is most likely to do.  Sure, the players all have to be prepared for anything regardless of package, but stopping the run when you are in dime or stopping the pass if you are in short yardage puts you at a disadvantage.  If a team always runs on first down and always passes on 3rd down, then a DC can always put 9 in the box on first down and nickle or dime it on third.  And that makes a difference.  Perhaps I don't follow your meaning?


RE: ATTN: David Shaw; RE: Run-Run-Pass - Hulk01 - 01-13-2019

So what if we run 63% of the time on first down?

What difference does it make to a defense? 

None.

What if often go R, R, P?

It makes no difference.

If we always ran on first down--always--yes, that would be helpful. 
We don't and never have.

People, this is the most wasted use of Stanford intelligence of which I am aware. 

PS: True story. 
The last person who criticized Stanford's play calling to me--honest, I swear--didn't know where linebackers lined up.


RE: ATTN: David Shaw; RE: Run-Run-Pass - OutsiderFan - 01-13-2019

I think the greater reality is if you are in position to RRP too frequently, you aren't gaining enough yards on 1st and 2nd down, and you've got more problems than just whether you run or pass too frequently. Because this demonstrates you can't run the ball, which is a much larger issue than running too frequently.