[washingtonismoney]

It is not possible in any system I am aware of to audible into any play in the playbook.

Air Raid offenses frequently do this. And reputedly Luck could audible into any play he wanted.

[washingtonismoney]

It is not possible in any system I am aware of to audible into any play in the playbook.

Air Raid offenses frequently do this. And reputedly Luck could audible into any play he wanted.

[fullmetal]

It would have been great to have called a time out there, as Shaw was apparently trying to do the same but failed.

[fullmetal]

It would have been great to have called a time out there, as Shaw was apparently trying to do the same but failed.

[Goose]

[quote author=Goose link=topic=16412.msg185997#msg185997 date=1484162128]
It is not possible in any system I am aware of to audible into any play in the playbook.

Air Raid offenses frequently do this. And reputedly Luck could audible into any play he wanted.
[/quote]
ANY play? It is hard to audible a pass to a tight end when you don't have one on the field, so I assume you don't mean that. Do you know how the air raid people designate the plays in a way that is understandable to them but not to the defense? The systems I am familiar with allow checking into a subset of different plays, but not the whole playbook. Changing the whole formation is possible, but not from anything to anything.  These systems also allow adjustments to called plays and announce the quarterbacks read of the defense (which may produce pre-formulated play changes), but once again, the set is limited.

[Goose]

[quote author=Goose link=topic=16412.msg185997#msg185997 date=1484162128]
It is not possible in any system I am aware of to audible into any play in the playbook.

Air Raid offenses frequently do this. And reputedly Luck could audible into any play he wanted.
[/quote]
ANY play? It is hard to audible a pass to a tight end when you don't have one on the field, so I assume you don't mean that. Do you know how the air raid people designate the plays in a way that is understandable to them but not to the defense? The systems I am familiar with allow checking into a subset of different plays, but not the whole playbook. Changing the whole formation is possible, but not from anything to anything.  These systems also allow adjustments to called plays and announce the quarterbacks read of the defense (which may produce pre-formulated play changes), but once again, the set is limited.

[BostonCard]

[quote author=Goose link=topic=16412.msg185997#msg185997 date=1484162128]
It is not possible in any system I am aware of to audible into any play in the playbook.

Air Raid offenses frequently do this. And reputedly Luck could audible into any play he wanted.
[/quote]

I thought Luck could call any play he wanted (at least, any play with the personnel in the huddle), but not necessarily that he could audible any play.  Seems like calling a whole new protection and route scheme on the line might be difficult to execute.  I believe, however, that Luck could (and did) get a chance to call a set of plays in the huddle (I believe three) and pick one based on the defensive alignment at the line.

BC

[BostonCard]

[quote author=Goose link=topic=16412.msg185997#msg185997 date=1484162128]
It is not possible in any system I am aware of to audible into any play in the playbook.

Air Raid offenses frequently do this. And reputedly Luck could audible into any play he wanted.
[/quote]

I thought Luck could call any play he wanted (at least, any play with the personnel in the huddle), but not necessarily that he could audible any play.  Seems like calling a whole new protection and route scheme on the line might be difficult to execute.  I believe, however, that Luck could (and did) get a chance to call a set of plays in the huddle (I believe three) and pick one based on the defensive alignment at the line.

BC

[Goose]

I thought Luck could call any play he wanted (at least, any play with the personnel in the huddle), but not necessarily that he could audible any play.  Seems like calling a whole new protection and route scheme on the line might be difficult to execute.  I believe, however, that Luck could (and did) get a chance to call a set of plays in the huddle (I believe three) and pick one based on the defensive alignment at the line.

BC

That also matches my understanding, but I thought the number was two. Three would certainly be manageable.
Most teams I know of have a certain subset of the playbook that is "permanently installed". It is expected that for any game, the quarterback can call these plays and expect them to be correctly executed. Then, for each particular game, plays are "installed" for that opponent. These plays are practiced during the week, and the quarterback can expect the players to run them well. Now, in principle, Andrew Luck could call any play in the playbook, including those that were not "installed". But I expect he seldom did. The coaches have put together what they think will work. If you need something else, the defense is doing something unexpected. While the team should be familiar with the entire playbook, (and the method Stanford uses is positional and constructive, allowing plays to be made from "pieces"), if you haven't practiced the play that week, or maybe for the previous opponent, it may not go so well.
Additionally, the quarterback can call "several" plays in the huddle and select among them at the line of scrimmage. He can also "automatic" to a subset of the permanently installed plays (a quarterback sneak being a prominent example). Some formations also have associated "automatics", like in Ogre sets you can call a pass play at the LOS to any one of four possible receivers. That sequence is not active in other formations, or means something different. So it can be real complex, or fairly simple, depending on what the coach allows and the quarterback can remember!
The one thing the QB has to worry about is that if he audibles to a pass play and the freshman LT doesn't understand the call.

[Goose]

I thought Luck could call any play he wanted (at least, any play with the personnel in the huddle), but not necessarily that he could audible any play.  Seems like calling a whole new protection and route scheme on the line might be difficult to execute.  I believe, however, that Luck could (and did) get a chance to call a set of plays in the huddle (I believe three) and pick one based on the defensive alignment at the line.

BC

That also matches my understanding, but I thought the number was two. Three would certainly be manageable.
Most teams I know of have a certain subset of the playbook that is "permanently installed". It is expected that for any game, the quarterback can call these plays and expect them to be correctly executed. Then, for each particular game, plays are "installed" for that opponent. These plays are practiced during the week, and the quarterback can expect the players to run them well. Now, in principle, Andrew Luck could call any play in the playbook, including those that were not "installed". But I expect he seldom did. The coaches have put together what they think will work. If you need something else, the defense is doing something unexpected. While the team should be familiar with the entire playbook, (and the method Stanford uses is positional and constructive, allowing plays to be made from "pieces"), if you haven't practiced the play that week, or maybe for the previous opponent, it may not go so well.
Additionally, the quarterback can call "several" plays in the huddle and select among them at the line of scrimmage. He can also "automatic" to a subset of the permanently installed plays (a quarterback sneak being a prominent example). Some formations also have associated "automatics", like in Ogre sets you can call a pass play at the LOS to any one of four possible receivers. That sequence is not active in other formations, or means something different. So it can be real complex, or fairly simple, depending on what the coach allows and the quarterback can remember!
The one thing the QB has to worry about is that if he audibles to a pass play and the freshman LT doesn't understand the call.

[washingtonismoney]

ANY play?

From Michael Lewis's profile of Leach:

Leach is unusual in giving his quarterback the authority to change every play, wherever the line of scrimmage. "He can see more than I'll ever see," Leach says. "If I call a stupid play, his job is to get me out of it. If he doesn't get me out of it, I might holler at him. But if you let him react to what he sees, there's a ton of touchdowns to be had." All Leach is really saying to Hodges when he sends in the play is, "Line up in Ace, see how they line up against it and call a good play."

http://www.nytimes.com/2005/12/04/magazi...-deep.html

From ESPN in 2016:

It doesn't happen often, but Leach says there are games when his quarterback heads onto the field about 60 percent of the time with nothing more than an offensive formation. In those cases, it is the quarterback's responsibility alone to give the offense his own play call, unless he decides to audible on the formation altogether.

It's a level of trust between Leach and his quarterbacks that he calls “one of the strengths of our offense.” According to several former quarterbacks who played for Leach, it also serves to form what they describe as the most unique relationship between a coach and a player in all of college football.

“You don’t find that in a lot of offensive coordinators or head coaches,” former Texas Tech quarterback B.J. Symons said. “Some of them might be a little egotistical that you’re going to run what they call. … That freedom that Mike gave you -- and it came from him trusting that you could make the right call -- that was a big part of my success.”

http://www.espn.com/blog/pac12/post/_/id...lse-unique

[washingtonismoney]

ANY play?

From Michael Lewis's profile of Leach:

Leach is unusual in giving his quarterback the authority to change every play, wherever the line of scrimmage. "He can see more than I'll ever see," Leach says. "If I call a stupid play, his job is to get me out of it. If he doesn't get me out of it, I might holler at him. But if you let him react to what he sees, there's a ton of touchdowns to be had." All Leach is really saying to Hodges when he sends in the play is, "Line up in Ace, see how they line up against it and call a good play."

http://www.nytimes.com/2005/12/04/magazi...-deep.html

From ESPN in 2016:

It doesn't happen often, but Leach says there are games when his quarterback heads onto the field about 60 percent of the time with nothing more than an offensive formation. In those cases, it is the quarterback's responsibility alone to give the offense his own play call, unless he decides to audible on the formation altogether.

It's a level of trust between Leach and his quarterbacks that he calls “one of the strengths of our offense.” According to several former quarterbacks who played for Leach, it also serves to form what they describe as the most unique relationship between a coach and a player in all of college football.

“You don’t find that in a lot of offensive coordinators or head coaches,” former Texas Tech quarterback B.J. Symons said. “Some of them might be a little egotistical that you’re going to run what they call. … That freedom that Mike gave you -- and it came from him trusting that you could make the right call -- that was a big part of my success.”

http://www.espn.com/blog/pac12/post/_/id...lse-unique

[BostonCard]

By the way, Luke Faulk announced he is returning to WSU.

BC

[BostonCard]

By the way, Luke Faulk announced he is returning to WSU.

BC

[JJJ]

[quote author=Goose link=topic=16412.msg186010#msg186010 date=1484164930]
ANY play?

From Michael Lewis's profile of Leach:

Leach is unusual in giving his quarterback the authority to change every play, wherever the line of scrimmage. "He can see more than I'll ever see," Leach says. "If I call a stupid play, his job is to get me out of it. If he doesn't get me out of it, I might holler at him. But if you let him react to what he sees, there's a ton of touchdowns to be had." All Leach is really saying to Hodges when he sends in the play is, "Line up in Ace, see how they line up against it and call a good play."

http://www.nytimes.com/2005/12/04/magazi...-deep.html

From ESPN in 2016:

It doesn't happen often, but Leach says there are games when his quarterback heads onto the field about 60 percent of the time with nothing more than an offensive formation. In those cases, it is the quarterback's responsibility alone to give the offense his own play call, unless he decides to audible on the formation altogether.

It's a level of trust between Leach and his quarterbacks that he calls “one of the strengths of our offense.” According to several former quarterbacks who played for Leach, it also serves to form what they describe as the most unique relationship between a coach and a player in all of college football.

“You don’t find that in a lot of offensive coordinators or head coaches,” former Texas Tech quarterback B.J. Symons said. “Some of them might be a little egotistical that you’re going to run what they call. … That freedom that Mike gave you -- and it came from him trusting that you could make the right call -- that was a big part of my success.”

http://www.espn.com/blog/pac12/post/_/id...lse-unique
[/quote]

Thanks for sharing this. I love pretty much anything Michael Lewis writes and hadn't seen this yet. My one new thing learned for the day.  8)

[JJJ]

[quote author=Goose link=topic=16412.msg186010#msg186010 date=1484164930]
ANY play?

From Michael Lewis's profile of Leach:

Leach is unusual in giving his quarterback the authority to change every play, wherever the line of scrimmage. "He can see more than I'll ever see," Leach says. "If I call a stupid play, his job is to get me out of it. If he doesn't get me out of it, I might holler at him. But if you let him react to what he sees, there's a ton of touchdowns to be had." All Leach is really saying to Hodges when he sends in the play is, "Line up in Ace, see how they line up against it and call a good play."

http://www.nytimes.com/2005/12/04/magazi...-deep.html

From ESPN in 2016:

It doesn't happen often, but Leach says there are games when his quarterback heads onto the field about 60 percent of the time with nothing more than an offensive formation. In those cases, it is the quarterback's responsibility alone to give the offense his own play call, unless he decides to audible on the formation altogether.

It's a level of trust between Leach and his quarterbacks that he calls “one of the strengths of our offense.” According to several former quarterbacks who played for Leach, it also serves to form what they describe as the most unique relationship between a coach and a player in all of college football.

“You don’t find that in a lot of offensive coordinators or head coaches,” former Texas Tech quarterback B.J. Symons said. “Some of them might be a little egotistical that you’re going to run what they call. … That freedom that Mike gave you -- and it came from him trusting that you could make the right call -- that was a big part of my success.”

http://www.espn.com/blog/pac12/post/_/id...lse-unique
[/quote]

Thanks for sharing this. I love pretty much anything Michael Lewis writes and hadn't seen this yet. My one new thing learned for the day.  8)

[BostonCard]

Upon further review, beyond the two or three plays to the huddle and audibling at the line of scrimmage, Andrew Luck did call his own plays:

http://www.espn.com/blog/stanford-footba...ay-calling

Following Stanford's win over UCLA, Shaw said Luck had the entire playbook available to him. Of course, the personnel on the field certainly dictates what sorts of plays he can call.

Having tight ends who can moonlight as wide receiers and/or fullbacks helps.

"When you have those guys on the field, you can run multiple looks and pretty much get to anything in the playbook," Luck said.

Looks like against UCLA that year, we went no huddle, Luck would look at the defensive alignment and call the play.

BC

[BostonCard]

Upon further review, beyond the two or three plays to the huddle and audibling at the line of scrimmage, Andrew Luck did call his own plays:

http://www.espn.com/blog/stanford-footba...ay-calling

Following Stanford's win over UCLA, Shaw said Luck had the entire playbook available to him. Of course, the personnel on the field certainly dictates what sorts of plays he can call.

Having tight ends who can moonlight as wide receiers and/or fullbacks helps.

"When you have those guys on the field, you can run multiple looks and pretty much get to anything in the playbook," Luck said.

Looks like against UCLA that year, we went no huddle, Luck would look at the defensive alignment and call the play.

BC

[teejers1]

BC, there is no hindsight bias going on here on those atrocious calls.  In both the Utah and SC games, I was yelling run the damn ball at the TV screen real time.  And in both instances, so was every other Stanford fan watching the game (and those that might not have been saying that, or thinking that, weren't paying attention).  Sorry, but that's just the truth.

And your antiseptic season stats about how "Team X plays against the run on third and short" is of very little value.  I'd put much more reliance on what is going on in the game, and particularly at that time in the game.

Honestly, the unending efforts by some to defend Shaw at all costs, no matter the lunacy of a particular play call, is amusing.  You know, it's possible to think Shaw is a great coach for Stanford and still believe that he has cluster-farked his share of plays/games over the years.

But I guess that concept is just too difficult to embrace in these parts.

[teejers1]

BC, there is no hindsight bias going on here on those atrocious calls.  In both the Utah and SC games, I was yelling run the damn ball at the TV screen real time.  And in both instances, so was every other Stanford fan watching the game (and those that might not have been saying that, or thinking that, weren't paying attention).  Sorry, but that's just the truth.

And your antiseptic season stats about how "Team X plays against the run on third and short" is of very little value.  I'd put much more reliance on what is going on in the game, and particularly at that time in the game.

Honestly, the unending efforts by some to defend Shaw at all costs, no matter the lunacy of a particular play call, is amusing.  You know, it's possible to think Shaw is a great coach for Stanford and still believe that he has cluster-farked his share of plays/games over the years.

But I guess that concept is just too difficult to embrace in these parts.

[BostonCard]

Yes, there is hindsight bias because unless you are in the wrong profession, you almost certainly had plays that you thought were headscratchers a priori, that wound up turning out ok, or that went for big gains.  In order to properly evaluate this, you would first have to have a consensus of what the headscratcher plays were, and my guess is that if you asked 10 people to identify "headscratchers" (in real time) you would wind up with very little overlap.  For example, as a general principle (regardless of my Shaw defensiveness), I tend to really like play action passes on second and short for reasons I've stated above.

Secondly, even if you could identify a priori a set of consensus head-scratchers, you would have to evaluate, if, in aggregate, those headscratchers produced worse outcomes than regular plays.

And even then, you would have to somehow figure out how going against type (for example, passing on a second or third and short, rushing on third and long, etc.) affects other plays.  A third and short pass might wind up incomplete, but might get the defense to pull a safety out of the box on a subsequent drive that allows you to convert it rushing the ball.

The bottom line is not that I think Shaw is infallible.  I think, for example, some of his punting decisions can rightly be called into question, with a reasonable amount of math to support it (even though in a few of those we downed the punt inside the five, and in one case we wound up scoring a safety on the opponent's ensuing drive directly attributable to the IMO bad decision to punt).  Of course Shaw has made boneheaded playcalls, as did Harbaugh, as does Nick Saban and Bill Belicheck, but I would be very hard pressed to identify them, in real time.  I also don't think Shaw needs defending; he's a big boy getting paid lots of money and being attacked on message boards goes with the territory. 

I do find criticism of individual play calls somewhat annoying because they are so prone to cognitive biases, and I would even say that of someone like Buddy Teevens, who clearly, in aggregate, called some atrocious offenses (though even then, I wonder if he would have been a hell of a lot more successful with the dreaded shotgun draw if he had the TWU clearing space for Christian McCaffrey or Toby Gerhart running with the ball).  It's not that I don't think Shaw makes mistakes; it's just that I don't think anybody can reliably identify them in real time.

BC

[BostonCard]

Yes, there is hindsight bias because unless you are in the wrong profession, you almost certainly had plays that you thought were headscratchers a priori, that wound up turning out ok, or that went for big gains.  In order to properly evaluate this, you would first have to have a consensus of what the headscratcher plays were, and my guess is that if you asked 10 people to identify "headscratchers" (in real time) you would wind up with very little overlap.  For example, as a general principle (regardless of my Shaw defensiveness), I tend to really like play action passes on second and short for reasons I've stated above.

Secondly, even if you could identify a priori a set of consensus head-scratchers, you would have to evaluate, if, in aggregate, those headscratchers produced worse outcomes than regular plays.

And even then, you would have to somehow figure out how going against type (for example, passing on a second or third and short, rushing on third and long, etc.) affects other plays.  A third and short pass might wind up incomplete, but might get the defense to pull a safety out of the box on a subsequent drive that allows you to convert it rushing the ball.

The bottom line is not that I think Shaw is infallible.  I think, for example, some of his punting decisions can rightly be called into question, with a reasonable amount of math to support it (even though in a few of those we downed the punt inside the five, and in one case we wound up scoring a safety on the opponent's ensuing drive directly attributable to the IMO bad decision to punt).  Of course Shaw has made boneheaded playcalls, as did Harbaugh, as does Nick Saban and Bill Belicheck, but I would be very hard pressed to identify them, in real time.  I also don't think Shaw needs defending; he's a big boy getting paid lots of money and being attacked on message boards goes with the territory. 

I do find criticism of individual play calls somewhat annoying because they are so prone to cognitive biases, and I would even say that of someone like Buddy Teevens, who clearly, in aggregate, called some atrocious offenses (though even then, I wonder if he would have been a hell of a lot more successful with the dreaded shotgun draw if he had the TWU clearing space for Christian McCaffrey or Toby Gerhart running with the ball).  It's not that I don't think Shaw makes mistakes; it's just that I don't think anybody can reliably identify them in real time.

BC

[washingtonismoney]

The other funny thing is, let's stipulate some Shaw playcalls are Bad. How many of them are they? Given the harping over the Utah playcalls, which was, jeez, three years ago now, it seems Bad playcalls are pretty rare. Once or twice a game, we'll say. Is that really that big a deal?

That being said, I'm pretty skeptical of the notion we can identify good playcallers reliably anyway. I think Lane Kiffin has metamorphosed three or five times between Bad Playcaller and Good Playcaller, like the world's most indecisive butterfly/caterpillar.

[washingtonismoney]

The other funny thing is, let's stipulate some Shaw playcalls are Bad. How many of them are they? Given the harping over the Utah playcalls, which was, jeez, three years ago now, it seems Bad playcalls are pretty rare. Once or twice a game, we'll say. Is that really that big a deal?

That being said, I'm pretty skeptical of the notion we can identify good playcallers reliably anyway. I think Lane Kiffin has metamorphosed three or five times between Bad Playcaller and Good Playcaller, like the world's most indecisive butterfly/caterpillar.

[BostonCard]

The other funny thing is, let's stipulate some Shaw playcalls are Bad. How many of them are they? Given the harping over the Utah playcalls, which was, jeez, three years ago now, it seems Bad playcalls are pretty rare. Once or twice a game, we'll say. Is that really that big a deal?

That being said, I'm pretty skeptical of the notion we can identify good playcallers reliably anyway. I think Lane Kiffin has metamorphosed three or five times between Bad Playcaller and Good Playcaller, like the world's most indecisive butterfly/caterpillar.

As I pointed out, if Buddy Teevens had Christian McCaffrey running behind Murphy, Garnett, and company, his shotgun draws would look like genius plays.

Someday, someone is going to turn towards AI/neural networks to use the all-22 views to determine the optimal playcalls in every situation.  But until then, I agree that whether a play is good or bad play seems to be more susceptible to an individual's own biases.  We used to hate the fade, but now we seem to rarely complain when it is thrown.  Likewise, the Wildcat has been the object of much derision, except if I recall, someone went through and showed that we gained more yards out of the wildcat than on regular running plays despite the complete lack of deception (and, of course, every once in a while, you get a big completion out of it).

BC

[BostonCard]

The other funny thing is, let's stipulate some Shaw playcalls are Bad. How many of them are they? Given the harping over the Utah playcalls, which was, jeez, three years ago now, it seems Bad playcalls are pretty rare. Once or twice a game, we'll say. Is that really that big a deal?

That being said, I'm pretty skeptical of the notion we can identify good playcallers reliably anyway. I think Lane Kiffin has metamorphosed three or five times between Bad Playcaller and Good Playcaller, like the world's most indecisive butterfly/caterpillar.

As I pointed out, if Buddy Teevens had Christian McCaffrey running behind Murphy, Garnett, and company, his shotgun draws would look like genius plays.

Someday, someone is going to turn towards AI/neural networks to use the all-22 views to determine the optimal playcalls in every situation.  But until then, I agree that whether a play is good or bad play seems to be more susceptible to an individual's own biases.  We used to hate the fade, but now we seem to rarely complain when it is thrown.  Likewise, the Wildcat has been the object of much derision, except if I recall, someone went through and showed that we gained more yards out of the wildcat than on regular running plays despite the complete lack of deception (and, of course, every once in a while, you get a big completion out of it).

BC

[gocard14]

[quote author=washingtonismoney link=topic=16412.msg186087#msg186087 date=1484198953]
The other funny thing is, let's stipulate some Shaw playcalls are Bad. How many of them are they? Given the harping over the Utah playcalls, which was, jeez, three years ago now, it seems Bad playcalls are pretty rare. Once or twice a game, we'll say. Is that really that big a deal?

That being said, I'm pretty skeptical of the notion we can identify good playcallers reliably anyway. I think Lane Kiffin has metamorphosed three or five times between Bad Playcaller and Good Playcaller, like the world's most indecisive butterfly/caterpillar.

As I pointed out, if Buddy Teevens had Christian McCaffrey running behind Murphy, Garnett, and company, his shotgun draws would look like genius plays.

Someday, someone is going to turn towards AI/neural networks to use the all-22 views to determine the optimal playcalls in every situation.  But until then, I agree that whether a play is good or bad play seems to be more susceptible to an individual's own biases.  We used to hate the fade, but now we seem to rarely complain when it is thrown.  Likewise, the Wildcat has been the object of much derision, except if I recall, someone went through and showed that we gained more yards out of the wildcat than on regular running plays despite the complete lack of deception (and, of course, every once in a while, you get a big completion out of it).

BC
[/quote]

A play calling AI would be really interesting to see, as so much of play calling success depends not just on being in the correct play for a given defense, but a measure of unpredictability. An AI that would always counter the defensive alignment with the correct play could be extremely predictable, and as such would require a degree of randomness incorporated into the programming. Because of this, it would sometimes call the "wrong" play, as an unexpected draw or screen on 3rd and 6 sometimes looks like a great call and sometimes blows up in your face.

[gocard14]

[quote author=washingtonismoney link=topic=16412.msg186087#msg186087 date=1484198953]
The other funny thing is, let's stipulate some Shaw playcalls are Bad. How many of them are they? Given the harping over the Utah playcalls, which was, jeez, three years ago now, it seems Bad playcalls are pretty rare. Once or twice a game, we'll say. Is that really that big a deal?

That being said, I'm pretty skeptical of the notion we can identify good playcallers reliably anyway. I think Lane Kiffin has metamorphosed three or five times between Bad Playcaller and Good Playcaller, like the world's most indecisive butterfly/caterpillar.

As I pointed out, if Buddy Teevens had Christian McCaffrey running behind Murphy, Garnett, and company, his shotgun draws would look like genius plays.

Someday, someone is going to turn towards AI/neural networks to use the all-22 views to determine the optimal playcalls in every situation.  But until then, I agree that whether a play is good or bad play seems to be more susceptible to an individual's own biases.  We used to hate the fade, but now we seem to rarely complain when it is thrown.  Likewise, the Wildcat has been the object of much derision, except if I recall, someone went through and showed that we gained more yards out of the wildcat than on regular running plays despite the complete lack of deception (and, of course, every once in a while, you get a big completion out of it).

BC
[/quote]

A play calling AI would be really interesting to see, as so much of play calling success depends not just on being in the correct play for a given defense, but a measure of unpredictability. An AI that would always counter the defensive alignment with the correct play could be extremely predictable, and as such would require a degree of randomness incorporated into the programming. Because of this, it would sometimes call the "wrong" play, as an unexpected draw or screen on 3rd and 6 sometimes looks like a great call and sometimes blows up in your face.

[Goose]

A play calling AI would be really interesting to see, as so much of play calling success depends not just on being in the correct play for a given defense, but a measure of unpredictability. An AI that would always counter the defensive alignment with the correct play could be extremely predictable, and as such would require a degree of randomness incorporated into the programming. Because of this, it would sometimes call the "wrong" play, as an unexpected draw or screen on 3rd and 6 sometimes looks like a great call and sometimes blows up in your face.

Yes, the idea that there is an "optimal" call on any given play is clearly false. First of all, optimal in what sense? Optimal in that we get the most yards on this play or optimal in that we win the game? As others have pointed out, sometimes getting the most yards on a given play (a "greedy" algorithm) dose not necessarily result in a global optimum (winning the game). But who is ever going to admit/claim that they called a play that didn't work to set up another play later? There is also the bugbear of decision theory: "good decision, bad outcome". There is a non-zero probability that even the "best" possible call gets stuffed. There is also the opposite, where the play shouldn't work but does (because of incredible execution by the offense or poor execution by the defense, or dumb luck). So football games have no analytic "best" solution, at least until now. If they ever do, I don't expect we will pay much attention to the game any more.

[Goose]

A play calling AI would be really interesting to see, as so much of play calling success depends not just on being in the correct play for a given defense, but a measure of unpredictability. An AI that would always counter the defensive alignment with the correct play could be extremely predictable, and as such would require a degree of randomness incorporated into the programming. Because of this, it would sometimes call the "wrong" play, as an unexpected draw or screen on 3rd and 6 sometimes looks like a great call and sometimes blows up in your face.

Yes, the idea that there is an "optimal" call on any given play is clearly false. First of all, optimal in what sense? Optimal in that we get the most yards on this play or optimal in that we win the game? As others have pointed out, sometimes getting the most yards on a given play (a "greedy" algorithm) dose not necessarily result in a global optimum (winning the game). But who is ever going to admit/claim that they called a play that didn't work to set up another play later? There is also the bugbear of decision theory: "good decision, bad outcome". There is a non-zero probability that even the "best" possible call gets stuffed. There is also the opposite, where the play shouldn't work but does (because of incredible execution by the offense or poor execution by the defense, or dumb luck). So football games have no analytic "best" solution, at least until now. If they ever do, I don't expect we will pay much attention to the game any more.

[stupac2]

We used to hate the fade, but now we seem to rarely complain when it is thrown.

Fades are weird because they're something that the casual fan can easily identify, and (barring unusual combinations of QB and receiver) at best 50-50 plays, so with a dash of confirmation bias (which I think explains 99% of teejer's problems with playcalling) you can easily wind up on "I LOVE FADES" or "FADES ARE THE WORST". Personally, I don't have a problem with them in general, except in do-or-die situations, and that's precisely because they're 50% plays. First and goal? Great time for a fade! Fourth and goal? BOOO FADE. But, as has been mentioned, sometimes you have to go against tendencies. If you never, ever throw a fade on 3rd/4th down, then in the highest-stakes environment that's probably the best time to do it, since the defense won't be expecting it, and an uncontested fade is wade better than 50-50.

Also, even very high-level players are prone to this. Sherman got all up in Carroll's face because they threw from the 1 earlier this year, despite the fact that in the past few years the Super Bowl play is either the only interception on a throw from the 1, or one of very few. The success rate is higher than runs! But because they lost the Super Bowl on a throw from the 1, he doesn't ever want to do it. Not rational!

[stupac2]

We used to hate the fade, but now we seem to rarely complain when it is thrown.

Fades are weird because they're something that the casual fan can easily identify, and (barring unusual combinations of QB and receiver) at best 50-50 plays, so with a dash of confirmation bias (which I think explains 99% of teejer's problems with playcalling) you can easily wind up on "I LOVE FADES" or "FADES ARE THE WORST". Personally, I don't have a problem with them in general, except in do-or-die situations, and that's precisely because they're 50% plays. First and goal? Great time for a fade! Fourth and goal? BOOO FADE. But, as has been mentioned, sometimes you have to go against tendencies. If you never, ever throw a fade on 3rd/4th down, then in the highest-stakes environment that's probably the best time to do it, since the defense won't be expecting it, and an uncontested fade is wade better than 50-50.

Also, even very high-level players are prone to this. Sherman got all up in Carroll's face because they threw from the 1 earlier this year, despite the fact that in the past few years the Super Bowl play is either the only interception on a throw from the 1, or one of very few. The success rate is higher than runs! But because they lost the Super Bowl on a throw from the 1, he doesn't ever want to do it. Not rational!

[teejers1]

You guys crack me up.  In what world is passing on third and 2 knocking on Utah's end zone late a good call?  In the world where it might have been completed for a TD?  Gee, that's real analytical thinking.  A triple reverse from the 2 also might score a TD . . . but it doesn't make it a good call.

In what world is shifting to a shotgun on third and six inches in the Sun Bowl and then pitching it to the short side a good call, especially for a "power running" team?

In what world is calling a pass play on second and 2 from your own 35 against SC late in a tie game - when you KNOW you will punt the ball away if you don't get the first down - a good call, when you can run it twice and virtually guarantee a first down?  [Running also has the benefit of burning clock in the extremely unlikely even you don't get the first down after two rushes].

In what world is having Andrew Luck stop his march to the end zone in the Fiesta Bowl so you can have your "just getting back from injury" PK, who had already missed one FG, try a 37 yarder at the buzzer for the W? [Yeah, yeah, I know . . . in the world where he makes the FG].

And to the poster who said:  you're only complaining about a few games over the years.  That is right.  And I suppose that is good, better than complaining about tens of games (although I submit there were many other wrong calls that we've forgotten about because Stanford managed to win the game despite the play calling).

What it comes down to is are you giving the team the best chance to win?
And those plays are magnified in crunch-time, with the game clearly on the line.

This regime has had too many clusterfarks that cost games.  I agree that most coaches make mistakes like that (Harbaugh made at least 2 memorable ones - one didn't cost the game against ND (not taking a knee or two); the other very well might have (having Luck throw on first down late against Cal in BG after a long catch and run by Toby).

Coach Shaw has been great for Stanford, which has had a remarkable W/L record under his leadership.  But it could have been even better, in BHA ways (to borrow from a Tedism), with just a little better sense of game situations by the play callers.

[teejers1]

You guys crack me up.  In what world is passing on third and 2 knocking on Utah's end zone late a good call?  In the world where it might have been completed for a TD?  Gee, that's real analytical thinking.  A triple reverse from the 2 also might score a TD . . . but it doesn't make it a good call.

In what world is shifting to a shotgun on third and six inches in the Sun Bowl and then pitching it to the short side a good call, especially for a "power running" team?

In what world is calling a pass play on second and 2 from your own 35 against SC late in a tie game - when you KNOW you will punt the ball away if you don't get the first down - a good call, when you can run it twice and virtually guarantee a first down?  [Running also has the benefit of burning clock in the extremely unlikely even you don't get the first down after two rushes].

In what world is having Andrew Luck stop his march to the end zone in the Fiesta Bowl so you can have your "just getting back from injury" PK, who had already missed one FG, try a 37 yarder at the buzzer for the W? [Yeah, yeah, I know . . . in the world where he makes the FG].

And to the poster who said:  you're only complaining about a few games over the years.  That is right.  And I suppose that is good, better than complaining about tens of games (although I submit there were many other wrong calls that we've forgotten about because Stanford managed to win the game despite the play calling).

What it comes down to is are you giving the team the best chance to win?
And those plays are magnified in crunch-time, with the game clearly on the line.

This regime has had too many clusterfarks that cost games.  I agree that most coaches make mistakes like that (Harbaugh made at least 2 memorable ones - one didn't cost the game against ND (not taking a knee or two); the other very well might have (having Luck throw on first down late against Cal in BG after a long catch and run by Toby).

Coach Shaw has been great for Stanford, which has had a remarkable W/L record under his leadership.  But it could have been even better, in BHA ways (to borrow from a Tedism), with just a little better sense of game situations by the play callers.

[TRD]

A play calling AI would be really interesting to see,

I think this is a very good idea.

Watson > Shaw, any day of the week.  (And I don't mean DeShaun Watson).

[TRD]

A play calling AI would be really interesting to see,

I think this is a very good idea.

Watson > Shaw, any day of the week.  (And I don't mean DeShaun Watson).

[TRD]

In what world is having Andrew Luck stop his march to the end zone in the Fiesta Bowl so you can have your "just getting back from injury" PK, who had already missed one FG, try a 37 yarder at the buzzer for the W? [Yeah, yeah, I know . . . in the world where he makes the FG].

This ^^^ was completely mind-boggling.

[TRD]

In what world is having Andrew Luck stop his march to the end zone in the Fiesta Bowl so you can have your "just getting back from injury" PK, who had already missed one FG, try a 37 yarder at the buzzer for the W? [Yeah, yeah, I know . . . in the world where he makes the FG].

This ^^^ was completely mind-boggling.

[Goose]

[quote author=gocard14 link=topic=16412.msg186100#msg186100 date=1484224794]


A play calling AI would be really interesting to see,

I think this is a very good idea.

Watson > Shaw, any day of the week.  (And I don't mean DeShaun Watson).
[/quote]

I am sure somebody, probably several groups, have tried it. Hard to score though, not like chess where the play call determines the outcome. "Good enough" plays well executed >> "Optimal" plays poorly executed. And it is hard to know, and tell the computer, which players can reliably do what etc. Also, you have to get a team to actually use the AI result or it won't be known if the plays were any good. It would be a fun experiment though.

[Goose]

[quote author=gocard14 link=topic=16412.msg186100#msg186100 date=1484224794]


A play calling AI would be really interesting to see,

I think this is a very good idea.

Watson > Shaw, any day of the week.  (And I don't mean DeShaun Watson).
[/quote]

I am sure somebody, probably several groups, have tried it. Hard to score though, not like chess where the play call determines the outcome. "Good enough" plays well executed >> "Optimal" plays poorly executed. And it is hard to know, and tell the computer, which players can reliably do what etc. Also, you have to get a team to actually use the AI result or it won't be known if the plays were any good. It would be a fun experiment though.