Mike McCarthy's Analytics Fraud

kumizi said:

This is a hilarious reach.

Coach makes a vague generalization that shows he doesn't understand analytics.

Homer fans: He's doing to cause he doesn't want to give away our strategy.

LOL WUT

Click to expand...

Prove to me that will be his gameplan on every Sunday. You can't. If you're niave enough to think that a NFL head coach will give away what his offensive tendencies will be on Sundays, then that's just too bad.

conner01 said:

Going into the last week we was 7-1 in games we ran 29% or more play action
I thought that was interring that half the games we didn’t
Sometimes you get behind and I get that but I think play action should be huge in our offense

Click to expand...

Yes, when you have a running threat, play-action should be a big part of your game.

That's really what we should be caring about rather than whether McCarthy is an analytics fraud or not.

The questions I want to see answered are:

If we run on first down and gain 3 yards or fewer, does that pretty much automatically mean the next play is a pass?

If we pass on first down and it's incomplete, does that pretty much automatically mean the next play is a run?

Are we going to run between the tackles on 60-plus percent of first downs?

I know there were exceptions to those rules under Garrett, but in most games, I could call whether the next play was going to be a run or a pass, and if I could do that, then defenses certainly could.

I don't care if it's analytics-based or just good coaching, I want to see this offense be unpredictable. Think it's a handoff? No, it's play-action. This is passing down, right? No, it's a draw.

Toruk_Makto said:

I don't have this data, no.

Click to expand...

What do you mean you don't have that data? Aren't you supposed to have all this bookmarked, interpreted, ready in case for some silly reason somebody says, "Prove it?" Asking for a friend.

ScipioCowboy said:

Let me preface my comments by saying I’m not contradicting you or arguing any point. I’m thinking aloud.

According to pundits and players, when an offense can run the ball well (however you reach that assessment), defenses are forced to commit extra defenders near the line of scrimmage, presumably making it more difficult to defend the pass.

In the aggregate, statistics show that running ball well does not correlate to passing the ball well. But I wonder how much those stats are skewed by differences in quarterback talent—ie Aaron Rodgers is going to be more efficient than Marcus Mariota no matter how many pass defenders he’s throwing against.

Do the metrics attempt to adjust for QB talent?

Click to expand...

it is not just that, there are several other issues with these stats that I have discussed with Adam before that he has had no answer for

these stats are collected without attention to time and context in game (ie 4th quarter or with lead or from behind or running out the clock etc), they are done with no correction for multiple analyses (which is a standard in large datasets) and with no sophisticated analysts like multiple logistic regression or C statistic which correct for biases

this publicity available data is absolutely garbage in and garbage out which is why you never hear coaches or DCs say stupid stuff like @Toruk_Makto is spouting. All good HCs and DCs always say that their first priority is to stop the run.

I’m sure NFL franchises have real data analytics that tell them the real FACTS

I’m supposed to believe @Toruk_Makto instead of what a SB winning HC in the NFL is saying?

as Clint Eastwood said: get off my lawn

gimmesix said:

Yes, when you have a running threat, play-action should be a big part of your game.

That's really what we should be caring about rather than whether McCarthy is an analytics fraud or not.

The questions I want to see answered are:

If we run on first down and gain 3 yards or fewer, does that pretty much automatically mean the next play is a pass?

If we pass on first down and it's incomplete, does that pretty much automatically mean the next play is a run?

Are we going to run between the tackles on 60-plus percent of first downs?

I know there were exceptions to those rules under Garrett, but in most games, I could call whether the next play was going to be a run or a pass, and if I could do that, then defenses certainly could.

I don't care if it's analytics-based or just good coaching, I want to see this offense be unpredictable. Think it's a handoff? No, it's play-action. This is passing down, right? No, it's a draw.

Click to expand...

I would sit on the couch and tell my son in law what was coming next
You have to get every competitive advantage in today’s nfl and that comes by being unpredictable

FuzzyLumpkins said:

The data was derived from an average. those stats do not apply to anything real. those were also biased for within games not for the season or other splits.

Click to expand...

So when 24 of 28 qbs all point in the same direction regardless of rushing efficiency or attempts you simply close your eyes and ears.

Here is the water. Can't make you drink.

conner01 said:

I would sit on the couch and tell my son in law what was coming next
You have to get every competitive advantage in today’s nfl and that comes by being unpredictable

Click to expand...

Right. I don't think what I was doing in predicting the plays was any different than every Cowboys fan was doing because we were so predictable. I absolutely hated it. I wanted to be wrong. I wanted to say, "This will be a run" and be surprised by a pass. Sometimes I was, but it rarely happened.

CATCH17 said:

I hope he does well but he isn’t changing from what he did in GB that got him fired.

Click to expand...

Doesn’t change the Superbowl he won either

So McCarthy has always called pass heavy in the first half to begin with. I don’t see what the problem is here except the crying from OP that he didn’t get the HC he wanted.

Toruk_Makto said:

Look. I'm an engineer by training and now work at a big bank in investments. So yeah I like math and "analytics." I think they are instructive especially when careful study shows where what you believe to be true is in fact false. People who understand these counter-intuitive truisms claim a consistent edge.

There was much talk about how much Mike has learned in his year hiatus and how analytics was a big part of the thing he leaned on and has incorporated. And good lord he is not off to a great start. In fact it seems he is a fraud.

See below:



This is an awful quote. What he says here is the exact opposite finding from the analytics community.

In fact this is something I looked into when studying Zeke's pedestrian efficiency numbers last year (As a reminder we greatly overpaid Zeke after greatly over drafting him). But I digress....I posted this in August of last year...



What this shows is that play action passing is the most effective passes a QB can throw. And it hardly matters if you run the ball a lot. Or if you run the ball a little bit. And it does not matter if you run the ball well. Or if you run the ball poorly.

The point? If McCarthy is getting this basic thing wrong what confidence do we have that he actually learned anything in the last year, analytics or otherwise? And just like the Garrett era we can expect to give away the small edges that smart coaches understand and benefit from.

And that's a shame.

Click to expand...

MM plans to have a staff of 12 (I think that was the number may have been 14) doing his analytics.

I dont believe the point is how proficient he is, but more so that he is taking it seriously. What will be more important imo is his staff and the choices he does or doesnt make based on the information he receives.

https://www.___GET_REAL_URL___/s/co...gressiveness-passing-tendency-challenges/amp/

visionary said:

it is not just that, there are several other issues with these stats that I have discussed with Adam before that he has had no answer for

these stats are collected without attention to time and context in game (ie 4th quarter or with lead or from behind or running out the clock etc), they are done with no correction for multiple analyses (which is a standard in large datasets) and with no sophisticated analysts like multiple logistic regression or C statistic which correct for biases

this publicity available data is absolutely garbage in and garbage out which is why you never hear coaches or DCs say stupid stuff like @Toruk_Makto is spouting. All good HCs and DCs always say that their first priority is to stop the run.

I’m sure NFL franchises have real data analytics that tell them the real FACTS

I’m supposed to believe @Toruk_Makto instead of what a SB winning HC in the NFL is saying?

as Clint Eastwood said: get off my lawn

Click to expand...

Agreed. Some stats may help you improve by pointing out things you need to work on, but they don't help in game planning. You need context to understand what those stats mean, how they can be applied to game situations. And some stats, completions, yards per carry, YAC mean very little since they will change game to game.

CowboysWillRise said:

MM plans to have a staff of 12 (I think that was the number may have been 14) doing his analytics.

I dont believe the point is how proficient he is, but more so that he is taking it seriously. What will be more important imo is his staff and the choices he does or doesnt make based on the information he receives.

Click to expand...

Right, analytics could be the wrong thing if it makes McCarthy develop tendencies. Hopefully, the purpose will be to keep us from having tendencies as much as it is to provide statistical data, to make McCarthy more aware, because most everyone develops predictable patterns if they are not self-aware enough to keep from it.

Not even one week and our resident "experts" are already dogging McCarthy worse than Garrett. So predictable... hate the qb, owner, rb, linebackers. Just hate everything. ..

THEHEREAFTER said:

Not even one week and our resident "experts" are already dogging McCarthy worse than Garrett. So predictable... hate the qb, owner, rb, linebackers. Just hate everything. ..

Click to expand...

I'm waiting for a FIRE MCCARTHY post any minute now!

Toruk_Makto said:

Rushing attempts hardly matters. Rushing effectiveness doesn't matter.

Taking a small Dak sample size or a few outliers doesn't disprove that correlation. Often times it confirms it.

In any case McCarthy is saying something demonstrably untrue.

Click to expand...

Toruk_Makto said:

Look. I'm an engineer by training and now work at a big bank in investments. So yeah I like math and "analytics." I think they are instructive especially when careful study shows where what you believe to be true is in fact false. People who understand these counter-intuitive truisms claim a consistent edge.

There was much talk about how much Mike has learned in his year hiatus and how analytics was a big part of the thing he leaned on and has incorporated. And good lord he is not off to a great start. In fact it seems he is a fraud.

See below:



This is an awful quote. What he says here is the exact opposite finding from the analytics community.

In fact this is something I looked into when studying Zeke's pedestrian efficiency numbers last year (As a reminder we greatly overpaid Zeke after greatly over drafting him). But I digress....I posted this in August of last year...



What this shows is that play action passing is the most effective passes a QB can throw. And it hardly matters if you run the ball a lot. Or if you run the ball a little bit. And it does not matter if you run the ball well. Or if you run the ball poorly.

The point? If McCarthy is getting this basic thing wrong what confidence do we have that he actually learned anything in the last year, analytics or otherwise? And just like the Garrett era we can expect to give away the small edges that smart coaches understand and benefit from.

And that's a shame.

Click to expand...

Boy I read that same quote from Mike and can take a SLIGHTLY different meaning out of it. First let’s state that analytics is a tool to help formulate a decision. It is not the tell all. Analytics does not and in no way can take in account the quality of the players (offensive or defensive) and the execution of the plays. Example, teams in the early nineties did not need analytics but if used it would tell them that Dallas was going to run the ball right down your throats. They knew it and couldn’t do anything about because the Cowboys players was so much better then the other teams and the execution of the plays was formidable. Again it’s a tool used to help give an edge where possible.

But let’s get back to Mike’s quote and the article mentioned. Yes he stated that you have to run the ball to have play action work effectively and the numbers show there wasn’t much difference if the team rushed ten times or upwards of twenty. Mike stated you had to run the ball, he never stated how many times but just that you had to try. So I am going to assume that you assume he is referring to a large amount.

Now let’s get to the second part of his quote that I think you are trying to call him out on. He stated that “when you are able to run the ball effectively, there is an opportunity to get more chunk plays”. The article was referencing passer rating on whether there were a few attempts or many, it never touched on large gaining plays. With that said my interpretation was that Mike believes through his analytics that when the run game is effective, there is a much better chance for a big gain play.

Without being able to question him and dig into his theory more, it’s not fair to label him a fraud.


Last note: If you research my posts going back several months to a year, you will find several that state I have a upper managerial position in one of the largest analytical firms in the country.

just saying!

Captain-Crash said:

yeah, let's get Garrett back. I already miss the loser guy.

Click to expand...

Crazy right?

#DontpayDak

PhillyCowboysFan said:

Boy I read that same quote from Mike and can take a SLIGHTLY different meaning out of it. First let’s state that analytics is a tool to help formulate a decision. It is not the tell all. Analytics does not and in no way can take in account the quality of the players (offensive or defensive) and the execution of the plays. Example, teams in the early nineties did not need analytics but if used it would tell them that Dallas was going to run the ball right down your throats. They knew it and couldn’t do anything about because the Cowboys players was so much better then the other teams and the execution of the plays was formidable. Again it’s a tool used to help give an edge where possible.

But let’s get back to Mike’s quote and the article mentioned. Yes he stated that you have to run the ball to have play action work effectively and the numbers show there wasn’t much difference if the team rushed ten times or upwards of twenty. Mike stated you had to run the ball, he never stated how many times but just that you had to try.

Now let’s get to the second part of his quote that I think you are trying to call him out on. He stated that “when you are able to run the ball effectively, there is an opportunity to get more chunk plays”. The article was referencing passer rating on whether there were a few attempts or many, it never touched on large gaining plays. With that said my interpretation was that Mike believes that when the run game is effective, there is a much better chance for a big gain play.

Without being able to question him and dig into his theory more, it’s not fair to label him a fraud.


Last note: If you research my posts going back several months to a year, you will find several that state I have a upper managerial position in one of the largest analytical firms in the country.

just saying!

Click to expand...

Very well said. I was just about to post something very similar because GB's bread n butter chunk play under McCarthy was the PA double move/deep pass to Jordy Nelson.

Anyhow, this whole thread is complete non-sense and really is a good case study into how fickle this fanbase has become.

Imagine hiring a HC with SB pedigree and then subsequently calling him a fraud based on a 20 sec answer to an interview question he gave less than a week on the job.

Bluestang said:

Imagine hiring a HC with SB pedigree and then subsequently calling him a fraud based on a 20 sec answer to an interview question he gave less than a week on the job.

Click to expand...

Let alone being called a fraud by a “banker”. SMGDH!

brooksey1 said:

What a terrible thread

Click to expand...

True, but it is slightly amusing to read along as Fuzzy swats this guy around.

A domestic cat is the most ruthless killer in the world.
It will kill for fun. Give'em all the cat food you can.
They will still kill for sport and the enjoyment of watching their prey die.

That is my visual. Fuzzy is the 20lb tabby cat that will make a bug suffer a
painful death... because he can.
Die bug, die... lol (metaphorical, of course... smh)