Will Linehan go more no huddle?

[khiladi]

Anecdotal evidence is not the way around that, though. When you select certain games which seem more characteristic of the criteria you want to examine, it isn't because they have the virtue of being games, it's because you haven't really clearly defined those criteria. Why not just set the criteria first, then go back to the games (all the plays of all the games, not just the ones you remember) and find out if you're right or wrong? That way you eliminate confirmation bias and get a little closer to objectivity.

When you become more interested in finding out than in finding a way to be right, you'll end up being right a lot more often anyway.

I'm objecting to the validity of your criteria, based upon these examples that aren't anecdotal. We scored 14 pts on TDs against Chicago when the game was already out of hand. Those 14 pts add to our average, but also obscures our offensive performance to a degree when comparing against other teams. Chicago wasn't playing aggressive defense anymore.

You don't win a game by targetting some mathematical criteria of averages over 16 games in relation to other teams. Why should I accept that as the indicator of a good offense?

 

[khiladi]

i'm objecting to the validity of your criteria, based upon these examples that aren't anecdotal.

This should say "I'm simply objecting". The point being is I'm not trying to establish an alternate criteria in this context. The only point I'm refuting is the notion that the defense is solely to blame for our ills, because our offense was this high-octane powerhouse that couldn't be stopped (hyperbole) and if the defense did it's job we would have made the play-offs. That is an absolute farce. We lost games because of the offense sucking as well, and I mean sucking.

 

[Redball Express]

Thus, being a Garrett homer...

Demoted?

because he's letting somebody else call plays..?

That's not demoted.

Now Callahan..he was demoted.

I didn't care at all for his playcalling and it made the games way too close all the time with the 3 and outs.

Tony Romo was having problems throwing anything more than 15 to 20 yards down field.

So when the QB is hurt and not 100%..that's a problem.

It changes everything offensively.

So the playcalling gets very conservative.

Hopefully with Linehan..we will change that..

and a healthy Romo would be helpful, too.

We'll see.

 

[percyhoward]

I'm objecting to the validity of your criteria, based upon these examples that aren't anecdotal. We scored 14 pts on TDs against Chicago when the game was already out of hand. Those 14 pts add to our average, but also obscures our offensive performance to a degree when comparing against other teams. Chicago wasn't playing aggressive defense anymore.

You're talking about a specific game, or games, rather than every play of every game. You're using a small sample (or several small samples) to try to support your point. That's the definition of anecdotal evidence, or cherry picking.

You don't win a game by targeting some mathematical criteria of averages over 16 games in relation to other teams. Why should I accept that as the indicator of a good offense?

We're not trying to win a football game here. You're trying to say the offense wasn't as good as the numbers suggest. You're using as evidence some plays of some games. I'm using all the plays that you're using, plus the ones you're ignoring.

Again, set your parameters first. Filter out the factor(s) you believe to be skewing the numbers. (Say, only the drives that began inside our own 35, or only the plays that took place when the margin was + or - 8 points, or points scored only against defenses ranked in the top half of the league). Then look at every play. That's the only fair and accurate way to do it. If you don't know how to find this information, give me your parameters, and I'll look it up for you.

 

[Brooksey]

It's hard to come up with any evidence that Garrett is more responsible (or even as responsible) for the offense's success than Romo, or any other key player(s). So here, you may actually have a very good point.

There is a way to figure out how much of the offense's scoring was due to field position. Simply look only at the offense's drives that did NOT benefit from good starting field position. If you look at all drives in the league last year that began inside teams' own 35-yard line, Dallas ranked 3rd in the NFL in the percentage of drives that resulted in a TD. That's an average starting field position of the 20-yard line, and that's where more than 75% of all the league's drives started.

So that offensive ranking is real, and who gets the credit for it is more open to interpretation. If Garrett hurt that ranking, the obvious implication is that this offense would have otherwise been second only to the Broncos.

Everything can't be explained with stats, like situational failures in big/spots games like SD,KC, DET, GB etc. Do you have a stat for when it counted most? Because that stat is not a seasonal average. You think it's hard to come up with any evidence Garrett is more responsible? He's the head coach, that's all the evidence you need. The play calling imbalance was 65-35% pass/run ratio, while the ratio of the NFC powerhouses, SF and Seattle is 55%-45% Run/Pass. These two teams are balanced, like we were in the 90's winning three Super bowls. Passing more does not equal winning more, in fact I'm sure your research tool box can spit out a stat thats says balance wins more. Not targeting DEZ in big spots/games because of double teams is a fail, the time management has been a fail.

Do you have stats for these three failures? No you don't because they are situational big spot failures that seasonal averages don't account for.

Also for your seasonal averages, there's an old statistic that hangs true to this day when evaluating stats or profit and loss for any company. Take out the one time expenses or revenue's for a true look at the P&L. We never repeated the Denver performance so it means nothing, it was just an anomaly that shows there is questionable coaching decisions involved here for never running it again.

When evaluating the offensive seasonal averages, you must account for the Denver game and the added turnovers from our defense for a true look at how good we were from year to year. If not your stats for seasonal averages are not even accurate.

 

[Brooksey]

This should say "I'm simply objecting". The point being is I'm not trying to establish an alternate criteria in this context. The only point I'm refuting is the notion that the defense is solely to blame for our ills, because our offense was this high-octane powerhouse that couldn't be stopped (hyperbole) and if the defense did it's job we would have made the play-offs. That is an absolute farce. We lost games because of the offense sucking as well, and I mean sucking.

Nice job owning the thread Khiladi. The Pass Happy Garrett Homers can't admit to any of this so they bring in the research tool for the "seasonal average argument". The average points per drive and differential covers up the big-game, big-spot spot failures.

Funny thing is JJ doesn't try to hide it, he pulled JG from play calling, then Callahan.

Ignore this, everyone knows its nonsense and what their agenda is.

 

[percyhoward]

Everything can't be explained with stats, like situational failures in big/spots games like SD,KC, DET, GB etc. Do you have a stat for when it counted most? Because that stat is not a seasonal average. You think it's hard to come up with any evidence Garrett is more responsible? He's the head coach, that's all the evidence you need. The play calling imbalance was 65-35% pass/run ratio, while the ratio of the NFC powerhouses, SF and Seattle is 55%-45% Run/Pass. These two teams are balanced, like we were in the 90's winning three Super bowls. Passing more does not equal winning more, in fact I'm sure your research tool box can spit out a stat thats says balance wins more. Not targeting DEZ in big spots/games because of double teams is a fail, the time management has been a fail.

Do you have stats for these three failures? No you don't because they are situational big spot failures that seasonal averages don't account for.

Also for your seasonal averages, there's an old statistic that hangs true to this day when evaluating stats or profit and loss for any company. Take out the one time expenses or revenue's for a true look at the P&L. We never repeated the Denver performance so it means nothing, it was just an anomaly that shows there is questionable coaching decisions involved here for never running it again.

When evaluating the offensive seasonal averages, you must account for the Denver game and the added turnovers from our defense for a true look at how good we were from year to year. If not your stats for seasonal averages are not even accurate.

You can account for the influence of your best game (like the Denver game), and I have. You can account for the effect of the defense on field position, and I have. I'll repost these for you when I have time. When you filter out that interference, nothing changes relative to the rest of the league's offenses. That's because you also have to filter out every other team's best game, and every other team's advantageous field position.

What you cannot do is filter out games that you consider "big" solely on the basis that we lost the game. That's why you set your parameters before you look at the plays/games. You kind of have to have guts to do this if you're worried that your suspicions may not be correct, because remember...you're looking at every play of every game--not just the failed plays in the losses. What do you think skewed the numbers? Filter out all those situations, and whatever is left should tell you what you want to know. You may not like what you find.

 

[ShiningStar]

You can account for the influence of your best game (like the Denver game), and I have. You can account for the effect of the defense on field position, and I have. I'll repost these for you when I have time. When you filter out that interference, nothing changes relative to the rest of the league's offenses. That's because you also have to filter out every other team's best game, and every other team's advantageous field position.

What you cannot do is filter out games that you consider "big" solely on the basis that we lost the game. That's why you set your parameters before you look at the plays/games. You kind of have to have guts to do this if you're worried that your suspicions may not be correct, because remember...you're looking at every play of every game--not just the failed plays in the losses. What do you think skewed the numbers? Filter out all those situations, and whatever is left should tell you what you want to know. You may not like what you find.

while this is good, we also may not find the whole truth. Stats dont always tell the whole story. Good players vs bad, schemes vs schemes, one good day vs a bad day, the factors go on and on.
its like how i try to show people, the stats dont tell the whole story. You want to get to the whole truth sometimes, we have to look at reality, sometimes what we think of a big game or a good player are not just that good.

 

[Idgit]

while this is good, we also may not find the whole truth. Stats dont always tell the whole story. Good players vs bad, schemes vs schemes, one good day vs a bad day, the factors go on and on.
its like how i try to show people, the stats dont tell the whole story. You want to get to the whole truth sometimes, we have to look at reality, sometimes what we think of a big game or a good player are not just that good.

Stats are measurements. You can measure the wrong thing, and you can do the wrong thing with the data. But give me smart people, measuring the right things, and drawing sound conclusions from the data over unsubstantiated impressions or strong opinions any day of the week. That's the debate going on in this thread, as near as I can tell, and the likelihood of who's more likely to be right and who's just guessing is pretty obvious.

 

[ShiningStar]

Stats are measurements. You can measure the wrong thing, and you can do the wrong thing with the data. But give me smart people, measuring the right things, and drawing sound conclusions from the data over unsubstantiated impressions or strong opinions any day of the week. That's the debate going on in this thread, as near as I can tell, and the likelihood of who's more likely to be right and who's just guessing is pretty obvious.

i dont disagree, i just peronally dont always buy into the stat hype. Sometimes its relied on to much of those in the media, to compensate for something else. before any posters get startled, i said media.

 

[percyhoward]

while this is good, we also may not find the whole truth. Stats dont always tell the whole story. Good players vs bad, schemes vs schemes, one good day vs a bad day, the factors go on and on.
its like how i try to show people, the stats don't tell the whole story. You want to get to the whole truth sometimes, we have to look at reality, sometimes what we think of a big game or a good player are not just that good.

All stats are not created equal. Obviously, the best stats, correctly applied, will let you compare team to team, or season to season, or player to player better than anyone's opinion ever would. There are many, many stats published that have no bearing on winning whatsoever. Unfortunately, there's no law that says that every time a bad stat is published that it has to carry a disclaimer. Instead of simply throwing up your hands, you could look into which stats correlate highly with winning. They'll tell you a lot.

Points per drive differential, and pass rating differential are the best ones. Teams that rank the highest in those categories win the most games, year in and year out. The factors that you say "go on and on" either aren't nearly as important as those two, or haven't yet been identified.

 

[percyhoward]

i dont disagree, i just peronally dont always buy into the stat hype. Sometimes its relied on to much of those in the media, to compensate for something else. before any posters get startled, i said media.

We've all seen a writer throw out a bunch of stats just to meet a deadline. We've also seen that they can make things worse by misinterpreting the stat, or by just getting it flat wrong. We've even seen writers who seize on one stat that seems to back up their existing opinion, and ignore other, better stats that contradict it. A good rule of thumb is "does this make sense?" Does it make sense, for example, to focus on 3rd down conversion percentage without also looking at 1st and 2nd down? Does it make sense to value a receiver's total yardage over the pass rating on his targets? Not really.

 

[khiladi]

You're talking about a specific game, or games, rather than every play of every game. You're using a small sample (or several small samples) to try to support your point. That's the definition of anecdotal evidence, or cherry picking.

I would argue that it is you who are cherry-picking by choosing a statistic that has no bearing of when we scored. You do realize that 2 games that I mentioned above, in terms of absolutely useless scoring in how they demonstrate some superior offensive scoring production, are enough to show that our numbers are skewed. I really don't have to go into more to demonstrate the point. When Sean Lee gives you an INT to score from 7 yards out, when two of your TDs are when your getting blown out by Chicago, one of them a fly pattern on a play where Romo puts Dez in the slot changing the play, outside the context of the system, then your totally cherry-picking context. That's 21 points, which is a games worth of offenseive production for us in many contexts.

All your simply doing is presenting averages without context.

We're not trying to win a football game here. You're trying to say the offense wasn't as good as the numbers suggest. You're using as evidence some plays of some games. I'm using all the plays that you're using, plus the ones you're ignoring.

Yes, we are trying to win football games here. When you state the reason we lost is because of the defense and using statistics to prove it, that is exactly what we are talking about. Averages don't take context into account in this respect. Further, you don't win by hitting targets of mathematical formulas. We lost our last game, though our passing yardage dwarfed Philadelphia and our total offensive yardage did as well.

Again, set your parameters first. Filter out the factor(s) you believe to be skewing the numbers. (Say, only the drives that began inside our own 35, or only the plays that took place when the margin was + or - 8 points, or points scored only against defenses ranked in the top half of the league). Then look at every play. That's the only fair and accurate way to do it. If you don't know how to find this information, give me your parameters, and I'll look it up for you.

I don't have to do a full statistical analysis to prove my point, because I'm not getting paid for it and the above was sufficient to demonstrate my point that averages don't necessarily mean what people claim them to mean. Averages don't tell me anything about the context of a game.

 

[ShiningStar]

We've all seen a writer throw out a bunch of stats just to meet a deadline. We've also seen that they can make things worse by misinterpreting the stat, or by just getting it flat wrong. We've even seen writers who seize on one stat that seems to back up their existing opinion, and ignore other, better stats that contradict it. A good rule of thumb is "does this make sense?" Does it make sense, for example, to focus on 3rd down conversion percentage without also looking at 1st and 2nd down? Does it make sense to value a receiver's total yardage over the pass rating on his targets? Not really.

i agree. with all that, but i see the other side as well. im glad you made that post because it explains the whole ordeal better. My point is the stats show the CB gave up a lot of yards, and everyone screams well he had a bad day, true, but he didnt give up points either. The other CB could have given up less yards, but more points were scored on him. Other things factor in but im sure you are wise enough to know. Thats all im saying.

 

[khiladi]

When somebody says "his eyes don't lie" ir doesn't necessarily mean he's saying something subjective, it simply means he doesn't buy the stats argument and how it's used to adequately reflect reality. For the last 7 years, we've heard about how Garrett has had a top 5 offense because of passing yardage. No, his offense was a joke, especially in the context of the talent level he's had.

 

[khiladi]

i agree. with all that, but i see the other side as well. im glad you made that post because it explains the whole ordeal better. My point is the stats show the CB gave up a lot of yards, and everyone screams well he had a bad day, true, but he didnt give up points either. The other CB could have given up less yards, but more points were scored on him. Other things factor in but im sure you are wise enough to know. Thats all im saying.

The Detroit game is a perfect example. We got 4 TOs on defense, with Sean Lee gifting the offense a TD. The Cowboys only capitalized once from what i remember. And you negate that TD, the offense wouldn't have been up for the defense to 'lose the game' at the end. If the offense was good and considering we were playing a hobbling Detroit game, how come Detroit was even in the game and we had to score 17 points in the the late 3rd and 4th to take the lead when we clearly went up-tempo and pushed the pace? When we when with our crappy system, crickets were chirping in terms of scoring.

But all we hear was about how Megatron abused the defense.

 

[percyhoward]

When evaluating the offensive seasonal averages, you must account for the Denver game and the added turnovers from our defense for a true look at how good we were from year to year. If not your stats for seasonal averages are not even accurate.

To filter out the effect of the Denver game, I looked at every offense that averaged over 2.00 points per game, and took out their best game.
1 Den 2.83 vs Phi: 45 points/9 drives (5.0)
2 SD 2.33 vs NYG: 37 points/9 drives (4.1)
3 NO 2.29 vs Dal: 49 points/9 drives (5.4)
4 Dal 2.18 vs Den: 48 points/11 drives (4.4)
5 Chi 2.16 vs Dal: 45 points/8 drives (5.6)
6 GB 2.15 at Min 37 points/8 drives (4.6)
7 Phi 2.12 vs Chi 45 points/10 drives (4.5)
8 Sea 2.12 vs Jac 45 points/13 drive (3.5)
9 NE 2.10 vs Pit: 55 points 13 drives (4.2)

Here are the rankings when you take out that game.
1 Den 2.70
2 SD 2.23
3 NO 2.13
4 Dal 2.04
5 GB 2.04
6 Sea 2.02
7 Chi 2.00
8 Phi 1.99
9 NE 1.95

Every team's number goes down .10 -.16 of a point per drive. With the Denver game included, Dallas ranks 4th, hundredths of a point ahead of the Bears. Without it, Dallas still ranks 4th, thousandths of a point ahead of the Packers. The Denver game didn't skew our numbers any more or less than it did those of any other team's offense, and it didn't change our ranking at all.


To filter out the effect of the added turnovers from our defense, I also looked at every drive in the NFL in 2013 that did NOT begin after a turnover.
1. Broncos 179 drives 59 TD 33.0% 2.68 points/drive
2. Saints 163 drives 42 TD 25.7% 2.21
3. Cowboys 162 drives 39 TD 24.1% 2.09
4. Bears 162 drives 38 TD 23.5% 2.05
5. Eagles 175 drives 41 TD 23.4% 1.95

So if you take out all the drives that started after turnovers in the NFL in 2013, we'd have actually moved up to the 3rd best offense in both TD percentage and points per drive.

 

[Idgit]

When somebody says "his eyes don't lie" ir doesn't necessarily mean he's saying something subjective, it simply means he doesn't buy the stats argument and how it's used to adequately reflect reality. For the last 7 years, we've heard about how Garrett has had a top 5 offense because of passing yardage. No, his offense was a joke, especially in the context of the talent level he's had.

"Eyes don't lie" is paraphrasing for something being completely subjective.

Your supporting anecdote is an example of a statistic that doesn't correlate with winning football games being used as a measure of offensive efficiency. In other words, it's an example of misinterpreting the importance of the data collected and not an argument against using statistics properly.

 

[percyhoward]

I would argue that it is you who are cherry-picking by choosing a statistic that has no bearing of when we scored. You do realize that 2 games that I mentioned above, in terms of absolutely useless scoring in how they demonstrate some superior offensive scoring production, are enough to show that our numbers are skewed. I really don't have to go into more to demonstrate the point. When Sean Lee gives you an INT to score from 7 yards out, when two of your TDs are when your getting blown out by Chicago, one of them a fly pattern on a play where Romo puts Dez in the slot changing the play, outside the context of the system, then your totally cherry-picking context. That's 21 points, which is a games worth of offenseive production for us in many contexts.

All your simply doing is presenting averages without context.

Yes, we are trying to win football games here. When you state the reason we lost is because of the defense and using statistics to prove it, that is exactly what we are talking about. Averages don't take context into account in this respect. Further, you don't win by hitting targets of mathematical formulas. We lost our last game, though our passing yardage dwarfed Philadelphia and our total offensive yardage did as well.

I don't have to do a full statistical analysis to prove my point, because I'm not getting paid for it and the above was sufficient to demonstrate my point that averages don't necessarily mean what people claim them to mean. Averages don't tell me anything about the context of a game.

Since there's no point in talking about Lee's INT as it relates to the offense, when the discussion is past that already, I assume you haven't understood. I've already shown you that you can completely leave out not just the score following Lee's INT, but any score that followed any opponent's turnover all season long. And we actually move up the offensive rankings.

I also showed that we move up in the offensive rankings when involved in close games (margin + or - 8 points). More of our scores happened in the context of close games than did the scores of most teams. That means it makes no sense for you to go on about scores in blowout losses. Fewer of our scores happened in that context than most teams' did, after all. You do get this, right? I'm taking your parameters and looking at everything that falls within them, for every team. That's the only way do it that makes any sense. You can't just look at one team, and assume these same scenarios aren't repeated over and over across the league. They are, to varying degrees.

For example, Dallas was 5th in total offensive TD. When you remove all drives that began after an opponent's turnover, Dallas is still 5th in total offensive TD. For the sake of comparison, the Chiefs were 9th in total offensive TD. When you remove all drives that began after an opponent's turnover, the Chiefs drop to 13th. Want to see an offense whose numbers are skewed by their defense's takeaways, look at the Chiefs.

Want to see an offense that scored a lot of garbage points? That would be Philly. They were 2nd in the NFL in total offensive TD. (Remember, Dallas was 5th). When you look only at TD scored when the margin was 8 points or less, the Eagles drop to 7th. Dallas moves up to 2nd.

No fancy math formulas here, I'm just counting touchdowns.

 

[ConstantReboot]

We need more No huddles.

The more we use less of Garrett's playbook, the better the offense becomes.

I know more people say things like protecting the defense. Well the offense seems better with the no huddle than Garrett's 3 pass plays that chew 6 seconds off the clock. Just say no to Garrett's playbook.