re: Predictability of LSU on Offense since 2008 - Why the box is stacked

Nice fish Choctaw, when and where did you earn them?

Cpt. Bengal, Would it be accurate to interpret your data like this: For 2010, 2011, 2012, 2013 and 2014 so far, LSU ran the ball at least (about) 70% of the time on 1st down?

Without info. on the distance needed for a 1st down, it seems hard to make any inferences on tendencies on 2nd and 3rd downs. But on first down, it appears other teams can really stack the box to stop the run.

Your post says it all.

quote:

---

2008: Offense(55th) Defense(32nd)
2009: Offense(112th) Defense(26th)
2010: Offense(86th) Defense(12th)
2011: Offense(86th) Defense(2nd)
2012: Offense(85th) Defense(8th)
2013: Offense(35th) Defense(15th)
2014: Offense(60th) Defense(17th) <-Thus far

---

So Miles, how many excuses are there left ????

Cholly Mac had a winning record too, but could not win the big one.
YOUR style will not win it either... FRANKLY, you were lucky in 2008 that it was Ohio State and Big 10 football.

CptBengal states:

quote:

---

We have all seen opposing defenses stack the box on first down against Les Miles since 2008. Why is this...simple, because we are exceedingly more likely to run than pass....

You would expect this to be true if we were running the ball well over this time. Funny enough, however, the yardages tell a different story:

---

CptBengal then uses skewed facts and analysis to misrepresent the facts.



CptBengal states:

quote:

---

Do tell.... as I have already pointed out those flaws myself.

---

No, CptBengal didn't. He said that he wished he had stats for every game. Those stats are available. He simply choose not to use them.

quote:

---

First, you don't understand the flaws in your data set. (Or maybe you do understand the flaws, and you prefer to present misinformation to advance your agenda.)

---

CptBengal’s data set starts in 2008. LSU lost five games in 2008. LSU lost four games the next year, 2009. LSU lost two games in 2010 and one game in 2011. LSU lost only three games in 2012 and 2013. Are you seeing a trend? More on that in a moment.

quote:

---

Second, you misanalyse the data set to present misinformation to advance your agenda.

---

Rushing attempts increase when a team is successful. The more comfortable the lead in a game, the more likely a team is to run the ball. Conversely, a team that is losing in a game is more likely to pass. LSU was less successful in the beginning years of CptBengal’s chart and generally more successful as the years progress.

quote:

---

Third, you chose presentations that misrepresent the facts to provide misinformation to advance your agenda.

---

YPC/YPA presents a decrease in YPC even if it really is an increase in YPA. If the YPC remain constant and the YPA increase, CptBengal’s quotient (YPC/YPA) will decrease. Put in more concrete terms, if LSU rushes for the same number of yards every season, but increases the YPA every season, CptBengal will get a downward slope in his graph even though the rushing has remained constant and the offense as a whole has improved.

As you know, as we all know, Jennings has been killing it in YPA. Mettenberger also tended to complete the long ball. In short, the YPA increased over the last few years making the YPC/YPA decrease. Thus, his graph shows a downward trend.

However, YPC has generally increased over that timeframe.



Why would you look at YPC/YPA to analyze running success? Why, not look solely at YPC? Why, unless you wanted to misrepresent the data?

quote:

---

Fourth, you include incomplete facts to provide misinformation to advance your agenda.

---

2014 is only four games. Why compare it to the rest of the years that are complete years? It's only four games, less than a third of the total games to be played (12+bowl). You even said that "five data points sucks."

I think everyone can figure out why you included it. It helps CptBengal misrepresent the facts.

First, everyone knows that Jennings is killing it in YPA this year, so if he includes this year's YPA it helps his graph. (Again, why would he include YPA to analyze the success of a rushing attack? Oh yeah, to push his agenda and misrepresent the facts.)

Second, everyone knows that teams tend to run more and pass less when they are breaking in a new QB. Since LSU is breaking in two new QBs this year, LSU was going to run it more and pass it less this year. Thus, it helps his graph to include this year even though less than a third of the games have been played.

quote:

---

You would expect this to be true if we were running the ball well over this time. Funny enough, however, the yardages tell a different story:
CptBengal then uses skewed facts and analysis to misrepresent the facts.

---

I didnt "skew" or "misrepresent" anything. I calculated a statistic and then discovered a relationship of that statistic through time. The fact you dont like what the result of that analysis is..isnt a skew (A skew is a mathematical term to describe a feature of a distribution) and it isnt misrepresented (misrepresentation is providing one statistic and then calling it something else, like you did yesterday)

quote:

---

No, CptBengal didn't. He said that he wished he had stats for every game. Those stats are available. He simply choose not to use them.

---

What I said is I would need every play in every game. Those stats are available, play by play, but it would take far too long to collate from the game summaries. If you have access to a nice database, by all means, share and we can work on the logistic regression together. It would be fun!

quote:

---

First, you don't understand the flaws in your data set. (Or maybe you do understand the flaws, and you prefer to present misinformation to advance your agenda.)
CptBengal’s data set starts in 2008. LSU lost five games in 2008. LSU lost four games the next year, 2009. LSU lost two games in 2010 and one game in 2011. LSU lost only three games in 2012 and 2013. Are you seeing a trend? More on that in a moment.

---

By all means, if you would like to include wins/losses thaty can be added into modeling. My model made no assumptions on wins/losses. An analysis deals with a hypothesis, ONE. Mine was that less miles is predictable, and the data bore that out. If you would like to propose another hypothesis on wins and losses we can also discuss how to analyze that.

quote:

---

Rushing attempts increase when a team is successful. The more comfortable the lead in a game, the more likely a team is to run the ball. Conversely, a team that is losing in a game is more likely to pass. LSU was less successful in the beginning years of CptBengal’s chart and generally more successful as the years progress.

---

But you were talking wins/losses, not margin of victory.

See, this is your problem, you dont even understand the dependency of the variables on the marginal probabilities, yet you want to claim my analysis is wrong. It's laughable.

quote:

---

YPC/YPA presents a decrease in YPC even if it really is an increase in YPA. If the YPC remain constant and the YPA increase, CptBengal’s quotient (YPC/YPA) will decrease. Put in more concrete terms, if LSU rushes for the same number of yards every season, but increases the YPA every season, CptBengal will get a downward slope in his graph even though the rushing has remained constant and the offense as a whole has improved.

---

Go read my OP again. I SPECIFICALLY REFERENCED YPC:YPA. I made no distinction on YPC, because that wasnt the analysis.

It's funny that you think the analysis is wrong because I didnt use the variable you wanted, then are foolish enough to claim I am "skewing" things by not doing so.

Moreover, I used the ratio because we are looking at the hypothesis of predictability, NOT yards per play. In that binomial outcome there are two options and BOTH need to be represented in a probabilistic framework for the analysis to make sense.

quote:

---

2014 is only four games. Why compare it to the rest of the years that are complete years? It's only four games, less than a third of the total games to be played (12+bowl). You even said that "five data points sucks."

I think everyone can figure out why you included it. It helps CptBengal misrepresent the facts.

---

5 data points sucks because of the overall N you blithering tart. Good god.

I'll tell you what, you give me access to a database and instead I'll do 2014 as a posterior Bayesian analysis....deal?

quote:

---

Second, everyone knows that teams tend to run more and pass less when they are breaking in a new QB.

---

LINK?

aTm doesnt seem to be doing so? I like how you just make up things with no evidence, it's comical.

quote:

---

Reserving this spot to destroy your post when I'm done in my meeting.

---

Waiting patiently for you to explain why YPC/YPA is a more correct way to measure rushing success than simply YPC.

quote:

---

Cholly Mac had a winning record too, but could not win the big one.
YOUR style will not win it either.

---

he already has.....along with 3 SEC West titles and 2 SEC championships.

quote:

---

FRANKLY, you were lucky in 2008 that it was Ohio State and Big 10 football.

---

nothing better than seeing an LSU "fan" shite all over a NC winning team. gotta love it

I think he remembers running it on Arkie behind 14 points and then again against Georgia in the SEC CG. Both knew we were going to run it and they just couldn't stop it. That line was a thing of beauty with Faulk at LT.

Below you find my responses, but to this question:

quote:

---

Waiting patiently for you to explain why YPC/YPA is a more correct way to measure rushing success than simply YPC.

---

I wasnt measuring rushing. Are you really that stupid where as I lay out the hypothesis you cant even read what I wrote?






MY REPLY TO YOUR bullshite:

quote:

---

You would expect this to be true if we were running the ball well over this time. Funny enough, however, the yardages tell a different story:
CptBengal then uses skewed facts and analysis to misrepresent the facts.

---

I didnt "skew" or "misrepresent" anything. I calculated a statistic and then discovered a relationship of that statistic through time. The fact you dont like what the result of that analysis is..isnt a skew (A skew is a mathematical term to describe a feature of a distribution) and it isnt misrepresented (misrepresentation is providing one statistic and then calling it something else, like you did yesterday)

quote:

---

No, CptBengal didn't. He said that he wished he had stats for every game. Those stats are available. He simply choose not to use them.

---

What I said is I would need every play in every game. Those stats are available, play by play, but it would take far too long to collate from the game summaries. If you have access to a nice database, by all means, share and we can work on the logistic regression together. It would be fun!

quote:

---

First, you don't understand the flaws in your data set. (Or maybe you do understand the flaws, and you prefer to present misinformation to advance your agenda.)
CptBengal’s data set starts in 2008. LSU lost five games in 2008. LSU lost four games the next year, 2009. LSU lost two games in 2010 and one game in 2011. LSU lost only three games in 2012 and 2013. Are you seeing a trend? More on that in a moment.

---

By all means, if you would like to include wins/losses thaty can be added into modeling. My model made no assumptions on wins/losses. An analysis deals with a hypothesis, ONE. Mine was that less miles is predictable, and the data bore that out. If you would like to propose another hypothesis on wins and losses we can also discuss how to analyze that.

quote:

---

Rushing attempts increase when a team is successful. The more comfortable the lead in a game, the more likely a team is to run the ball. Conversely, a team that is losing in a game is more likely to pass. LSU was less successful in the beginning years of CptBengal’s chart and generally more successful as the years progress.

---

But you were talking wins/losses, not margin of victory.

See, this is your problem, you dont even understand the dependency of the variables on the marginal probabilities, yet you want to claim my analysis is wrong. It's laughable.

quote:

---

YPC/YPA presents a decrease in YPC even if it really is an increase in YPA. If the YPC remain constant and the YPA increase, CptBengal’s quotient (YPC/YPA) will decrease. Put in more concrete terms, if LSU rushes for the same number of yards every season, but increases the YPA every season, CptBengal will get a downward slope in his graph even though the rushing has remained constant and the offense as a whole has improved.

---

Go read my OP again. I SPECIFICALLY REFERENCED YPC:YPA. I made no distinction on YPC, because that wasnt the analysis.

It's funny that you think the analysis is wrong because I didnt use the variable you wanted, then are foolish enough to claim I am "skewing" things by not doing so.

Moreover, I used the ratio because we are looking at the hypothesis of predictability, NOT yards per play. In that binomial outcome there are two options and BOTH need to be represented in a probabilistic framework for the analysis to make sense.

quote:

---

2014 is only four games. Why compare it to the rest of the years that are complete years? It's only four games, less than a third of the total games to be played (12+bowl). You even said that "five data points sucks."

I think everyone can figure out why you included it. It helps CptBengal misrepresent the facts.

---

5 data points sucks because of the overall N you blithering tart. Good god.

I'll tell you what, you give me access to a database and instead I'll do 2014 as a posterior Bayesian analysis....deal?

quote:

---

Second, everyone knows that teams tend to run more and pass less when they are breaking in a new QB.

---

LINK?

aTm doesnt seem to be doing so? I like how you just make up things with no evidence, it's comical.

quote:

---

CptBengal

---

As far as I'm concerned, you win the internets today. Hats off to you for the statistical data and graphs. I honestly think Miles doesn't really give a shite about offensive performance. He can't, because my four year old can pick whether we are going to run or pass correctly 90% of the time when we are on offense.

I'm not sure why Miles changed, but we used to be more explosive, have more complex passing routes, and less predictable on offense. Our offense is a fricking bore to watch and he refuses to change it. We have the athletes to run a more explosive offense, but he wants to slow shite down and make us a Big 10 offense.

I'm maddened by it all.