I think a few YPC haiku's would also be appropriate.
Does Te'o get his job back?
Collapse
X
-
Originally posted by Yubaking View PostHave you ever been in a science class? When you try to see what happens as a result of changing one variable, you do your best to keep the other variables fixed. Why? Because if you let the other variables change, then your result is a reflection of a combination of factors and not necessarily the result of just the change in the variable you are trying to study.
You and Steve have been guilty of violating this concept with respect to assessing what the effect of having a good run actually is--from day one.
All I have said is that if you make the run defense better, the whole defense overall will be better. That assumes everything else stays the same. If you change other variables, you have just left the scope of my assertion.
I saw all of the playoff losses I cited. Lack of run defense in the second half killed us in those games against NE, PIT and the NYJ. I was at the Jets game. I can guarantee you that we win that game if Greene does not run all over us for a TD and I would have liked our chances even with the Greene run if we had made the final third down stop that we blew because our run defense was not very good.
In your examples, all of the other variables are changed. We were more suspect against the run against CIN than against DEN, but we overcame that because of better time of possession, better net turnovers, better pass defense and better offensive production. That does nothing to undermine any of the points I have made. That doesn't mean we played good run defense against CIN.
Further, YPC against is my go to stat for run defense, not for all defense. It may not be perfect, but over a whole season, it usually does a pretty good job of sorting out the good run defenses from the bad ones. Total yards is crap.
In the spring I plotted a series of graphs, by year, of ypc allowed, vs points allowed. When I ran a linear regression line through it, the line did not describe the data points. The regression equation does indeed put a line through the points (it is an mathematical algorithmn, that means nothing), but the R squared value (goodness of fit) was not very much above zero About 0.05 (where 1 or -1 are perfect fits and zero is completely uncorrelated). If there was a relationship, then there should be some sort of trend. But the line had about as many points above it as below it. Teams with high YPC who gave up a lot of points, teams with low YPC who gave up a lot of points, and teams in the middle who spanned the spectrum. The ONLY reason that there was any association found by the regression equation is that really impossibly bad defenses are hopeless against the run and pass.
Finally, as for Soliai, I said that I thought he could help us but that ATL paid too much for him. You do get that when I heard the price paid by the Falcons for Soliai, I immediately stated that he was not worth that much, right? He was believed by me to be the best true NT out there on the free agent market at that time. Even if Soliai's individual performance has fallen off, a proposition that has not been established by any cited stat in this thread, all that would mean is that Soliai is not as good as he used to be. That does nothing to undercut anything I have said about the importance of having a good NT.
And even if he was the player he once was, how much value would he be? How many plays would he get? 25 a game? He might have helped us some vs AZ, although I doubt it. But the rest of the games, the big runs have come outside and off tackle. How many big runs have we allowed up the middle? He could be the best damn NT in all of football, and he would still be a waste of money if he had prevented us from signing the other FA, like Johnson, Donald Brown, or Flowers.
-
👍 3
- Top
- Bottom
Comment
-
-
More upside down tigers can't be a bad thing.
Originally posted by KNSD View Post:sprite: :sprite: :sprite: :sprite: :sprite: :sprite: :sprite: :sprite: :sprite: :sprite: :sprite: :sprite:Prediction:
Correct: Chargers CI fails miserably.
Fail: Team stays in San Diego until their lease runs out in 2020. (without getting new deal done by then) .
Sig Bet WIN: The Chargers will file for relocation on January 15.
- Top
- Bottom
Comment
-
-
Originally posted by Yubaking View PostI have heard of it, but have no practical experience with it. It has literally been 28 years since I last studied mathematics of any kind (advanced calculus then). If you count statistics, then 26-27 years ago, a long time ago either way.
I have not seen an application of multivariate analysis to what we have been discussing. I am sure it would take considerable study and effort for me to break it down (given my lack of exposure to it) to have an opinion about it if such an analysis were to exist herein.
- Top
- Bottom
Comment
-
-
Originally posted by Steve View Post
In the spring I plotted a series of graphs, by year, of ypc allowed, vs points allowed. When I ran a linear regression line through it, the line did not describe the data points. The regression equation does indeed put a line through the points (it is an mathematical algorithmn, that means nothing), but the R squared value (goodness of fit) was not very much above zero About 0.05 (where 1 or -1 are perfect fits and zero is completely uncorrelated). If there was a relationship, then there should be some sort of trend. But the line had about as many points above it as below it. Teams with high YPC who gave up a lot of points, teams with low YPC who gave up a lot of points, and teams in the middle who spanned the spectrum. The ONLY reason that there was any association found by the regression equation is that really impossibly bad defenses are hopeless against the run and pass.
- Top
- Bottom
Comment
-
-
Originally posted by Yubaking View PostSorry, but the fact that teams chose to run versus pass against ATL in the red zone has nothing to do with defining where ATL stacks up against the run. Teams have been able to pass and run on ATL at will and have gotten the ball in close where they coincidentally tried to run, making ATL the least scored upon team in the league despite having the league's worst pass defense. The fact that the TDs happened via the run instead of the pass does not make the run defense worse or the pass defense better.
The same is true of the first downs generated, which is further misleading when compared to our team's first downs because of the difference in rushing attempts being against us versus ATL.
If you rather I refer to what you posted as "your stat citing non-analysis", then have it your way. I don't think that changes any of the points that I made concerning your stat citing non-analysis.
- Top
- Bottom
Comment
-
Comment