I never said anything about wanting to know points per play. I was asking for yards per play.
Sorry, that's
yards per play. You requested that I post yards per play, and I told you that it doesn't correlate to wins. Neither does TOP for that matter, and you already know why. Because there are two ways a defense can get off the field fast -- force 3-and-outs, or give up big plays.
Take that same wariness of a stat and apply it to yards per play. There are two ways a defense can allow a lot of yards per play -- give up one big play, or give up several smaller ones. Go back and
really read my example in post #323 if you don't get this. If you understand why TOP isn't a great stat, you're capable of understanding the same thing about yards per play.
Saying that a stat doesn't correlate to wins is just another way to say it's a bad stat. TOP per drive is worth looking at if you want to know which unit (offense or defense) is more responsible for the TOP for a game. That's what I was doing. But TOP per game can be a misleading stat, just as yards per play can be, for the reasons described above.
The reason they got themselves off the field was because they gave up huge chunks of yards.
We know that's not the main reason, otherwise points allowed per drive would have gone up, not down. The 2014 defense allowed fewer points per drive (ranked 16th) than it did from 2011-13 (ranked 25th). How do you explain this? Think about it, and see if you can.
Look at this list of defenses that allowed plays of 20+ yards in 2014. You don't judge a defense solely on how many big plays it allows.
Big Plays per drive, 2014.
Arizona .41
New England .36
Baltimore .34
Dallas .33
Was Dallas' defense better than those teams'? Of course not.
And the only reason why the defense wasn't worse than it's below averageness was because the offense ran the ball so well and controlled the TOP.
That's the point of looking at "per drive" numbers. You can separate the offense's portion of TOP from the defense's. And doing this, we know that the defense faced average drives that were of a shorter duration than the previous three seasons.