The benchmarks listed apply nicely when you have a ball-control offense but don't even fit into the philosophy of a large-chunks-of-yard offense, of which there are plenty.

In other words, success or failure in a ball-control offense, whether that is moving the ball through ground or air, is aligned with those metrics. Or, you could turn that around and say that those metrics align with (and might be derived from) the bountiful ball-control offenses in the game.

However, there exist a handful of offenses that are not ball-control but vertical stretch offenses. Success on a given play isn't dependent upon "down & distance" because most plays, if successful, have the potential to break more than 10 yards. The philosophy is to be 'just successful enough' to string together enough almost successful plays (2,3,4 yard gains) to hold a drive together until a big-play occurs. In this philosophy, the metrics you listed don't even make sense. I'm not inclined to argue in support of this philosophy, but it does exist.

I guess this doesn't really add much to your point other than a sidebar to consider.

I'll admit it's been a long time since I looked at the DVOA formula (even though I know it's a staple for the pro analyses at FO).

50% on first down and 70% on second down makes sense for most plays (as does 100% on 3rd & 4th, obviously). But do they still make sense if you're in a first-and-long or second-and-long situation?

If you've got 1st & 15 due to a false start, is getting 7.5 yards enough to set you up for success on subsequent downs? I don't think so. Now you're at 2nd and 8. Since first & 10 is the most common down & distance, and we want 5 yards in that situation, doesn't it follow that our "ideal" second down situation is 5 yards or less? If that's the case, on 1st & 15, should we want 10 yards instead? Just thinking aloud here.

Yes I do.