In an IM race, I know...

I should target a NP of 68-70% of my FTP
I should target a VI of around 1.05
I should keep MaxP ~ FTP

Intuitively, all of these same concepts of optimal average and maximum power production, and variance, seem like they would apply to running as well.

Specifically, it would stand to reason that a robust running power meter would answer the question of how fast an IM racer should run up a hill, just as a power meter already does for the same athlete on his bike. Since there is an optimal power output when climbing a hill on the bike course, surely there is a corollary for climbing a hill on the run course.

Given that, I don't at all understand why you don't see value in a device that would provide athletes the same data on the run as they use (often to the exclusion of all else) on the bike. Naturally the target ranges would need to be recalibrated for running, but the underlying principles would be the same -- average X watts, don't exceed Y watts. Why don't you see that as an improvement on things like pace, HR, RPE?