So for the first element - I don't know the answer. Indeed, Jeff's chart on the shoes worn at the Trials (linked at the bottom to keep it out of the way in formatting) says only 4 athletes - 3 women, 1 man - wore the MetaRacer. And of those, you can tell from the photos that there are a couple different stages of prototypes (though some were confirmed to be in the production version). So - could be personal preference, could be ASICS targeting this as a shorter-than-marathon racer, could be production issues in time for the Trials, or could be something else.
Regarding the plate - it has a scoop, it's not planar. We had a call with some of the ASICS designers and they showed us the production model, but we weren't allowed to use the photos in the review, but I can confirm there is some 3D depth to it. That said - the scoop is not particularly pronounced. This might be just owing to fitting it into the limited stack height, due to optimization, or due to IP compliance or concern with what Nike has produced. I didn't ask, and I suspect they couldn't tell me. All I can relay here is that the place is curved, to some degree.
Finally, regarding the midsole - I think ultimately you're right in that a higher stack, ultra-cushioned shoe (like the AlphaFly) is more the direction the marathon market seems to be heading. That said, we've seen strong performances from Adidas runners in their non-stacked shoe (Houston leaked prototypes notwithstanding), and I don't think it's crazy to say that a lighter, less-stacked shoe can't compete with a heavier, well-stacked shoe. There has to be some middle ground.
I do hope to get out there and get a 20 miler in this shoe in the near future, in which case I'll (at the very least) update this thread. I did 18 in the Brooks Hyperion Elite (a high-stack, but very firm) racer, and came away impressed. Lower fatigue than expected. Compared to do 18 on asphalt in something like the Nike Streak LT, there's no comparison. The cushion helps. But like I said, I assume there will be a middle ground of the weight vs. stack equation. ASICS may have found it, they may not have.
Here's the RW chart:
https://www.runnersworld.com/gear/a31180532/olympic-marathon-trials-shoe-count/