Here's an example of where sinificantly lifting the shadows was needed.
We had a beautiful evening here the other day and the moon just rose beautifully inbetween the trees and our neighbours house.
I exposed for the moon and at first blush the entire image, except for the moon, was pitch black.
However with +3.5 EV, +100 shadows and a significant lift of the lower part of the curve i reproduced how I experienced the scene.
So there are practical circumstances that exposure protecting the highlights gives significantly underexposed shadows.
Two remarks:
-1- The shadows show mild banding and noise, but I really don't care, it's not the greatest shot anyway and it preserves the memory I had from this beautiful evening in a very acceptable manner
-2- If I wanted a technical better foreground I should have gotten out my tripod and made two exposures (one for the moon, the other for the sky and foreground) and blended them into one picture later. But I was too busy enjoying the beautiful moonrise that I didn't care to do that. Sometimes I find it better to take it all in rather than miss the scene because I'm bogged down in chasing a better technical end-result.
So my bottom line point is that having some data on banding and noise when lifting the shadows can be useful to know, ETTR of very high DR scenes can give very underexposed shadows, but for me it's not an issue to choose between camera A and B.