I find designers often assume things which are trendy/etc do not actually have a factual basis in what works better, nor any sort of references to back it up.

I admit they often make prettier things (and I'm talking UX designers), but they often do not deign to come down to the level of us mortals who have to implement their wireframes with cost considerations or the like. It feels like they're unwilling to put constraints on their process, so it can make it largely painful rather than helpful.

It feels like the architecture astronauts you get some times in corporate walla walla land in software. Some people have grandiose (and costly visions) and do not feel they can be questioned on those; refusing to accept questions makes everyone else think they're full of crap, because it's usually the guy who doesn't know what he's talking about who starts going off when challenged.

Perhaps you're doing the same sorts of things? You're regurgitating popular layouts, etc, and not really making decisions which can be explained well nor taking into account an economy of development resources.

It's great working with designers under you. But as equals, it's very taxing to work with people who won't articulate reasoning. Questions which may figure that out:

Can you explain how much time different things take to do?

How do you know this?

Do your developers agree?

How much of your dev's portfolios have you gone over (and therefore know what they can do easily/cheaply)?