Thought about this a couple of times, actually! Mostly in the context of removable OP modifiers, as well.
I think there's a solid chance it would skew the gameplay and what's optimal towards OP-heavy builds, wherever the sweet spot in the CR reduction is. And then there are things like Hardened Subsystems, where it gets really weird, since those would almost *give* you OP. +% max CR skills effectively become +OP skills, as well.
Thought about having it increase deployment costs instead, but that gets weird too. For example, D-mods could become something 100% beneficial - e.g. you've got compromised armor, but there'd likely be cases where you could get Heavy Armor by going over OP, and fully negate the d-mod while still having a lower deployment cost.
Basically, it feels like there's a fundamental problem with the idea. If we're trying to increase the pool of points that make the ship better by making the ship worse, there are going to be cases where it's no-brainer, i.e. the thing you can buy with the points outmatches the "worse" part. I think for a lot of ships it would result in a small number of clearly-optimal over-OP loadouts.
Perhaps it could be made to work, though. Some tweaks to Hardened Subsystems and skills and how it works so that nothing can *directly* counter the downside, and a sufficiently high magnitude downside... but it's still a solution looking for a problem, isn't it? If removable OP modifiers were a thing, then sure, it'd make sense to look at how to handle it, but absent that?