You didn't have to, though. If you could legitimately use it as a mobility tool, it was a damage LOSS to waste that opportunity by saving it for raidbuffs and instead walking back.
On a 200p skill, using your gap-closer under raidbuffs could net you an extra 25 potency, tops. Each GCD for which you were forced to use a ranged filler instead of continuing your optimal damage would cost most tanks over 300 effective-potency.
Removing the potency only simply offers, in effect, one extra charge of mobility (since you would otherwise face damage loss from overcapping it), negligible when already at 2 charges. The primary changes were to APM and cognitive load (no longer being rewarded for knowing whether there was a gap to close within your skill's recast time).
Guess we'd better separate that self-heal/barrier from Brutal Shell, the 4s miti layer and excog each out of Heart of Corundum, the parry chance from the base mitigation of Camouflage, etc.
If we keep having any two effects not perfectly alike (not necessarily sharing exactly the same optimal timings even if separated) together under a single skill to make gameplay more than "monkey see gap, monkey close gap", we'd certainly cause ourselves cognitive fatigue before we can hit our daily PvP fix.
Thinking that could possibly weigh multiple factors together -- what an abomination! ¯\_(ツ)_/¯
/s to be clear.
I don't mind having discrete gap-closers, but they're no more needed than for a %DR and barrier to be separate, or damage and healing, etc.. There's extra control in having those as separate tools, certainly, but having the most streamlined, straight-foward path to every capacity (as to reduce any value for/in componential strategy, etc.) isn't always for the best. And as PvP aims to be button-efficient, it makes all the more sense that flavor sometimes bundles in some extra cognitive load rather than all that is not discrete and plain being pruned.