I agree for the most part on your views on what PLDs should be, role-/niche-wise, though I must say I'm not to fond of your stacking system. Part of the appeal of the Knight motif for me has always been that it should have a strong stand-by. Back in ARR, whereas WARs had to spend a combo on buffing their damage, PLDs did not. WAR had one oGCD that they could use during the gather. Our ST oGCD hit like a truck and then we had an AoE oGCD to top it off. Everything about it shouted "ready at arms". The ramp up of stacks, while it works for sort of the Heroic motif of "Paladins" or "Sword Saints" moves away from the more established "Knight" in my mind. In ARR, we were practically Riot Guards. We were durable, responsive, and had ways of strengthening each other's defenses. We were damn durable, but we were never built to be "in the thick of it" for the purposes of our own output, unlike WAR (or now, just as much if not more so, DRK). We were there only because others needed us to be.
Now, in Heavensward, Clemency has since moved us more a bit more towards the Holy Defender motif, while Goring Blade and Royal Authority have moved us more towards that sort of Sword Saint idea as well. Both in a sense have moved us away from the Knight, or, especially, the Riot Guard (which hinged on staying nearly topped off and where Stoneskin was more useful than Cure, and where there was nothing else to spend our combos on but straightforward enmity). It's different, and in some ways I certainly like it more, but it doesn't feel... finished... either. (Though, arguably, of all the tanks, only Warrior really does. DRK remains "close", much like PLD, in my mind.)
- Our reduced SB and RoH enmity values hinged on their belonging to our only combo before, yet now, when given the ubiquitous three combo choices, remain unchanged, such that RA is more sparsely usable than Delirium or Storm's Eye, for instance.
- Shelltron, while appreciated, especially with a tower shield, feels like a wasted opportunity to accomplish something more.
- Divine Veil feels like a toss, and a bit unresponsive at that, especially in comparison to our memories of being able to actually use Cover inventively and importantly.
- In a continually DPS-focused meta, PLD, mostly as MT, falls well behind.
- In dungeon runs, PLD really feels the weight of only having a substantial damaging AoE once every 25 seconds.
- Paladin still feels little identity within its Sword and Shield Oaths. It's nothing huge, but again feels like something is somehow missing, especially now that the job has moved out of its previous supportive and minimalistic motifs.
That said, again, I don't think stacks are the solution. I imagine the solution would come from a combination of general changes to the game and to Paladin's specific abilities. At most, Shield Oath and Sword Oath might be revised to provide something similar, but likely without stacks for fear of feeling too similar to WAR, and quite possibly without using mana as its ramp-up spectrum, for fear of seeming to similar to DRK. I'll throw out my own suggestion once I'm finished with them. I'd hate to critique and provide nothing of my own, after all; I just need a bit of time. (I'll post them in Edit as I think them up. Just keep in mind they might not have other, balancing components yet.)
I'd list PLD-specific solutions I'd think appropriate, but ultimately I think these issues go a lot deeper and broader. Though I've nothing against the traditional, bare-bones way the on the way the game handles enmity, I can't help but wonder if other ways would allow a tank to feel and act more tankily, or, better yet, allow more group tactics to be played with and executed creatively. I wonder at the lack of cleaving damage when every tank animation cuts through every nearby enemy to hit its target. Blocks, dodges, and parries, and their resident RNG-dependence, seem archaic and underutilized. The enmity focus/variance of tank combos is as limiting as it is diversifying. Any excess Vitality has no place in the current meta. Dually skill & spell classes face unique and shortsighted issues due to split Skill and Spell Speed. Magic remains a single type, whether a lightning bolt, an energy fog, a curse, or a literal boulder launched at high speed. Blocking a twig with a tower shield will still allow through 70% of said twig's damage; yet the same block will mitigate a third of a Bahamut abruptly sitting on you. You can dodge a physical attack that slashes literally every area of a room simultaneously, or parry a shockwave from a landing mech, but you cannot dodge a magical wrench.
I feel like it would be so much easier to get a clear picture of what each tank needs if the general mechanics of the game were addressed first...