This thread is made as a response of various challenges made to Paladin potential viability as a raid tank in their current state that I have seen and tried to reply to on these forums.
The content below began as a discussion of the comparitive benefits of Avoidance stats verses Stamina after uncrushability has been reached, and the first section relating to that is now newly contained in my guide thread, found here:
WoW Forums -> The Light Works in Mysterious Ways
The discussion of the value of stamina directly paraphrases some of the writings by Ciderhelm in his guides "Hold the Line" and "Fortifications".
I will quote that section directly, and feel free to skip it to get to the meat of my observations about our current direction, but I do often reference concepts that I discuss in that part, which is why I am quoting it here.
Oh, and skip the tldr. we know.
thanks.
"Judge a man by the dents in his shield, not the shine of his sword."
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~~
For cutting edge content, the primary obstacle is that bosses hit hard. Really hard. When an encounter pushes the limits of the capabilities of the raid, DPS needs to go as rapidly, as absolutely overburn as possible to overcome the fight pushing it to the limits of its endurance.
Threat-generation is the most important quality a tank can have to the point where it has reached the threshold of the DPS-dealing’s threat at its maximum output.
Second is the ability to take hits and stay alive despite them. Yes, avoidance and mitigation assist in this, but first and foremost is the attribute of a large enough health pool to absorb a hit. Health is what makes you not get one-shotted. Avoidance serves more secondarily as factors which prolong the longevity of the healers’ ability to heal you to full again. They represent a “chance not to die”.
~~~~~~~~~~~~
Think about it this way: incoming damage represents not a static number, but a percentage of your total health pool. The threat posed by an enemy’s attacks is determined by the extent to which they can decrease your health pool. Stats change the effective percent of your total the same damage would remove. Whether you tank something naked or in full tier 6, the damage a boss causes is a constant, but not the proportional effect it has on you.
This isn’t quite the same thing as comparing through saying “with this stat I take X damage and without it I take Y damage”, an outlook which would clearly favor avoidance-stacking but which neglects the whole picture, and that difference is pivotal.
~~~~~~~~~~~~
We think along these lines instinctively when, confronted by a fury warrior and a mage, we attack the mage first since we realize that the mage has a smaller health pool, meaning that our attacks, which don’t change in power, are more dangerous to the mage’s survival than to the warrior’s. Attacking the warrior first and then the mage requires exactly the same amount of total damage to kill them both as the other way round, but we instinctively attack the mage first because the same amount of damage is more likely to kill the mage in a given length of time than the warrior. Also, we know that although the fury warrior might deal larger numbers against a naked AFK target than the mage, the “casting….casting…casting….BOOM” damage of the mage is more dangerous to us than the far steadier flurry of auto-attacks and heroics dealt by the fury warrior. It is a spike.
The core of the point is actually manifested on the healer’s perspective.
Imagine for a moment that you are a healer. Your regular heals are a two and a half second cast; unfortunately, on new encounters, the boss may very well kill your under-geared Main Tank in that period of time. So you set up a healing rotation: you will begin your heal, then another healer will begin a heal, then another -- maybe not so explicitly, but to the effect that heals are landing every second or less on the Lead Tank.
As a healer, you don't have a crystal ball. You know that not healing your tank could result in a raid wipe. Perhaps halfway through that heal, you notice the tank's health bar doesn't seem to be dropping -- he has not taken damage by good fortune of his avoidance statistics. However, you still cannot cancel that heal, because the next split second could very well see his life bar fall drastically!
The 2.5 second heal is the root of the Stamina gearing theory. In almost every new boss encounter, before healers really get a chance to work out what's going on and the ebb and flow of a battle, they are going to be slamming you with heals... Whether you avoid the damage or not has literally no effect on whether you receive heals or not -- you are still sapping the healers’ mana bars.
Avoidance only saves you in single-mob encounters when your life is critically low and you have to rely on the chance of dodges and parries to survive the encounter.
~~~~~~~~~~
The reality is, whether you simply had taken the hit and taken a percentage of your life bar in damage before immediately receiving a full heal is no different than dodging the same attack then taking a 4k over-heal.
Whether you where hit then healed or avoid an attack makes no difference in terms of either your chances of survival or the healers’ ability to hold back and conserve.
~~~~~~~~~~
Avoidance is relying on good fortune in new encounters, where stamina is a steady, reliable base.
The goal of a tank is to buy their healers reaction time. It should be to smooth the flow of incoming damage, making it more recoverable and predictable.
Smoothing the pattern damage in-flow is what keeps a tank alive, and all other considerations including reducing the total amount of healing you required over an encounter are secondary.
This is done much more effectively by increasing your total health (and thereby decreasing the percentage of that total that a single un-parried/dodged hit represents) than to increase avoidance stats which spread damage out, but make a similar single unparried/dodged hit much more dangerous to your survival than otherwise.
It is impossible to reach a point of 100% avoidance. Because of this, avoidance, which by simply possessing it (i.e. using the itemization budget of your gear) is a sacrifice of potential health, is at best a damage spreader.
While taking fewer hits in a less frequent average pattern may sound appealing, look at it again from a healing perspective.
The hits taken will be random, and the healers cannot predict when they land. Furthermore, the lower stamina totals of an avoidance-stacker mean that there is less time to recover from one when it does arrive. Healing will be forced to continue landing on you, in a chain of over-healing, since the risk of dying to spike damage is unacceptable. Also, there is always the possibility of the law of averages failing you, and getting a string of un-avoided hits no matter how high your avoidance was stacked.
~~~~~~~~
This is something the warrior community was brought to realize before now, but it holds another special element for paladins.
Namely, we do not gain mana back from over-healing come the patch.
~~~~~~~~
This means an avoidance-stacking paladin is a crippling handicap for his threat-generation. The avoided attack preceding an overheal or the hit connecting and immediate being topped-off no longer are even equal in results, the topping-off is substantially better. It represents a quantifiable increase in threat-generation potential.
It is a misconception to compare the situations and conclude that the avoidance tank requires less healing.
As a warrior tank, and as a healer too, I unequivocally assure you that the heals still land. Cutting edge content is not forgiving, and no other practice would be tolerated by the extremity of the encounter.
The question becomes merely “how am I making best effective use of that constant healing and making it so that a steady stream is sufficient without there being major emergencies and spikes?”
While avoidance stats seem very appealing on the surface due to taking 0% damage, they are actually much more harmful to a tank to stack than the alternatives.
~~~~~
Avoidance is important because it allows a state where crushing blows and critical hits have been eliminated, which is vital for a tank’s job of stopping spikes in incoming damage, but it becomes useless from a tank-viability perspective once the point of uncrushability has been reached.
Crushes and Critical hits are bad under the very theory I outlined above, since they increase the percent of your total health that is removed at once; an upward-spiking deviation from the steady incoming flow. To negate their possibility is a very useful thing, and overcomes the disadvantages of stacking the avoidance itemization which results in damage stuttering.
~~~~~
This is the reason why Druid tanks, while viable for their massive stamina and armor advantages, also have the disadvantage of this spike damage being near impossible to avoid, making their tanking strategy to decrease the percent of their health pool a spike represents, making it less dangerous, as oppose to eliminating the spikes, which is the approach used by warriors and Paladins.
I believe stamina is the key to guild and raid progression. I feel our healers can recover from regular, sustained damage on a high health bar and high AC rating more than they can handle massive damage from the blows I didn't quite parry on a lower health bar.
(end of guide extract)
(And now we come to the main part)
It is significant to note that the cheapest itemization for eliminating Crushing Blows (i.e. the spikes), is NOT, in fact, Avoidance. The stat is Block Chance.
By exploring the properties of Block Chance, it becomes clear that this supports the damage smoothing concept completely. Blocking reduces a blocked hit by a static amount, determined by your Block Value. This clearly makes for reduced damage across the board for ALL hits (assuming uncrushability) by a small amount, as opposed to avoidance which decreases SOME hits by 100%.
Block further lends to a damage-smoothing strategy since it frees up item value for more stam, compared to the available item value budget remaining after achieving an equal uncrushability through avoidance.
In a hypothetical gearing comparison, if a Paladin achieved uncrushability entirely through Block Rating added to the mitigation gained from the minimum Defense needed to be uncrittable, they would have the item extra value budget to gain over 3.3k (3.5k with talents) more health through stamina itemization than a Paladin who did the same thing using Dodge instead of Block Rating.
And dodge is the next least expensive mitigation stat which helps achieve uncrushability after Block.
It is often stated on these forums that Paladin Tanks have a problem with gaining high Stamina levels.
The reason generally given is the gap of crush-pushing stats needed caused by the difference between the block chance provided by Holy Shield for Paladins and Shield Block for warriors.
I’d like to point out a few implications for this.
While it is true that a fast-attacking NPC has the ability to burn through the charges of Shield Block before it becomes available again, making the greater number of charges and resulting higher viability of Holy Shield for smoothing damage be a large advantage, this is not something that can be easily or commonly implemented into boss encounters.
Here’s why;
A burnt-through Shield Block opens the warrior to receiving Crushing Blows, critical hits being reduced to zero through Defense fairly early. Crushing Blows deal 150% of what a normal hit would cause in damage, i.e. a spike.
However, if a boss is given a significantly faster attack speed, enough to make this situation realistic, the damage dealt by an individual hit must be reduced to counter-balance. If high attack speed is paired with similarly high damage of individual hits, the encounter is no longer realistically survivable. Attack speed and attack damage must be balanced so that damage dealt over time is constant within survivable levels when either variable is changed.
If, then, a fast attack speed means a low damage for individual attacks, we run into the problem of Crushing Blows (and Critical Hits) being a multiplicative effect on an already small value. A high number multiplied by 150% adds a much, much larger numerical amount of damage than a low number multiplied by 150%.
We reach a situation where, by modifying boss statistics so that spikes start occurring on the warriors, the actual spiking has a much more narrow range and a much less dangerous variation from the base-line flow of hits.
We covered before how what is threatening to a tank’s life is wide fluctuations in incoming damage, rather than the static levels of a smoothed-out stream.
In end-game, if smoothed out damage flow is killing your tank, it simply means that your raid is under geared for the encounter.
So in the end, rapid-attacking bosses are not actually all the dangerous, because they must be balanced to deal very smooth and constant damage out-put even before any mitigation kicks in and they just are hitting a paper doll target.
By increasing attack rate to overcome the leveling effect of Shield Block eliminating crushing blow spikes, you make it so that those very spikes themselves are no longer a serious danger, while also forcing the warrior into the exact same tanking strategy as the bear druid.
Either scenario, the warrior tank is preferable to the paladin, due to higher health pool levels.
Designers of encounters know that the event most dangerous to a tank’s survival, and therefore the ability of a raid to beat a boss, is spike damage, and spike damage is much more achievable through slow attacks which deal very high individual numbers and have threat of instant death.
It just works out that this kind of design is the exact kind paladin tanking is weakest against. It also diminishes the Paladin tank in the views of others because the encounters we DO have an advantage in are already extremely “easy” due to the steady damage rates that are the natural characteristic of them.
I do not believe that this fact is a death sentence for Paladin main tanks though, and neither do I think that any change is necessary to the mechanics of the Paladin class in order to make it viable.
We simply need to make like a bear.
The trap I see many of us falling into is the self-deception of thinking that the warrior tanking outlook is the only one that could ever work. The natural result of basing an analysis on this concept is that Paladins are not Warriors, and as long as they remain distinct from warriors they will be lesser tanks; we must become a warrior with different names for our abilities.
But we forget the bear.
No one contests the viability of a feral druidic grizzly. It makes sense due to the fact that it possesses strengths that are different from those of the warrior tank, and while accomplishing essentially an equal extent of filtering damage-flow into survivable streams, the druids work from one end of the scale, while warriors work from the opposite one. They meet in the middle, and both are viable tanks which are very different in play style and gearing priorities and give advantages and disadvantages to various possible encounters or raid make-ups.
We, as a community, must recognize an innate mechanic of our class which allows us to approach the problem of tanking from our own unique direction, and roll with it.
That is the secret to equal status.
That mechanic is Block.
The very nature that Blocking mechanics suggests to me a unique avenue for our Paladin class. It reduces damage by a static amount regardless of the actual per-hit damage, and we are funneled by the path of least resistance for crushing blow elimination to stack A LOT of it. At least we should be if we’re being thrifty with our itemization.
Now, Block is often dismissed as useless for mitigation purposes because it pales in the face of large individual hits. With a block value of 200, the percent of a 10,000 damage hit I am taking that is reduced by nature of blocking it shrinks to insignificance.
Furthermore, the situations where Holy Shield shines over Shield Block, namely with fast attacking bosses, make much greater use of the benefit of Blocking for mitigation since individual hits are much lower. 200 reduced from 1000 is a much more impressive figure. But as shown above, bosses must have spike damage that is not just present but possessing of a wide damage range in order to be threatening to a tank’s survival, something that encounter designers want to achieve with the mechanics they introduce.
So the implication is, there needs to be a way to give bosses a mechanic to cause burst damage with a spiky, unpredictable, irregular pattern, but that a Paladin would have a natural advantage in their ability to survive and smooth over the other options of druid and warrior.
The answer occurred to me one day recently while I was at work.
In a massive twist of irony, it came from the warriors themselves.
Flurry.
Flurry mechanics are essentially a triggered effect which increases attack speed by a large amount for a very short period. Warriors have a talent in the Fury tree which grants a flurry effect when we get a critical hit on our target. Shamans also have a similar talent in the Enhancement tree.
The important element of Flurry is that it doesn’t last long, and the power of your attacks while under it’s effect suffers no reduction, like what would be necessary with a pervasive attack rate buff for a boss’s baseline hits.
It is, at the very core, Spike Damage.
Imagine a boss with a triggered Flurry ability. A Paladin would suddenly have an advantage over a warrior in tanking the boss both in the case of a paper-doll, baseline abilities, no-gear comparison, as well as fully geared at an equal level with the implications of the trends warrior and paladin gearing takes.
Flurries would represent a source of spike damage that would still work perfectly well with a druid tanking strategy of absorbing and rolling with spikes, but would not be as successfully mitigated according to the warrior strategy.
The Paladin angle would become damage reduction at a flat rate, regardless of the size of hits. A huge amount of Paladin tanking abilities lend themselves already directly to this approach, and the old functionality of Reckoning even mirrors the kind of mechanic we are directly suited to excel against.
At a baseline level, when facing a flurry-capable boss, a warrior tank would again find themselves facing attacks with the potential to burn through their Shield Block occasionally. The difference is that now a flurry proc which hits at the tail end charges of Shield Block would finish off the charges and continue doing a veritable Reckoning Bomb of auto-attacks in a short length of time, which are hitting for normal damage, causing a spike which is large enough to actually cause a warrior who is on the receiving end some concern for their life, which was impossibly to achieve with mere attack rate increases.
The warrior would not be invalidated though, due to a fairly rarely recognized element of warrior tanking. Warriors receive the best scaling threat generation from Shield Slam. Their second best innate-threat ability, Revenge, also requires a preceding block. Warrior have the lowest threat-generation potential of all three classes, and the most easily outpaced by the threat of the DPS or Healers. Recall the very first paragraph of this thread in relation of threat. With this consideration, Warriors stack as much available itemization in Block Value as possible, because the damage/threat of Shield Slam, their best ability, scales with the value of this stat.
Warriors will almost always have a higher Block Value total than Paladins, not because of mitigation concerns, but because their threat generation is based on it to a great extent.
Paladins, in comparison, would have less Block Value, but a much higher Block Rating, which was necessary to achieve an uncrushable state (ironic again how the very weakness we hate about Holy Shield is the source of this advantage), and a blocking ability with much higher number of charges innate as well as available through talents.
There is also the threat effectiveness distinction.
Warriors must make the conscious choice during battle of whether to spend limited amounts of rage on mitigation abilities or on threat abilities. A warrior spamming Demo shout, Commanding Shout, and Shield Block will soon lose aggro. A warrior spending every single scrap of rage on threat-generation will die to spike crushing blows, not to mention find themselves still having aggro difficulties through inability to Revenge.
Not so with Paladins.
Paladins generate a significant amount of their threat with the exact same ability which provides a significant portion of their mitigation.
Another ability, Reckoning, effectively allows the same tool to be used to both deliver huge amounts of threat by using a damage seal, or also used to provide a significant amount of survivability by procing JoL at the same time as using JoL using double-hits which are not factored into the proc-per-minute chance calculation.
Blessing of Sanctuary (while it could seriously use being applied after damage reduction is calculated) deals threat-producing Holy damage on every blocked attack including ones made without Holy Shield active at the same time as providing damage reduction of a flat amount, which is hauntingly familiar to blocking mechanics, and similarly dismissed as useless against large single hits for the same reasons.
Our threat-generation directly scales upwards with higher numbers of blocks both directly through Holy Shield, Retribution Aura, or Blessing of Sanctuary, and also indirectly through proc’ing abilities like Reckoning which allows potential for more threat.
So warriors would compare to paladins in that when tanking a Flurrying boss, they would have more powerful damage reduction for the baseline attacks and opening hit or two of a Flurry before SB charges run out, but would take far more pronounced spike damage than a Paladin would under the same circumstances.
This creates a very healthy strategic choice based on Paladins being more effective at damage smoothing consistently and eliminating spike damage, while warriors would provide the strategy of higher levels of mitigation over-all granting more potential for healer efficiency in an endurance fight, the ability to counter particularly nasty spikes with their myriad of “Oh Sh!t” abilities which Paladins lack, and the approach of using their higher health pools to reduce the significance of the inevitable spikes to a manageable level.
In this way, without any change in the core mechanics or the itemization trends of either class, there is an intelligent selection of options open to a raid providing distinct and unique advantages and disadvantages based on approaching tanking the encounter from a different direction, with all being equally viable. An intelligent choice such as that resulting from the current balance between druid and warrior.
To recap the previous section, the concept of a flurry mechanic for bosses, which would either be encounter specific like a cleave or fire breath, or a base combat table event kicking in if crushing blows are eliminated, would be beneficial to tankadin viability because:
-The lesser base block provided by Holy Shield becomes an advantage rather than a disadvantage due to the itemization distribution difference it naturally creates between warriors and paladins.
-Additionally, the unevenness of stamina levels between the two classes no longer is a crippling factor, but instead an alternate path by which the warrior compensates for his lesser suitability to damage-flow smoothing by affecting the relative effect of spikes on his HP, as opposed to affecting the damage intake its self, like the Paladin.
-Paladins would not require equal stamina levels to Warriors nor any gear redesigning, and having the least stamina would be fine since they would be operating on an approach to tanking on the opposite end of the spectrum to a stamina-oriented school of thought such as that outlined at the very beginning of the thread.
-The static damage reductions provided by Blessing of Sanctuary, Block Value and the greater portion of Block Chance Paladins would acquire, and even Blessing of Sacrifice, which are comparatively of little use against slow but powerfully-hitting enemies, now have a useful purpose and contribute a significant amount to a Paladin without changing the abilities, making them over-powered, or providing an equal benefit to Warriors which would make them irrelevant to paladin tanking viability.
-The above abilities, as well as the historical functionality of Reckoning, foreshadow this path for tankadins.
-The paladin gains a unique philosophy behind their tanking style which is very different from that druids operate on as well as the other one operated on by warriors. It is as viable as the other two, without overwhelming them and making either of the other tanks an outclassed underdog.
-Boss encounters now have something we’ve never seen before to play with and throw at us, allowing new possibilities with encounter design and wonderful things we’ve never had to deal with before in future raids.
I believe that the class developers in Blizzard have a fairly decent sense of what they are doing.
I also am convinced by my own observations that they really do read what we write in here.
It is clear to me that the obvious solution to Paladin not viability, but Value as raid tanks is through taking advantage of the innate, and so far dusty and unused, strengths of the basic way that Paladins work by use of a Flurry mechanic to achieve boss tank-killing spike damage.
Of course, it’s entirely possible that I’m completely wrong, and perceived a beauty and progression in the underlying mechanics that is completely accidental and unplanned, and Blizzard really does have no idea what to do about Pallies.
But I have more faith in the devs than that, and there’s always a chance that even if they didn’t think of it or plan it, they might read this idea and even use it.
I guess there’s nothing to do but wait and see.
One thing’s for certain. I love my Paladin. I am a tenacious, dedicated player playing the most tenacious, durable class in the game. It would take far more than what faces us now to get me to quit.
Regards.
~Brekke
