An issue that has plagued PvE in WoW for a long time is defense imbalance. It affects both raid and M+, and it hasn’t really been addressed, instead just getting worse. The most obvious example is high M+, where some specs are able to survive much higher key levels than others. Another is the raid death statistics, which shows another aspect of this defensive imbalance: the power of passive defense.
Defining defensive categories:
To start with, I think it’s relevant to split defense into different categories. First there’s different levels of “agency”: Active, Rotational and Passive. Then there’s 2 “timing” categories: Mitigation and Recovery.
In case it’s not clear what goes into which category or how I differentiate them, here’s some examples.
Active:
Anything that you make a decision to use to counter specific mechanics. This can be anything from Barkskin to Divine Shield. Ideally you plan out when to use these, but a lot of people end up using them after already taking damage or just not using them at all. These are the defensives that have the highest skill cap/requirement.
Rotational:
Buttons you press as part of your rotation that provide defensive value. This could be Shadow priest healing/absorbing damage rotationally thanks to Devouring Plague, Vampiric Touch and Mental Fortitude or Trickster rogues in TWW using Feint as part of their rotation. These aren’t quite passive, they do require player input, but they’re also essentially always running.
Passive:
You get full value just by selecting the talent or in some cases just by logging into the game. Here we have things like Sanctified Plates in the paladin class tree or Demonic Embrace in the warlock class tree. The player has no input here beyond (at most) selecting the talent, once you’re in combat it’s just always going.
Mitigation:
Preventing damage from happening in the first place. This includes both damage reduction like Astral Shift as any shaman spec or absorbs like Touch of Karma as a Windwalker monk. This obviously has to happen before the damage comes in for you to get anything out of it. Immunities also fall into this category.
Recovery:
Anything that helps you heal back up again after having taken damage. Here you have things like Alter Time as any mage spec, the leech tertiary stat or Fury warriors healing themselves with Bloodthirst.
How categories play out in practice:
This is a good opportunity to quickly loop back to the aforementioned death statistics in raids, because they highlight something quite important. A general trend in these statistics has always been that specs with 2 particular traits do very well: passive defense and Cheat Death effects.
Cheat Death should be obvious, you just get to make twice as many fatal mistakes as others, which is going to lead to fewer “real” deaths. They don’t require you to predict that damage is coming or actively do anything really, you just get a second chance.
Similarly, being passively tanky gives more room for mistakes as you might just be able to live abilities or combinations of abilities that are lethal to others. Recent examples here could be the Flame Waves (tornadoes) on Smolderon or various overlaps on mythic Fyrakk, particularly in the first phase. A spec with a lot of passive defense like Retribution paladin can get hit by a Flame Wave or a Fyrakk mythic frontal and then survive the unavoidable hit that comes afterwards in the form of Seeking Inferno or Wildfire. Meanwhile something like a Balance druid getting hit by the avoidable damage very easily leads to the unavoidable damage finishing the job.
A big advantage to passive defense is also that it doesn’t “scale” with player skill. A bad player with a passive 50% damage reduction and a good player with a passive 50% damage reduction will end up taking the same damage, while it looks very different if that 50% damage reduction requires you to actively use an ability before getting hit. So even if mages are incredibly durable when played well, weaker players won’t make use of that and have fewer passive defenses to fall back on, which in turn makes mages look rather squishy (until you open a RWF stream and look at Imfiredup seemingly effortlessly surviving everything).
Opportunity cost:
There’s also quite a bit of difference in what you give up in order to become more tanky in combat. We’ve all heard druids get told to “just go Bear Form”, which is obviously not entirely wrong, but it carries a much greater cost (you’re effectively disarming and silencing yourself) than a demon hunter pressing Blur. There’s nothing inherently wrong with there being opportunity cost to defensive abilities, but the actual defensive power needs to actually be proportional. A druid having to go Bear Form just to have as much health as a warlock or hunter gets for existing doesn’t really achieve this goal.
The creep:
The general trend over the past many expansions has been that more and more defensive power gets added to most specs. The problem is that the power level of this defense varies drastically, because there’s no consistent pattern to it. The Dragonflight talent rework is full of filler talents, and a lot of those happen to be seemingly random types and amounts of defense.
You have priests getting 3% magic damage reduction from Spell Warding while druids get 6% damage reduction from everything thanks to Thick Hide. Warlocks get 10% stamina and 5% max health while paladins get 2% AoE reduction, 4% armor, 20% armor, 10% stamina and another 10% AoE reduction. There doesn’t seem to be any rhyme or reason to any of these, they’re just numbers picked out of thin air. It’s not like the different classes are being hit by drastically different mechanics. Of course this isn’t comparing the full kits (that would make paladins look even stronger), but the point is really just: putting defense as filler talents is a problem. Filler talents that increase your damage might be boring, but at least damage is something that gets tuned as an overall package, so it doesn’t really matter where the damage comes from as long as the total number is the same. Defenses don’t get balanced like this, or seemingly at all.
The reworks that happened during Dragonflight have also generally increased defensive capabilities, like adding Sanctified Plates to paladin or Mass Barrier to mages.
Enter hero talents. The new shiny feature for The War Within, and yet again (nearly?) every tree contains some form of defensive power. This didn’t need to happen, the trees could’ve just been 1 or 2 points smaller. And unsurprisingly, the power levels of these defensive nodes are all over the place. The most absurd example here is probably Trickster, in particular for Outlaw, where Feint becomes a rotational button. That’s a 40% AoE reduction (60% AoE and 20% everything with a talent) with 40% uptime. That’s a lot of defense just for doing your regular rotation, and for Outlaw there’s cooldown reduction as well which pushes this uptime higher. Rogue, and again Outlaw in particular, was already not really a class that struggled defensively, so it’s mindboggling to see this addition. Other examples include Archon priest getting 5% stamina, Druid of the Claw powering up Frenzied Regeneration in a variety of ways or Fel-Scarred demon hunters taking 10% less fire damage.
Versatility:
This secondary stat also contributes to the defensive creep and imbalance, and it should be quite obvious how. If your spec happens to like versatility (which shouldn’t happen, that’s a failure in itself), congratulations you are now tankier just by stacking the stat you would be stacking anyway. In theory, and I believe Blizzard has confirmed this a few times themselves, versatility should never be a spec’s ideal stat for damage/healing. That way, you have the choice of trading some throughput for more survivability. The idea here is fine, but that requires the tradeoff to exist for every spec. As soon as some specs pick versatility for damage, that’s gone and should be a priority to resolve. I don’t believe this has actually played out at any point since the stat was introduced, there have always been some specs that had awful scaling with other secondaries so they just stacked versatility.
The stat also has a cute little wrinkle which is that it has increasing returns on the defensive side. Going from 0% to 10% reduces your damage taken by 10%. Going from 10% to 20%? That’s ~11% less damage taken for the same amount of rating. Now, the diminishing returns on secondary stat gains counteract this to some degree, but it still shouldn’t really be a thing.
Damage intake tuning:
Now we get to the core of the issue, which is that it’s impossible to tune the damage output of bosses/mobs when defensive power varies so wildly. If you want to truly threaten the Retribution paladin with extra health, multiple armor and avoidance modifiers, plate armor, an immunity, multiple regular defensives and who stacks versatility for damage, how is the Enhancement shaman supposed to not get their head caved in?
This is why we need defensives to be looked at across the board. Otherwise the realistic result is that some specs simply cap out at much lower key levels than others simply because they can’t survive, and that raid bosses have to be tuned so the frailest specs can make it through, which then makes surviving trivial for the tanky specs (which you would then just prioritize bringing)
Conclusion:
Defensives shouldn’t be completely homogenized obviously, that wouldn’t be interesting. Having unique abilities like Alter Time is great for the game because it allows skilled players to do super cool things. We do need them to be broadly comparable, and for proper use to lead to classes being at least in the same ballpark of survivability. Adding random defensive nodes to talent trees as filler should also really stop. Defensive talents are fine, but they need to exist for a reason and not just add extra defense because “a talent” was needed in that position in the tree.
Being able to modify your existing defenses to adapt to certain situations is good, examples of this are the Planes Traveler/Astral Bulwark choice node in the shaman class tree or Ice Cold for mages. There’s a tradeoff here, you get to feel smart about choosing the right one for the situation. Calming Presence for monks on the other hand? There’s nothing cool here. It just sort of exists and happens to make you tankier.
There also needs to be a bit more parity in the Active vs Passive defense between classes. A skilled player of any class should obviously be better at surviving things than an unskilled player, but the difference between the two varies too much between classes. The obvious examples here are still mage and paladin. Mages are very tanky when played well, but when played poorly they fall over basically immediately. Compare this to paladins, where a lot more of the defensive budget is passive, meaning the gap in survivability between differently skilled players is smaller.