World of Warcraft

We understand that raw patch notes don’t always provide enough context for changes to the game, so we thought we’d take a moment to explain some of our logic behind the buffs and nerfs coming in the 4.2 “Rage of the Firelands” patch. We particularly want to try to combat the perception that classes end up nerfed in PvE as collateral damage from PvP nerfs (or vice versa). The number of PvP buffs versus PvP nerfs were about equal this time around, and we make careful consideration to keep the impact outside of PvP minimal. So, if we nerfed your class’s PvE damage, it’s most likely because we thought your class was doing too much damage in PvE.

No doubt you’ll disagree with some of our logic – “But we were middle of the pack!” seems to be the common rebuttal these days. Balance is an art, though, not a science. (This topic probably deserves its own blog, but I’ll try to spell it out briefly.)

Balancing a Battleship

Player feedback certainly plays a role in our deciding to make balance changes, but it’s just one small part. (And remember that our public forums are not the only place where players voice their opinions.) We also do a lot of internal testing, both simulations and actual character performance, and we collect a lot of external data, which needs to include players at a wide variety of skill levels.

If a class is performing poorly except in the hands of one percent of the population, that’s a problem, but it’s not an excuse for that class to be overpowered when played by that one percent, especially in PvP. In PvP, your group composition (whether your whole Arena team or just the participants in a given fight in a Battleground) matters enormously. In PvE, the encounter specifics matter enormously (and we’re about to get a new cast of characters in Firelands). As a result, it can be challenging to decide which numbers are the right numbers. It doesn’t make sense to balance PvE damage solely around Patchwerk-style target dummies and assume everything else is a gimmick. Likewise it doesn’t make sense to buff and nerf for each individual encounter. (Say a boss buffs casters – do you then nerf casters as a consequence?)

Also realize that we don’t believe in the existence of any single thermometer for estimating character performance. By that I mean there is no single relatively-accurate measurement of real and true DPS (or tanking, or healing) upon which most players agree. Everything from simulations to target dummy DPS to Arena comp percentages to the top 100 raid parses all count as pieces of the puzzle, and you really have to consider the totality of all of that information in the proper context.

Resto druids “win” healing meters now in part because their raid cooldown, Tranquility, shows up as healing. Warriors do well on Chimaeron because they take a lot of raid damage. Resto shaman heal well on Chimaeron because everyone is wounded all the time.) Yet once you start eliminating data – “oh that’s a gimmick fight” or “oh, someone is inflating their meters by attacking an irrelevant target” – you risk skewing the results.

Statistics is a complex business. That doesn’t mean all of this stuff is inherently unknowable and nobody should bother. It just means you have to be careful; the discussions about these topics are never short and simple. It's wise to be skeptical when anyone attempts to boil down conclusions on class balance to very simple declarations.

Simple Declarations

Having said that, we have a lot of ground to cover below, so some of these notes are going to be very terse by necessity. My apologies in advance.

General

• We changed the way interrupts interact with spell schools. Our intent when we created dual school spells (like Mind Spike being Frost and Shadow) was to allow players locked out of one school to still have something to cast, and we now have the technology to deliver on that design. However, we maintained the rule that being interrupted while casting a dual-school spell will lock you out of both schools because we didn’t want players to only use those spells as a way of avoiding interruption.
• We changed all heals to have 200% crits because we wanted to make crit a more attractive stat to healers. Anything random is already at a disadvantage when viewed by a healer, and one point of haste just flat out increased throughput more than a point of crit.
• We removed the threat caused by buffs or crowd control because we wanted to make communication and coordination easier in dungeons, especially among strangers using Dungeon Finder. We want the challenge of a dungeon encounter to be the encounter mechanics, not marking targets. We also think this change will be a quality of life improvement for tanks, who inherited a lot of the responsibility for explaining fights, marking targets and otherwise setting the pace.
• We changed the values of Agility for plate-wearers and Strength for leather-wearers to reinforce which type of armor you should be using and so we wouldn’t have to spend diminishing returns trying to balance tanks wearing non-traditional armor sets.

Death Knights

• We added the cast time to Hungering Cold for PvP reasons. It is one of the most powerful forms of crowd control in the game, especially in Battlegrounds, and yet was impossible to prevent.
• The nerfs to Obliterate and Howling Blast were made because Frost damage was too high in both PvP and PvE. Note that these values were hotfixed – you shouldn’t see damage drop further when 4.2 goes live.
• The Glyph of Dark Succor change was to keep Death Strike from providing so much healing in PvP.
• The Might of the Frozen Wastes change was a small tweak to help keep one-handed Frost relatively competitive with two-handed style.
• The Unholy Might buff was to help catch Unholy up to Frost in PvE. Interestingly, we didn’t nerf Unholy damage at all in 4.1, but you can still see a small drop in their DPS because so many talented DKs went Frost. I’d love to have the discussion some time about how close two similar specs need to be before players will play the one that is most fun for them and not the one that does theoretical higher damage. Is it 5%? 1%? 0%?

Druids

• We boosted Feral damage to compensate for their losing the attack power from Strength. Net DPS shouldn’t change much overall, though burst may be slightly higher. (We didn’t want to buff bleeds since that was a problem before in PvP.)
• We cut back on the power of Innervates from Feral and Balance druids because we felt they were contributing to too much healer mana.
• We changed several Balance druid mechanics to cut down on the damage they could do while moving in both PvP and PvE and to cut back on some of their strength in multi-dot fights in PvE. Furthermore, we felt like druids were spending too much time at one end or the other of the Eclipse bar by using dots rather than moving the bar back and forth as intended.
• We toned down bear damage, because they were going to do more DPS than other tanks while tanking. Other changes were made to keep bears from neglecting certain core abilities.
• We redesigned Restoration’s mastery because it was devalued in situations where druids did a lot of raid healing by HoT-ing different targets, especially in 25-player raids.

Hunter

• Multi-Shot was doing too much damage in PvE given how simple it was to use.
• Careful Aim allowed Marskman hunters to do too much damage in PvE raiding, where the 80% health phase could last for a long stretch of time.
• However, the Careful Aim nerf also affected Survival, whose DPS was fine or even a little low, so we buffed Black Arrow to compensate.

Mage

• We nerfed Arcane Blast because Arcane’s damage was too high in PvE. We wanted Arcane to be competitive with Fire, especially given that Fire tends to perform better on fights with movement or multiple targets. However, it looked like many Fire mages were begrudgingly respeccing to Arcane, which wasn’t the intent. We wanted Arcane to be competitive, not the only serious mage spec for PvE. (See Frost vs. Unholy note above.)
• We originally tried nerfing Spellsteal’s cooldown, but that made it feel really random (for both sides) since the mage had no control over which spell was stolen. We instead nerfed the mana cost to encourage tactical use of Spellsteal and discourage spamming. We’d still like to try a model where dispels have a long cooldown but remove everything, but that is too big a change for now.
• We added the diminishing returns to Deep Freeze and Ring of Frost (after earlier trying some different nerfs) to tone down Frost mage control, especially in the mid and lower tiers of PvP when dispels can’t be assured.

Paladin

• We concluded PvE Holy paladins were ending fights with too much mana relative to other healers, so we raised the mana cost of their single-target heals.
• We tweaked Holy Light because we wanted it to compete a little more with Divine Light, especially when used with Beacon of Light.
• We buffed Word of Glory for three reasons: We felt Holy Power was mattering less to Holy paladins than it did at Cataclysm launch. We wanted to provide more uninterruptible healing in PvP. We knew Light of Dawn was trumping Word of Glory in almost all cases in 25-player raids.
• We changed the Denounce mechanic to give Holy paladins slightly more offensive utility in PvP. We felt that ignoring the other healing classes came at some risk which was not the case for paladins.
• We buffed Holy Radiance both to help PvE paladins feel like they could make larger contributions to raid healing (especially in light of the single-target nerfs) but also as part of a significant buff to Speed of Light to let Holy paladins have more mobility in PvP.
• We made a tweak to Holy’s mastery to allow its bubble to stack, so it would be wasted less often when healing a single target.
• We made Selfless Healer, Divine Protection, and Beacon of Light no longer dispellable so that paladins didn’t lose so much of their survivability in PvP to dispels. (At no point in this patch was Avenging Wrath undispellable – that was a myth.)
• We redesigned Holy Shield, partially because paladin mitigation was going to be too good in the Firelands raid, but also because many paladins (though of course not all) told us they wanted a more dynamic rotation and less passive mitigation.
• We buffed Seal of Righteousness to let Retribution use it for AE fights as intended, and also to buff Ret AE damage overall.
• We made a slight buff to Selfless Healer. While we thought the old model was unbalanced and turned Ret into too much of a healing spec, we heard from a lot of players who liked the utility of being able to help heal somewhat. This talent should provide that trade-off in a more balanced way.

Priest

• Shadow was doing too much DPS in PvE when multiple DoTs was favored, so we nerfed their DoT damage. We want Shadow to benefit from multi-DoTs, but Shadow’s damage was just too high under those conditions. We buffed Shadow cast-time spells to compensate.
• We changed the facing requirement of Psychic Horror as a PvP quality of life change to make it consistent with other non-projectile crowd control spells.

Rogue

• We nerfed Cloak of Shadows because it felt like Subtlety rogues could counter both casters and melee specs too easily. This change forces them to choose between Cloak of Shadows and Combat Readiness.
• However, we wanted to compensate rogues for this PvP nerf, and we concluded rogue damage in PvE was also too low, so we buffed their damage overall.
• The buff to Hemorrhage was designed to make it less punishing when it wasn’t possible to get behind a target, which often arises in PvP but also sometimes in PvE.

Shaman

• We previously nerfed Water Shield via hotfix because shaman were gaining too much mana in PvP when attacked (especially by pets to discourage drinking). The 4.2 change is just a more elegant implementation of the same nerf that should keep the same mana per time as they have currently.
• We recognized Fire Nova had some usability issues so we increased its throughput and added the Flame Shock refresh mechanic to help ease some of the inconvenience. This is a new mechanic and one we are still evaluating.
• We introduced the Glyph of Unleashed Lightning to help shaman feel less punished by movement in both PvP and PvE. The impact of changes like this are very difficult to model.
• We nerfed Mana Tide for the same reason we nerfed Innervate – it was just providing too much mana for the group’s healers as a whole. We didn’t want to decrease the benefit to the shaman, so we redesigned / added the talent of Resurgence to help offset the nerf to them personally.

Warlock

• We nerfed the Glyph of Soul Swap to reduce the ease of applying multiple DoTs in PvP.
• We nerfed Drain Life because Affliction was forsaking Shadowbolt in PvE, which wasn’t intended. We want Drain Life to be for utility, not primarily for damage, and we want all casters to have to hard cast at least some of the time. This was done via hotfix and players won’t see a change in 4.2.

Warrior

• We put Deadly Calm and Recklessness on the same cooldown to reduce warrior burst in PvP.
• We also nerfed Arms and Fury damage across the board because they were doing too much damage in both PvP and PvE. While we are sensitive to casters outperforming melee on several raid encounters, having warriors handily outperform all other melee isn’t the solution to that problem.
• The stance requirement changes on long cooldowns was a quality of life change.
• We didn’t want warriors using Charge as a rotational ability on some fights without actually having to move (which was a bug created as a result of a fix put in to help hunter problems with minimum range).

Raid Nerfs

• As discussed previously, now that players who have spent a lot of time in Throne of the Four Winds, Blackwing Descent, and Bastion of Twilight are moving on to Firelands, we wanted to make sure players who previously couldn’t progress on those raids are now able to experience them. In a way, this provides new content for everyone – if you’re done with the 4.0 raids, you now have 4.2. If you haven’t seen the 4.0 raids yet, now’s your chance.

As I write this, 4.2 has not been completed, so there is still time for us to make additional changes. Constructive feedback is always welcome. Even the most articulate, logical, and passionate argument won’t always force our hand, but it helps.

Greg “Ghostcrawler” Street is the lead systems designer for World of Warcraft. He successfully resisted the urge to bore the other designers with tales of obscure marine fauna for Vashj’ir, though he did suggest the pogonophorans.