The new tanking threat paradigm and you

If you're wondering what all the fuss about Ghostcrawler's latest dev watercooler post is about, well, you should probably go read it. Some of these changes have already gone live on the realms, while others won't until the next patch. The basic gist is as follows:

1. Threat generated by tanks has been increased from 300% of damage dealt to 500%. What this means in practice is if your tank is doing 5k DPS, you'd need to do over 25k DPS to pull threat off of him or her. (You need to do roughly 110% of tank threat to pull once he or she has aggro, so you'd actually need to do 27.5k DPS to pull off of a tank doing 5k DPS.) This change was hotfixed in, so if you're noticing your tank is suddenly doing a lot more threat per second, that's why.

2. The way Vengeance stacks is going to be streamlined. Vengeance currently ramps up somewhat slowly. In the current model, every time you take damage as a tank, you gain 5% of the damage you take as attack power. So if you're hit for 20,000 damage, you gain 1,000 attack power. As you take more and more damage, this stacks up to a maximum of 10% of your health, so for a tank with 165,000 health, this caps at 16,500 attack power. In the new version, when a tank takes that 20,000 damage, he or she will gain one-third of the damage of the attack as attack power immediately, or 6,600 AP. This is more than six times as much attack power gained as in the current model. Vengeance will otherwise work the way it does now.

These two things combined by themselves mean that, except in cases where the DPS simply blows all their cooldowns immediately upon seeing the trash coming or as soon as they see the boss while the tank is sitting down to eat, threat will be almost trivial for a tank to gain and maintain. In addition to this revelation (which we are already starting to play with right now, as I experienced in a recent pickup Zul'Gurub instance), Ghostcrawler talks about how tanking will be redesigned to remain active with this new design philosophy.

This is really groundbreaking stuff, and it means that patch 4.3 will see the complete dismantling of the legacy of vanilla WoW tanking design. Once, gaining and keeping threat was the most important role of the tank, more important even that survival, and many endgame tanks were warriors 31/5/15 specced into Defiance in the protection tree to ensure threat. These changes can be seen as driving a final nail into that kind of tanking's coffin.

Getting threat to matter to tanks didn't happen

What's really interesting here is that these changes effectively admit that tanks simply don't choose threat stats on their gear, even when it might have been necessary to hold threat. Undergeared tanks in 5-man dungeons can choose to pick up gear with a mix of threat and avoidance/mitigation, and die, or pick up as much pure avoidance/mitiagtion as possible and reforge what threat stats remain as much as they can, and possibly live. More interesting is the admission that tanking still suffers to some extent from an inherent scaling limitation caused by threat. Vengeance was created to keep tanks scaling as well as DPS when tanks choose avoidance and mitigation while DPSers choose stats that will, of course, increase their DPS.

Vengeance does its job just fine in raids or instances where the tank is comparably geared to his or her DPS. A tier 12 tank can hold threat against her tier 12 raid. The troubles begin when a tank in gear just good enough to get into a Zandalari pickup group gets a DPSer who simply outclasses him or her to degrees that Vengeance simply doesn't currently stack fast enough to overcome. In a situation where a trash pack of four mobs is going to eat a burst of pure DPS coming close to 70K in 4 seconds, there's very little a newcomer tank can do. With these changes, it'll be a lot less likely for highly geared DPSers to rip threat off of a tank, because the tank's 12k DPS will count for 60k threat per second.

If threat doesn't matter, will tanks matter?

With threat design now attempting to eliminate DPS throttling -- or at least greatly minimize it -- allowing undergeared tanks to hold threat in Dungeon Finder groups and removing the debate over threat stats on tanking gear (essentially saying to tanks, "We have heard you -- you don't want hit or expertise on gear"), the challenge becomes one of keeping tanks involved in their ability choices. At present, good tanks using their complete toolkits can hold aggro very effectively, but the fear is that with these changes, poor tanks will seem like good tanks and good tanks won't have anything to do but wait for huge damage spikes to use their cooldowns. The discussion then becomes one of what tanking will be like, and it's fascinating to consider just how far the changes could go for some classes.

Right now, resources generally generate threat. Using a protection warrior as an example, as rage comes in, it's spent to use Shield Slam, Revenge, Devastate, Thunder Clap, Shockwave and Rend, Cleave or Heroic Strike. These abilities all cost rage in most cases (Shield Slam can be free with a Sword and Board proc), while big tanking cooldowns like Shield Wall and Last Stand do not. Shield Block costs rage, but it's a fairly trivial amount; it still means that Shield Block works somewhat like the Death Strike example listed above already. If Shield Block could miss, cost more rage, and was more often active (at present Shield Block is a 1-minute cooldown that can be talented down to 30 seconds with full Shield Mastery), then it could serve in a similar role as an active mitigation ability.

The active mitigation model and its problems

Of course, the problem with the current DK model is that, just as most tanks will sacrifice threat stats to increase survivability, it's very compelling for DK tanks to hoard their resources (runes and runic power) to use Death Strike over and over again. In short, giving tanks the model of managing resources to maximize survival while also making threat effectively trivial means it is very likely for tanks to hoard their resources and only use the survival ability. The biggest threat of a system that combines resource management for survival and a complete loss of threat mechanics is if that system doesn't have significant options that will make tanks have to make decisions. Even now, DK tanks often spam Death Strike every single time the runes allow, and that's not choice any more than the old days of warrior tanks hitting Shield Block every 6 seconds to push Crushing Blows off the table was choice. It's not choice if you never decide not to do it. That design was admittedly not compelling or fun, and neither is spamming Death Strike.

One possibility is to give each tank two or three ways to increase survivability, perhaps keying these options into current offensive moves. At present, a warrior tank has three major debuffs that increase her survivability and/or threat (Demoralizing Shout and Thunder Clap reduce the enemy attack power and attack speed respectively, while Devastate applies Sunder Armor, reducing enemy armor and increasing the party or raid's damage) while also managing his or her threat moves. In order to retain the complexity of the warrior rotation, you could easily imagine extending the system already in place, where offensive moves like Thunder Clap or Devastate also provide survival. Shield Slam, for instance, could provide the warrior with an increased chance to block after it lands, while Revenge could boost parry similarly.

Instead of having one ability like Death Strike that fills every niche, spread the survival out amoung several and make it so the tank can choose which he wants to use for the situation. By using current attacks to provide specific kinds of survival boosts, the tank could choose between using them in order to increase his or her chance to endure specific kinds of damage. Making Spell Reflection provide direct magic damage reduction and a chance to reflect some damage back to a tanked target versus using Shield Slam to increase your block when tanking a fast swinging physical mob would be a way for a good tank to distinguish himself or herself.

This is just one possibility, of course. What we know is that we'll need to watch all four tanks to determine how their redesigns play into this new active mitigation model. Feral druids could become a particular sticking point for a couple of reasons. One, their gear is already DPS leather; they don't have nearly as many options to dump threat stats as the plate tanks do and end up taking a survival hit. Two, they could end up generating ridiculously higher amounts of threat for the same reason. It's not impossible to imagine ways to design around it, but it's just one issue out of a great many that this new approach to tanking design will bring to the fore.

If you're a tank or considering becoming one, patch 4.3 promises to be interesting indeed. Get ready to relearn tanking again.

World of Warcraft: Cataclysm has destroyed Azeroth as we know it; nothing is the same! In WoW Insider's Guide to Cataclysm, you can find out everything you need to know about WoW's third expansion, from leveling up a new goblin or worgen to breaking news and strategies on endgame play.