We have suggested several big changes coming up for the paladin class in patch 3.2. The paladin class Q&A is still in progress, so rather than making you wait, we wanted to give you a preview of what we’re trying to do with all three paladin talent trees. We thought some of the patch notes might not make sense until you understand the design goals.

Caveats: Some of these changes are still in progress and may ultimately end up in a different form or with different numbers. We definitely want to see these changes in action on the PTR and we will iterate on them accordingly based on player testing and feedback. None of this is set in stone. Furthermore, assume that talents, glyphs or set bonuses that affect any impacted abilities will also be changed accordingly. Finally, don’t expect that every class will see this many changes. We are making far fewer changes to class mechanics overall in 3.2, but paladins in particular have some issues we feel are in need of being addressed.

Protection
Currently, we think paladin tanks are almost there and that they just need slightly better cooldowns to handle some of the tougher boss fights. Rather than add a new ability that felt like a clone of another class’s ability, we decided to buff an existing talent that was no longer cutting it. Ardent Defender has two important changes. The first is that the damage can no longer “skip over” the 35% health level – it will always be reduced. Secondly, it has a new effect that if a blow would kill you, it instead sets you to 30% health. This portion of the ability cannot occur more than once every 2 minutes. Think of it as a Last Stand that you don’t have to push.

A second change to Protection is we want to make sure Blessing of Sanctuary is always the tanking blessing of choice. A likely change here is to have it boost stamina as well.

We also recognize that block does not provide the mitigation it once did. While we have long-term plans to change the way block works entirely, in the short term we are doubling the effect of bonus block value on items (so jewelry, but not shields).

Holy
We like that Holy paladins have a niche as single-target healers. The problem is we think this niche is a little too narrow at the moment. Furthermore, paladins don’t have a wide arsenal of healing spells so it’s important that all of them are being used.

First, we are changing Flash of Light so that it places a heal-over-time effect on any target with Sacred Shield on them (the effect will be similar to Sheathe of Light). This should make Flash see a little more use.

Second, and more importantly, we are changing the way Beacon of Light works. Currently, it does not count over-healing on the target. We are changing that. In other words, if you place Holy Light on a rogue who is already at full health, it will still have the full effect on a tank with Beacon of Light on them. This is a huge buff, particularly when you consider the Holy Light glyph which allows for a small amount of “splash” healing. This effect isn’t common when you use Holy Light on a tank because the tank is often standing alone. But if you heal the melee, you are likely to get a lot more total healing from this effect (while still healing the tank through Beacon of Light). We are also going to try to allow a target to have more than one beacon (i.e. from different paladins) on them.

Many players may surmise that this change would make paladins far and away the best healers in the game. This is partially because paladin mana regeneration is so potent. That isn’t an issue when the paladin can generally only heal the tank. With the Beacon of Light change, the paladin can provide a lot more raid or party healing. To adjust for this, we are going to reduce the amount of mana returned by Illumination. If you use Holy Light too recklessly (such as on targets who don’t require that much healing) you do risk burning out of mana too quickly. Of course, they will still have Divine Plea and other mana regeneration mechanics.

As a footnote, we are also likely to slightly adjust Replenishment and buff the amount of MP5 on gear.

These changes should lead to healing as Holy being more dynamic – you are going to be targeting a lot more group members than just the tank, while still providing massive healing on the tank. However, you won’t be able to just constantly spam your biggest heals, at least not on the more challenging encounters. There is an opportunity here for skilled players to really do some outstanding healing as paladins, but it’s going to take a little more effort. This is something a lot of players have been asking for so we hope that it delivers without completely changing what some paladins enjoy about their healing style.

Retribution
Depending on who you ask, Retribution paladins are either perceived as being in need of improvements to make them stronger or are in need of changes to make them less powerful. We are really happy with the overall changes to the Seal and Judgement system from Lich King. However, Ret still has some problems. There aren’t enough buttons to push and they are all limited by cooldowns, allowing Ret paladins to have a lot of up-front burst without requiring a lot of skill or timing to actually hit those buttons (now after the opening moves of a fight, we think things get a lot more interesting because now the paladin’s defenses, healing and dispelling can come into play). We have other problems to fix as well. Exorcism currently can’t be used on players, which we think is a weak design. We have also become less enamored with the drawback to Seal of Blood / Seal of the Martyr causing damage to the paladin.

Here are the design changes we hope will fix all of these problems:

First, Exorcism has a cast time of 1.5 seconds but can be used on players again. This will let paladins use it in PvP, but not while moving towards a target.

Second, we are changing The Art of War to make not only Flash of Light instant, but also Exorcism. Choose healing or damage. Paladins will have to watch for this proc and use Exorcism when it happens. The spell itself is still ranged, but the proc will only occur when the paladin is already in melee.

Third, we are lowering the cooldown and damage of Crusader Strike to four seconds from six seconds. This accomplishes a few things. It lowers burst, it gives the paladins more buttons to push since they aren’t always waiting on cooldowns, it requires a little more skill since the player will have to choose between Crusader Strike and other attacks more often (such as the new Exorcism procs), and it gives Retribution a chance to get more damage out of their Seals (providing a sustained DPS boost for PvE).

Fourth, we are removing Seal of Blood and Seal of the Martyr. The damage recoil increasingly felt like a liability in PvE, and wasn’t serving to offset the massive burst damage capable in PvP. We are buffing Seal of Vengeance / Corruption and redesigning Seal of Command with the expectation that these are now the seals of choice for PvP and PvE. Righteousness can remain a tanking seal.

Finally, we are replacing the current effect of Vindication with an attack power debuff that works like Demoralizing Shout. In PvP, Ret paladins can still debuff melee targets, while in PvE they can provide another necessary debuff in the case that the tank is not a warrior or druid. In addition, the talent may be more attractive to Protection paladins.

These changes do not close the door on adding additional attacks, PvP utility or even the oft-requested Crusader Strike debuff. If the changes outlined above finally accomplish the goal of balancing Retribution in PvE and PvP, then we can start exploring those other design issues.

Again, these are our current plans and they may change based on testing and feedback. The changes to all three trees are things we are working on right now. Some of them might not work out quite the way we intended and will have to undergo iteration before finally showing up in 3.2. Even once the patch goes live, we are still likely to have to tweak numbers. This is simply a preview of what may yet come and we hope that it will be viewed as such.