Dev Watercooler: Mists of Pandaria PvP

Now that Mists of Pandaria has been out a few weeks, we wanted to let you all know what we consider working well in PvP, what we want to improve, and what’s coming down the pike. I’m going to split this up into three parts: a discussion of Matchmaking Rating (MMR), class balance concerns, and a sneak peek at some of our PvP plans.

MMR

As I’m sure you’re aware, the intent behind MMR is to match players by skill. This is accomplished by comparing an assigned number (MMR) that adjusts based on the relative MMR of the opponents you win or lose to. Matching players by their effectiveness or performance helps make sure that less experienced players aren’t always getting stomped by expert players, and that competitors at all levels are earning their victories by battling players of roughly equivalent ability. MMR isn’t intended to be a “PvP score”—its primary function is in matchmaking. We’ve seen a lot of players asking for an MMR reset, and to be fair, we had originally planned on doing one (and even announced it beforehand). Some of you may be concerned that high-ranking players won’t have a reason to play and will be able to just sit on their high ratings, perhaps even earning Gladiator without having to fight opponents. While we think there’s some legitimacy to the concern, we don’t think resetting MMR is the right way to address it.

Because MMR is a measure of your performance as a player, we don’t think it makes a lot of sense to reset MMR at the end of a season or expansion. It’s unlikely your skill atrophied significantly more than that of your competition between seasons, so there isn’t a lot of benefit in measuring it again. If we did reset MMR then what you’d likely see is a handful of mismatched games, where experienced players rock weaker players until MMR is calibrated again. That’s a lot of pain for relatively little gain since the MMRs will likely settle back into the same places anyway. It is important to remember that the Conquest point cap, Elite rewards and Gladiator titles are all tied to Team Rating and not MMR. Your Team Rating increases and decreases as you win and lose, and your Team Rating is really the most important number when it comes to Arena PvP.

However, we took a hard look at the way we measure Team Rating, and came to the conclusion that our formula doesn’t do a good enough job of measuring the potential for improvement over the course of a season. While your skill may not necessarily improve over the course of a season, your gear definitely does, and that should have an effect on the relative power of your team. We’re currently working on a change to allow Team Ratings to grow faster over time, as long as you keep playing. We don’t have any specifics to share yet, but this correction will have the added benefit that players who continue playing will be capable of reaching higher Team Ratings, and therefore potential access to rewards, than someone who stops playing and sits on their rating.

As a potential example let’s say the Fatty Goat Steaks have a Team Rating of 2700 and the Mushan Tongues have a Team Rating of 2500. The Fatty Goat Steaks decide their Team Rating is high enough that they’re going to sit out the rest of the season. The Mushan Tongues keep playing each week, their gear gradually improves, and they continue to win many of their matches. Because of the new inflation component, their Team Rating rises, let’s say, 300 additional points to 2800. They are now higher ranked than the Fatty Goat Steaks. If the Fatty Goat Steaks decide to jump back into Arena matches, they can benefit from the inflation as well, but they’re going to have to play and win to catch up. It’s not going to be a system where you have to play every day just to defend your rating, but certainly one where you can’t sit at the top doing nothing and collect your rewards at the end. We’ll have more details on exactly how this system will work as we get closer to its implementation.

We believe this adjustment to the Team Rating formula will have a similar benefit to the “rating decay” that some of you have been asking for, but will feel more positive – rather than feeling like you must keep running just to stay in place (i.e. keep your current score), players that continue participating will be rewarded with higher Team Ratings. As a season wears on, this should also make upper brackets more active as well.

To support this change we’re going to expand the 5.1 feature of improving the item level of your raid armor and weapons using Valor to now include Justice, Honor, and Conquest points as well. A Gladiator who has purchased a full set of Conquest gear will now have a compelling motivation to keep playing in patch 5.1, since they’ll be able to improve that set by spending Conquest points.

Class Balance Concerns

We have 34 class specializations in the game now, and the intention here isn’t to go through every one to discuss their strengths and weaknesses. Instead, I’m going to address the ones that players seem to be most concerned about. That’s inherently a subjective call on my part, so I apologize if you think we missed something. Also, simply because something isn’t mentioned here doesn’t mean it’s being ignored. You can see some of the specific changes we are trying out in the 5.1 patch notes.

Beastmastery Hunters – We agree that stacking too many cooldowns to blow someone up is not interesting, skillful, or fair. (I also want to be clear that with all of these issues, we’re not blaming players for using the tools that we gave them; we’re blaming ourselves.)  We’re taking a hard look at the various Hunter cooldowns with the intention of reducing their burst. Hunters are receiving a buff in that they will no longer need to swap between Aspect of the Hawk and Aspect of the Fox. We also hope these changes help result in higher representation of Survival and Marksman Hunters.

Warriors – We don’t think Warrior burst is out of control the way Hunter burst is. It’s possible that the combination of Warrior control and burst are too hard to counter, though. The changes we’re looking at are reducing both burst and control. The Glyph of Gag Order in particular is just too powerful for PvP. We think that if you manage to keep a Warrior from you, you should be allowed to get off a cast (and if you’re a Warrior who just can’t stand the thought of that, you can still spec into Storm Bolt.) Shockwave, Avatar, and Recklessness are also mentioned a lot, and we’re looking at them as well.

Rogues – We think that Rogue damage in PvP is appropriate—the damage that some other classes are dealing is too high, and we want to adjust those cases before we do any tinkering elsewhere. We still like the Rogue suite of tools and abilities, but we feel like some of these mechanics might have been eclipsed by newer, shinier versions.  We are looking at improving Rogue mobility and control in conservative ways.

Mages – We are going to tone down Mage burst and control. We had hoped moving more of their control to talents would force some exclusivity in crowd control options, but it wasn’t sufficient.  We increased the cost of Spellsteal significantly to promote its use strategically instead of stealing everything available. We’ve also been trying out changes to Deep Freeze, Frost Bomb, and Pyroblast to reduce burst and control.

Monks – We know some players are concerned about Windwalker viability in Arenas. The Monk class has a high skill cap and they come with a learning curve. The Windwalker PvP bonuses are quite powerful and players aren’t really taking advantage of them yet. We absolutely want the newest class to be popular, but we’re also trying to be conservative and not recreate the situation where the Death Knight dominated PvP and PvE when it first launched. We do want to give the Monk a few more weeks and see where it stands, but we are keeping a close eye on how things develop.

Healing – We agree that off-spec healers are too competitive with dedicated healers. Specs such as Shadow, Elemental, and Balance are supposed to be able to provide some healing—that’s one of their perks—but they shouldn’t be able to substitute for a dedicated healer. The change we are going to try here is to have PvP Power benefit either your damage or healing depending on spec. We are concerned that healing may be too high even for dedicated healers, especially for Restoration Shaman and Holy Paladins, who are incredibly strong in PvP at the moment. After we tone down some of the burst damage coming from Hunters, Warriors, and Mages, we may try the PvP healing debuff at 30% instead of the current 15%, but that won’t happen until after we’ve resolved the damage issues.

Crowd Control – At the moment, we’re satisfied with the state of crowd control. We tried to not add new stackable forms of crowd control to any given spec, but because players have so many crowd control options (say, choosing a stun vs. a snare via talents) the overall number of crowd control spells that a PvP player needs to learn has grown. The pool just became deeper. We know that some of you have concerns, and we’ll keep an eye on it.

Burst – Aside from Hunters, and possibly Warriors and Mages, we don’t think burst is out of control. Other specs that get mentioned by players are Destruction Warlocks, Frost Death Knights, and Shadow Priests. Those are definitely on our radar, but we don’t have any changes to announce at the moment.

PvP Sneak Peek

We have a lot of irons in the fire on the development team, and we’re usually reluctant to discuss ideas that aren’t fully implemented in the game yet. Despite all of our caveats, ideas that we delay or cancel tend to get turned into “broken promises.” With that in mind, please understand that you may not see some of these ideas in the next patch, or the patch after that, or any time soon. These are just ideas we’re bouncing around:

Loss of Control UI – This is a feature we’ve wanted to build for a long time, and it’s finally close. If you’re feared, you probably realize you’re feared because your character runs around crazy with a big chattering skull over your head. However, if you’re stunned or silenced or disarmed, you don’t often realize it immediately and push your buttons which don’t do anything and as a result the game feels unresponsive. The loss of control UI makes sure you know when something prevents you from using your abilities. It has benefits for solo or instance play as well, but it’s largely a PvP-driven feature.

Rated Battleground Participation Rewards – If you’ve PvPed for any length of time, you’ve been in one of those matches where the final score is so close that victory was a breath away. When we do our jobs right and give you one of those really close matches, it’s sad to not get any reward for participating, so we want to give out some kind of small reward. The rewards will be based on the final score to discourage exploitation.

Battleground and Arena Maps in Dungeon Journal – While veteran PvP players already know all of the Arena and Battleground maps, we think Dungeon Journal is a good way to introduce the basic objectives and map overlay so newer competitors don’t have to learn to swim while drowning.

Improved Scoreboard – It might seem silly to some, but we think small touches matter. Right now, the entire way we end a Battleground hasn’t changed much: it feels a little archaic and isn’t very exciting. We want to have an actual “You win!” toast, followed by a more exciting scoreboard.

Smoother Battleground Brackets – We have technology now to scale down player items for Challenge Modes, and you might be aware that we scaled up player items during our beta raid testing. We’d like to use the same mechanic to scale up players within the lower level PvP brackets. For example, in the level 15 to 19 bracket, we could make all characters behave as if they were level 19 for purposes of the Battleground. If this idea works well, we could potentially condense lower level brackets and maybe reduce queue times.

Matchmaking – Our Battleground matchmaking system hasn’t changed much since it debuted and doesn’t yet benefit from many of the advances we developed for Dungeon Finder and Raid Finder. The original Battleground queuing system was designed for speed because it only had a single realm of candidates. By incorporating more of the new tech we’ve since developed, we can help ensure that there are a certain number of healers per team or at least a good class distribution.

Honor and Conquest Item Upgrades – We’ve talked about how we’re going to allow players to upgrade raid items with Valor. As mentioned above, we are going to let PvP players increase the item level of their Honor or Conquest gear by spending Honor or Conquest, respectively.

Small Groups joining Rated Battlegrounds – We still don’t want to let solo players queue for Rated Battlegrounds because it would undermine the intent of coordinated group PvP, and then they really just become normal Battlegrounds that reward better gear. However, we’ve heard your concerns that getting even 10 players together to queue for a Rated Battleground can be challenging. Our idea is to let a group of 5 players queue together, which we would then match with another group of 5 players. We think this can still provide a relatively balanced and viable team. One of the crazier ideas on the table would be to convert some of our smaller Battlegrounds, such as Gilneas, to have a 5v5 option.

Tol Barad and Wintergrasp – We are discussing making level 90 versions of these PvP zones that players could queue for to earn bonus honor.

As I said, this is a brain dump of a lot of ideas that a lot of different developers are working on. This is not a list of patch notes and not all of these may come to pass. Please let us know what you think, if we’re on the right track, or if you think we missed a big PvP issue. PvP balance is always going to be really subjective, and while balance is desirable, continual buffs and nerfs that don’t actually solve problems have the potential to be worse than the imbalances they’re trying to fix.

Greg “Ghostcrawler” Street is the lead systems designer for World of Warcraft. He has NEVER impersonated a female in the game in order to get loot.