06/21/2011 11:57 PMPosted by MoreloreA is definitely false. The issue is not that they don't theorycraft, or that they don't see the issues, but that their tolerance for difference is much greater than the player bases, especially the one that posts here. Think about how they came up with the curves for DR and appropriate allocations of item levels. Blizzard clearly does plenty of theorycrafting and balancing, and it's not like they don't know the forums where the player base does this exists.
Actually - his A is pretty much on the mark. They clearly stated in Wrath when they were trying to defend prot warrior imbalance issues - that they could NOT buff the "most popular tank" even though they were the worst tank in actual game performance.
They *do* look at insignificant data when determining buffs and nerfs. Class population is a factor that should never determine the balance of that class - but they value it highly.
However, the theorycrafting community, especially here on the official forums, tends to use artificial models that don't actually reflect real raid situations and then get very upset if their projected damage taken isn't within fairly narrow bounds for every class. Blizzard takes the stance that actual gameplay is more complicated than that, and they, based on my reading between the lines and my judgment of their actual implementations, give much more weight to their actual results than to their theoretical models.
But those theorycrafted numbers from the community almost always turn out to be correct. The community sees these problems long before they are actually a problem and Blizzard simply turns a blind eye - again - looking at Wrath and prot warrior issues that carried through almost the entire expansion. Or the fact that the community saw the DK problems before they went Live in Wrath.
If tank class Bob takes more damage than tank class Jim, but there's no statistical bias in raid success toward Jim, then the extra damage probably doesn't matter.
That's the stance they took in Wrath as well. It was a model that caused various tanks to become overpowered in various tiers of gear, and led to two tanks basically taking over the game in terms of survivability and easier progression.
They are clearly hesitant to make pre-emptive balance changes, once they're past the beta phase. I don't think it follows from this that they don't predict the problems - you can see from the dev discussions that they're aware of the potential - but I think it reflects a reasonable caution.
This is what causes the constant see-saw effect of class balance. Instead of analyzing the impact of ALL of their changes/fixes, they look at things on a 1 to 1 scenario. This doesn't work.
Take the Arms Warrior "charge bug" in 4.1 as a quick expample - it gives arms a boost in damage thanks to higher use of heroic strike. In some fights - it pushes arms way past fury thanks to no down time and constant charge spam.
This makes arms DPS look high (and it is VERY high on those fights). So what do they do, they basically cut 8% of arms damage off to help balance it out a little. But wait - they also fix the charge bug (which is all they really needed to do to begin with). Instead, they view the charge bug as a bug and don't factor it into the extremely high numbers arms can pull on specific fights, and instead lower their damage to address that.
From people on the PTR - arms has fallen behind fury by a good margin now and we're approaching Wrath again where arms is slowly becoming a non-viable raiding spec.