Arcane Mastery Problems

Mastery effects vary from spec to spec, and I’ve already highlighted the issues around change based masteries like Overload, but there’s another I want to look at: Savant for Arcane Mages. This increases maximum mana and regeneration, as well as increasing the damage bonus from Arcane Charges, but there are some interesting interactions in how the spec is designed.

First, it’s important to remember that prior to Legion there was a passive called Nether Attunement for all Mage specs, which increased their passive mana regeneration so that if you were spending 20% more mana because of haste, you were also generating 20% more mana. Both Fire and Frost had mana spending issues as a result of this, resulting in a few mana cost reductions for their spells. For Arcane this introduced an interesting gearing conundrum: if you get more mastery your spells hit a bit harder and you have more mana, but outside of cooldown burn phases more mana doesn’t really translate into more damage.

Earlier in the year I ran some stat weight analysis sims to see how stat weights vary as you change the amount of each secondary stat. There were fairly consistent behaviours:

  • Haste has a slight negative feedback loop on itself while crit/mastery/vers have no feedback loops.
  • Soft & Hard caps are clearly visible by large drops in the value of the stat being adjusted
  • All stats had a roughly linear effect on the value of the others.


The problem for Arcane is that haste & mastery didn’t follow this pattern, and have a weird polynomial relationship as you can see from these two charts:

chart (4)chart (5)

As you can see, haste ends up having a positive effect on itself at higher levels while also dragging the value of mastery down. Meanwhile mastery also has an initial negative loop while haste has a similar parabolic curve to mastery in the haste based chart. The problem is that even if you can manage both stats well enough to be towards the high point of each curve Crit & Versatility still contribute more damage. This could all be because of the carry over of burn/conserve phases from previous iterations of the spec, but the problem is that a large portion of damage is generated during cooldown based burn phases where extra haste or mastery has minimal impact, and neither has a significant change effect during conserve phases either.

Looking at the actual data rather than the trendlines you could make an argument that there is a Mastery “breakpoint” around 40% where it hits a flat non-feedback level. Mastery then has a positive effect on the usual haste negative feedback loop, but I’m not sure why this effect grows at higher haste levels. It’s possible that further data analysis may reveal more information/insights into how the two stats interact.

The problem is that as Arcane effectively has two “do more” stats it makes balancing them a bit difficult. This is further compounded by the possibility that micromanaging mana may not be completely necessary: experimenting with a “simplified” Arcane APL which uses Evocation on cooldown & AB4\ABarr cycles when not in “burn mode” results in output that’s within 5% of the standard burn/conserve APL. With further optimisation this new APL could be roughly equal with the standard, suggesting that the additional mana regen doesn’t contribute anything.


About binkenstein
I'm a geek, living in Christchurch, New Zealand.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: