The Game of Stat Values: Elemental Edition

Taking another look at Elemental stats, in particular how each one affects damage output. Please note that all of the work I’ve put in on this is using the T19P SimCraft profile, which has 10303 Crit (31.44%), 5068 Haste (30.09%), 1578 Mastery (25.02%) and 0 Versatility (0%).

The table of analysis & results can be found here, which compares the base profile to subsequent sims where I add 1000 of each stat in turn to see how it affects damage.

Mastery is still disappointingly falling behind

This is probably one of the biggest complaints about the 7.0 version of Elemental. Overload has been a “thing” since Mastery was introduced in Cataclysm. Even during the Warlords switch to Multistrike there was specific attention made to making sure that Multistrike still had that traditional Elemental feel too it, which is why Elemental multistrikes generated additional (but smaller) spell animations.

The problem isn’t a resource generation one. I’ve had a look at that and aside from 60%+ Overload chances on Lightning Bolt resource generation in Legion is better than in Warlords.

Part of the issue is that while Mastery increases damage for generating spells (Lightning Bolt & Lava Burst, plus Icefury & Elemental Blast talents) and increases resource generation it’s also reducing the time spent on those generator spells. Adding Mastery increased the cast count of Earth Shock by 1, or a 2.75% increase, but also slightly reduced the cast counts for Lava Burst and Lightning Bolt as that extra time spent casting an extra Earth Shock had to come from somewhere.

Looking at the total damage gains all stats look fairly similar, although 2.86% extra Crit is contributing more damage than 5.71% extra Overload chance. Similarly Haste increases the cast frequency by almost the same amount.

The only way Mastery wins out over other stats is if we exclude Elemental Pets from the total, as Mastery has just over 0.3% more player damage than either Crit or Haste, but the zero impact on Primal Fire Elemental means that the other stats have 0.33%-0.64% extra damage to make them “better”. This suggests that one possible solution for the Mastery question is to make it affect the Elemental Pets somehow, either a % damage bonus or its own “overload” function.

You can also see that while the 5.71% Overload chance increase translates to a 22.22% increase in Lava Burst overload damage it’s only a 10.57% increase for Lightning Bolt overload damage, even with similar proportional count changes. This is because Power of the Maelstrom generates Lightning Bolt Overloads that also count as “normal” overloads, so it reduces the proportion of damage increased. As a quick experiment, if the damage gain for Lightning Bolt overloads were the same as with Lava Burst overloads the gain from adding 1000 Mastery rating would be 3.13% overall. With such a low quantity (around 15 Crit rating per 1 Mastery rating) this would also bring the value of Mastery up to something a bit more even with the other stats.

There’s a similar issue with AoE damage too, as the Mastery value for HecticAddCleave fight types in SimulationCraft is almost half the value of the others. This is because Earthquake is a long & expensive spender, plus the fact that Static Overload has zero Mastery interaction (ie: the damage dealt & Maelstrom generation doesn’t change relative to Overload chance) & makes Haste a better Maelstrom generation stat than Mastery is.

It’s no surprise that when you take all of this into account that talents in a more “Mastery centric” build perform poorly when compared to other talents. Specifically looking at Echo of the Elements vs Ancestral Swiftness & Elemental Fusion vs Primal Elementalist. Ironically the only reason why Ascendance is competitive with Lightning Rod is because Crit chance increases Lava Burst damage while it’s up.

Thinking Critically, there’s too much

Part of the problem here is that Crit beats everything hands down, which is a completely unsurprising mechanics issue. You can blame Elemental Fury (50% bonus crit damage) and Elemental Focus (10% more damage to 2 spells after a crit), one of which (Fury) is also partially responsible for under performance in PvP (those 250% crits become 175% crits).

In order for Elemental to “scale” better (I dislike using the word scale in this context, as everything thinks it means something different) one of these passives needs to be removed to avoid the skewed stat priorities we’re currently seeing. My vote would be removing Fury as it’s “less flavour” & removing it would help in PvP. This would mean that Elemental Crits (primarily Lava Burst) would lose ~33% of their damage in line with other classes/specs rather than ~43%

Crit itself has traditionally been a “weak” stat because of Lava Bursts auto-crit function, and was only second equal-ish (with Haste, behind Multistrike) because of the Fury bonus & generally so little Crit vs the other two stats (plus we stacked Multistrike & had large haste bonuses via T18 4pc).

Rebalance mechanics, then buff

Unfortunately there’s not really a quick fix for all of this. Understanding why there’s an imbalance in stat priorities is the first step, but how to fix it and then balance damage outputs won’t be an easy task. Even with new hotfix tech I don’t expect miracles in the much discussed “Balance After 1 Week of Raiding” hotfix, especially as there’s already a lot of data from Heroic & Mythic Dungeons.

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

7 Responses to The Game of Stat Values: Elemental Edition

  1. Rythgar says:

    Thank you again Bink for doing the work you do. My question if you don’t mind is based on the template of stats that you used, does that imply that crit falls off after 30%? More of a crit (30%) > haste > crit/vers > Mastery?

    • binkenstein says:


  2. Krillin says:

    Thank you very much for your work. I have a huge respect for what you are doing. I was wondering: Is crit still the best stat (after int) if lightning rod was choosen instead of ascendence?

    • binkenstein says:

      Try your own sims to find out, but it probably is.

      • Krillin says:

        Thanks for your reply🙂 and as a sidenote I dont know how to simulate and i dont even know where to start^^ I guess its kind of complicated

        • binkenstein says:

          It’s slightly complicated, but generating a generic set of stat weights that apply to all talents/artifact traits/fights is pretty much impossible.

          There’s more information on the website, or if you’re on Earthshrine (Shaman Discord server) you could get some assistance there too ( )

  3. Trost says:

    I’m glad to have find your website. It is so cool to read your work.
    Big fan !

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: