Midnight Season 2 Tank Optimizer Updates

Today’s update had several adjustments for tank toughness optimization for Midnight season 2. We based this on early log data that we are seeing and a battery of tests that we run with our simulator and optimizer.

General

For all tanks, we have adjusted the default damage taken profile. Previously we were using 50% physical, and we have increased that to 65% physical. Note that this is raw, unmitigated damage taken. Your actual damage taken from physical will be much lower because tanks have very high physical mitigation.

We made this change after looking at logs for this season. This is a decent average – certain bosses you will take much less physical damage, and some you will take more. The main thing this affects is the value of mastery as a defensive stat for most specs.

We have also put less emphasis on the value of dodge/parry – we essentially apply a small penalty now due to its more random nature, which makes it not quite as good as e.g. armor or an absorb shield. The main impact this will have is a slight de-emphasis on critical strike for physical damage reduction.

Offense vs. Defense

Since the beginning of Midnight we have been defaulting to a “Mostly DPS” build for all tanks. This is how most people prefer to gear tanks (and how most guides out there will assume you are gearing/playing). We recommend that you skew more towards offense as the extra defense is often not necessary, and going more offensive doesn’t hurt your toughness that much.

It is worth repeating: if you do a mostly defensive optimization, you will NOT get results that look like what other guide sites recommend for tanks. Regardless of how any guide out there might present it, they are heavily skewed towards damage in their builds. Our optimizer is one of the few places that you can get a truly defensive build if you prefer that style, and it will be different.

Stat Priorities

Our website is a fully dynamic optimizer and simulator so we don’t hard-code “stat priorities” like you might see in a written guide, but we want to give you a rough overview of what to expect for each spec. This section is a generalization – the optimizer will adapt to everything from your chosen talents to your custom settings to give a personalized result.

Note on diminishing returns

Most guides that you see out there on stat priorities will make a note about diminishing returns on secondary stats. Yes, they are a thing, and no, you don’t need to know how it works or even think about them. Firstly, our optimizer handles it all for you. Secondly, the amount that they diminish is very small and of little consequence unless you are stacking a single stat to the complete exclusion of another. But even if diminishing returns didn’t exist you would never want to do that due to how secondary stats are designed (they all multiply each other), so it never becomes a real concern.

Offense

In general, crit+haste in roughly equal amounts will give you the highest damage for every tank spec. We’ll make notes below for cases where this might vary, but it is rare. And this makes sense – those two stats are designed to be more impactful for damage.

Versatility will almost always be the worst offensive stat due to its worse conversion rate compared to the other stats.

Mastery is a very decent damage stat, but will almost always sit slightly below crit and haste and slightly above vers.

Defense

There is more variation from spec to spec for defensive stats, but in general you can expect Mastery and Versatility in roughly equal amounts to be the default all-toughness build. As mentioned above, it is rare that a person would build for all defense which is why you rarely see anyone doing this. We’ll make notes below about exceptions.

Haste is a very popular stat for tanks, but it is rarely the best stat for pure toughness. People tend to like it because it is a good damage stat and it “feels” better to have some haste. It certainly does have defensive benefits for all specs, but just like Mastery is “decent” for damage, Haste is usually just “decent” for defense.

Blood Death Knight

Nothing unusual here, follows the general crit+haste for offense and mastery+ver for defense trend. In general you might find slightly higher-haste builds at some combos of offense+defense, but nothing too far out there.

Vengeance Demon Hunter

Mastery is a bad defensive stat for Vengeance. We ran all sorts of tests just to make sure we weren’t missing something… because it is a very simple stat in concept. Blizzard just didn’t tune it properly. Unless you are taking 90% or more of your unmitigated damage as physical (melee) damage, it can’t compete.

Thus we generally recommend avoiding Mastery for Vengeance – Haste takes its place as the 2nd best defensive stat next to Versatility.

Guardian Druid

Also pretty standard: crit+haste for offense and mastery+vers for defense. Some builds may favor slightly higher haste than crit for offense, but a decent amount of both is usually best.

Brewmaster Monk

Haste does not perform that well either for offense or defense for Brewmaster. Thus you will typically see lower haste builds no matter whether you go for damage or toughness.

Mastery is also a weak defensive stat for Brewmaster. This isn’t just because it’s dodge which people generally don’t like, it is also the “stack dodge until you get a dodge then reset” design that makes it tend to be weak… we would like to see Blizzard redesign this stat entirely.

Thus you will tend to see builds that have high Critical Strike and Versatility for Brewmaster.

Protection Paladin

Also follows the crit+haste for offense and mastery+vers for defense trend.

You might notice crit is slightly higher than haste for offensive-leaning strategies. Part of this is the emphasis on crit with the season 2 set bonuses.

Haste is also the weakest stat for defense for protection paladin, so if you start moving away from pure offense at all it will tend to be dropped by the optimizer. This might be surprising to some people as haste is often cited as the best stat from protection paladins… but we ran several tests and its defensive value just isn’t good in most scenarios. Its main value is offense.

One of haste’s main defensive benefits is more holy power generation, which lets you cast more Shield of the Righteous or Word of Glory. But even with very little haste you can maintain high SotR uptime (helped out significantly by the apex talents). Thus the holy power generation defensive value is tied largely to Word of Glory. That can be a little hard to assign a defensive value to… our intuition is that most people would prefer to burn any extra holy power on more SotR casts for more damage and stick to mostly free WoG from shining light.

Protection Warrior

Crit, haste, and mastery are all of very similar value for damage. Mastery+vers are the best for defense, but haste is pretty good too and you will see a decent amount.

It is worth noting that the value of crit and mastery for damage is very close, and it is not uncommon to see things tip in favor of mastery. The main drivers of this are a large chunk of base crit chance on Shield Slam from the apex talents which brings the value of crit on gear down a little, and the Tough as Nails talent which is a generally small amount of damage (more noticeable in Mythic+), but that chunk of damage scales very well with mastery and poorly with crit (the damage can’t crit, and more crit gives more parries which means less blocks – a true anti-synergy with crit).

Another thing worth noting is that Haste is generally better for Mountain Thane than Colossus. Most guides you will see out there are using Mountain Thane builds.

not for me…
seems like it favors:
haste > versa > mastery > crit

when picking “Tough as Nails” it favors:
mastery (~50%) > haste > versa > crit (~10%)

this is both with “Mostly DPS” and “Into the Fray” Talent which in previous versions made it heavily favor haste.

btw is there an overview which talents change the optimizer behavior?

I would have to see your specific case if you want to post a snapshot, there are a lot of factors that can influence the result.

Any talent can potentially change the optimizer behavior as we simulate your exact setup every time we optimize.

Without a specific case I updated the summary with some general notes about Protection Warriors.

Another thing to keep in mind when looking at Best in Slot results: a lot of BiS sets of gear include the “Venomcursed” items for season 2. These trigger a stat proc in one stat and a penalty in the others, e.g. the plate legs give a crit buff and haste/mastery/vers penalty. For a spec like Protection Warrior where all of the stats have similar value, this can shift the optimizer’s chosen gear – it might end up dropping a little crit and getting a little more mastery so that when this effect triggers the balance is more favorable.

Since posting this last week, Blizzard increased the mastery multiplier for armor for Vengeance – I guess they noticed how poor of a stat it was. Unfortunately it’s still not enough to make it overtake versatility for most cases.

Doing a few tests… I think they would be safe doubling the mastery coefficient, then it would be in the right ballpark to make it competitive.