Blood Death Knight Tank Rotation For M+ (shadowlands)

Core Ability Priority Overview

  1. [Death Strike] if you are below 60% health.
  2. (Necrolords) [Abomination Limb]
  3. [Marrowrend] if [Bone Shield] is not active or about to expire.
  4. (Venthyr) [Swarming Mist] with less than 67 RP (61 RP with [Bryndaor’s Might] equipped) or for threat generation or when defensive usage is not needed for the next minute.
  5. [Blood Boil] if a target does not have [Blood Plague] and (Venthyr) [Swarming Mist] or (Necrolords) [Abomination Limb] is not active.
  6. (Kyrian) [Shackle the Unworthy] (with [Combat Meditation] enabled).
  7. (Night Fae) [Death and Decay] when the duration of the [Death’s Due] buff/debuff is about to expire, but with enough remaining time to [Heart Strike].
  8. (Night Fae) [Heart Strike] while in [Death and Decay] when the duration of the [Death’s Due] buff/debuff is about to expire or (the duration of our [Death and Decay] ground effect is about to expire and the [Death’s Due] buff/debuff won’t outlast the [Death and Decay] by at least ~9 seconds).
  9. [Death Strike] when Runic Power is above 105 or above 85 when [Dancing Rune Weapon] is active.
  10. [Marrowrend] if below 8 stacks of [Bone Shield] (below 6 stacks when [Dancing Rune Weapon] (with [Crimson Rune Weapon]) has less than 60 seconds on its cooldown or when [Dancing Rune Weapon] is active or (Necrolords) when [Abomination Limb] has less than 20 on its cooldown).
  11. [Death and Decay] as (Night Fae) or with 3+ targets
  12. [Heart Strike] with, or when 1.5 second away from, having more than 3 Runes.
  13. [Blood Boil] with 2 charges and less than 5 stacks of [Hemostasis].
  14. [Heart Strike] with 3 runes or (with 8+ stacks of [Bone Shield] and 7.5+ seconds left of [Bone Shield] duration and (Night Fae) while standing in the [Death and Decay] ground effect).
  15. [Blood Boil] with 1 charge and less than 5 stacks of [Hemostasis].
  16. [Death and Decay] when [Crimson Scourge] procs.

Non-Global Cooldown Abilities

  1. [Blood Tap] if you have 2 charges, or are less than 5 seconds away from 2 charges of [Blood Tap], and you have less than 3 Runes.
  2. [Blood Tap] if you have less than 3 Runes and less than 63 RP (57 RP with [Bryndaor’s Might] equipped).
  3. Use [Raise Dead].

Core Ability Priority Details

[1] [Death Strike] if you are below 60% health.

Basis: [Death Strike] is our primary mitigation ability and using it effectively is the key to staying alive; this priority is meant to use [Death Strike] when it’s efficient. Using [Death Strike] under the right circumstances is our top priority, since you will either die or require an excessive amount of healing from your healer if you are not deliberate in your [Death Strike] use.

So, under what circumstances do we want to use [Death Strike]?

[Death Strike] healing is proportional to the damage we’ve taken in the last 5 seconds, though any overhealing will be lost (with the exception of the resulting [Blood Shield]). For a simple and effective method, it is recommended to use a fixed health %, alone, as a trigger for when to use [Death Strike], since situations where you drop below any given health % are almost always going to be when you have been taking significant amounts of damage in the last 5 seconds.

What health exact % should be used as a setpoint?

There are a few things we need to take into consideration: 

  1. If the threshold is too low, it becomes more likely that we’ll overshoot our threshold and die before we can reactively use [Death Strike].
  2. If the threshold is too low, too many of our [Death Strike]‘s may be used inefficiently to avoid capping RP, for minimal healing, when they could have been used more efficiently during a damage spike that didn’t push us below our overly-low threshold.
  3. If the threshold is too high, [Death Strike] might be wasted during low damage periods that still manage to push us below our overly-high threshold, when they could have been used more efficiently on higher damage periods.
  4. If the threshold is too high, our RP levels will be chronically low, since we’re using [Death Strike] too aggressively. This means we won’t have as much RP (our second health bar) saved up for when situations become dangerous.

As such, we want to get as close to that theoretical sweet spot as possible to maximize our overall [Death Strike] healing, which also has us pooling sufficient RP for truly dangerous situations and keeps our health pool high enough that we’re unlikely to instantaneously die before we have the chance to use [Death Strike].

In reality, the ideal threshold may change from situation to situation, but, for the purposes of this guide, a general trigger point of 60% health is recommended and has been found to be effective in the highest BDK M+ key levels.

[2] (Necrolords) [Abomination Limb]

Basis: [Bone Shield] provides a significant amount of armor and a respectable amount of haste. Generally, this buff reduces physical damage intake by over 20%, which is a large increase to our mitigation and EHP.

[Abomination Limb] provides up to 9 [Bone Shield] charges, though, because we generally don’t want to spend time below 5 charges and outside of [Ossuary] range, getting all of these charges is not always possible. Still, dropping down to even 5 charges of [Bone Shield] before casting [Abomination Limb] means that we’ll get most of the benefit, since we’ll drop additional stacks before we get the last tick of [Bone Shield] generation 12 seconds later. Even using a low estimate of 6 [Bone Shield] charges generated per cast, those 6 charges will result in a minimum of 28 additional RP through four additional casts of [Heart Strike] over two casts of [Marrowrend].

Why is this not ranked higher? Recovering from a low health % is going to do more for your immediate survivability than an armor buff or threat generation.

[3] [Marrowrend] if [Bone Shield] is not active or has a duration less than 5 seconds.

Basis: [Bone Shield] provides a significant amount of armor and a respectable amount of haste. Generally, this buff reduces physical damage intake by over 20%, which is a large increase to our mitigation and EHP.

Ideally, if [Marrowrend] is going to be used specifically to refresh the duration of [Bone Shield], you want to try and prevent a situation where the 20 generated RP is wasted due to having too much RP. It is suggested to track when [Bone Shield] is within 5 seconds of expiring so you have some time to efficiently sequence abilities without the immediate need of having to use [Marrowrend] on your next global and potentially overcap RP.

Why is this not ranked higher? Recovering from a low health % is going to do more for your immediate survivability than an armor buff and [Abomination Limb] provides a more efficient way of generating initial [Bone Shield] stacks.

[4] (Venthyr) [Swarming Mist] with less than 67 RP (61 RP with [Bryndaor’s Might] equipped) or for threat generation or when defensive usage is not needed for the next minute.

Basis: [Swarming Mist] is a versatile ability that can be approached in a few ways:

  1. As an ability that can be used to maximize raw RP generation and damage.
  2. As an ability that can provide “reserve” RP to be used *explicitly* when RP is low.
  3. As an ability that can be used to strongly establish AoE threat at the start of pulls.
  4. As an ability that can be used as a defensive cooldown between the provided dodge and *potentially* high RP generation.

While it is possible to use [Swarming Mist] in a way that satisfies #4 and at least one other strategy, there is no way to use it in a way that satisfies all four or even consistently to satisfy three of them.

No matter what, there doesn’t seem to be a downside to taking the approach outlined in #4 (using [Swarming Mist] as a defensive cooldown), so it is in our best interest to try not to stack defensive cooldowns on top of [Swarming Mist] unless absolutely necessary. After that, we’re mostly just looking at a damage vs. mitigation vs. threat tradeoff between #1 (using [Swarming Mist] on cooldown) and #2 (using [Swarming Mist] reactively at “low” RP levels), and #3 (using [Swarming Mist] at the start of pulls). For the purposes of this guide, it is advised to use it primarily for threat, when needed, in order to prevent group member deaths. When you have available casts outside of these circumstances, you next want to prioritize using it to maximize useful RP generation with #2.  Finally, if you have no immediate need to use the ability for the first two purposes, it is best to use it as soon as possible. For better or for worse, this ability requires a lot of in-the-moment judgement with its use.

What is the basis for the [Swarming Mist] RP thresholds?

When used for reserve RP we want the generated RP to gain as much value as possible. Since [Swarming Mist] can generate up to 145 RP over 8 seconds, it is especially valuable to use it at lower RP levels, since we are more likely than not going to have to spend some of that RP inefficiently to prevent overcapping resources. At the same time, we don’t want to hit the point where we might die before putting the RP to use. A simple threshold of less than 67 RP (61 RP with [Bryndaor’s Might] equipped) is suggested for this purpose, which prevents you from putting yourself in a situation where casting [Swarming Mist] does not give you enough RP to use a [Death Strike] on the following global. This threshold basis is subjective and does not attempt to account for the number of targets being hit.

Wait, can’t [Swarming Mist] cause us to overcap RP if it is still active while we are doing the rest of our rotation?

Yes. If possible you want to adjust your play to account for the extra RP generation while [Swarming Mist] is active to avoid overcapping RP. At the same time, you don’t want to go too crazy with casting [Death Strike] to the point where you are left with low levels of RP when [Swarming Mist] goes away.

Due to clutter concerns, revised RP thresholds aren’t listed for a [Swarming Mist] condition. However, if you want to apply something like that into your play it is safe to assume that [Swarming Mist] will pulse up to two times during any global cooldown, which can generate anywhere from 6 to 30 RP on 1 to 5+ targets (9-38 RP with [Rune of Hysteria]) before your next ability can be used. So, you can adjust your thresholds accordingly, if desired.

Why is this not ranked higher? Our immediate survivability and our primary form of proactive mitigation are more important than ramping up our RP generation.

[5] [Blood Boil] if a target does not have [Blood Plague] and (Venthyr) [Swarming Mist] or (Necrolords) [Abomination Limb] is not active.

Basis: The only basis for this priority is that [Blood Boil] is how we establish threat on multiple targets. If we waited 2-4 globals before using some AoE threat generation, then your targets might decide to kill a dps/healer in the meantime. In fact, if initial threat continues to be an issue, an additional [Blood Boil] near the start of a pull may be a consideration, but this situation isn’t common enough to account for it in the core priority list. In any case, even if this priority is suboptimal in one or two target situations, there is very little being lost using this same priority in all situations, since you will generally only be using this ability at this priority once per encounter and it only slows you down from putting runes on cooldown by one global cooldown.

[Blood Plague] provides low enough damage and healing that it is NOT justifying why we are going out of our way to apply it; we’ll be casting [Blood Boil] often enough, anyway, that using [Blood Plague] at this priority is only giving us an extra ~6 seconds of [Blood Plague] uptime, which is only healing us for ~0.5% of our HP, per target, over that duration.

The primary defensive benefit to using [Blood Boil] is actually to generate [Hemostasis] stacks. Let’s assume that, at most, 15% of our [Death Strike] healing overheals (not including [Blood Shield]) and that our Mastery is, at max, 28% (remember, [Blood Shield] is not affected by [Hemostasis] stacks); real overhealing values rarely get lower than 15% and 28% Mastery is a relatively low value. With these assumptions, each stack of [Hemostasis] increases the effectiveness of our next [Death Strike] by ~6% per stack gained. If we further assume that it will take 40 RP to use [Death Strike], then we can say that a cast of [Blood Boil] is equivalent to 2.4 RP per stack of [Hemostasis] gained; a value that gets worse if our assumed values of over-healing and Mastery get higher. These numbers *also* aren’t being used to justify this priority, though they will be used for some in the future.

If [Swarming Mist] or [Abomination Limb] is active, then initial threat is not a concern.

Why is this not ranked higher? We want to ensure that [Bone Shield] is active to reduce damage from incoming hits and to provide additional Haste before worrying about establishing AoE threat. Granted, [Swarming Mist] and [Abomination Limb] are capable of doing a better job of establishing threat when they are available and there are good conditions for their use.

[6] (Kyrian) [Shackle the Unworthy] (with [Combat Meditation] enabled).

Basis: In a single-target scenario, [Shackle the Unworthy] provides a 14 (17 with [Proliferation]) second 8% damage reduction, which doesn’t justify this placement from a mitigation contribution perspective, especially on single targets. However, this ability instantly refunds one rune on cast as well as an additional rune for each time [Shackle the Unworthy] spreads, so the long-term mitigation contribution of this ability is significant in addition to having an abnormally large damage contribution. While you’ll likely lose some rune regeneration at this priority, some level of resource loss is inevitable when playing (Kyrian) on multiple targets and trying to maximize regeneration will cause you to lose more throughput than you’d gain otherwise.

Why is this not ranked higher? Our immediate survivability, our primary form of proactive mitigation, and establishing initial threat are all more important than [Shackle the Unworthy], regardless of any longer-term mitigation or damage benefit it may have.

[7] (Night Fae) [Death and Decay] when the duration of the [Death’s Due] buff/debuff is about to expire, but with enough remaining time to [Heart Strike].

Basis: The Night Fae Covenant gives us 8% more Strength and 8% damage reduction if we can maintain our [Death’s Due] buff and debuff at full stacks. While some of the time we can maintain the buff and debuff off of [Crimson Scourge] procs, it is a Mitigation, EHP, and Damage increase to cast [Death and Decay] without [Crimson Scourge], if the alternative is losing our [Death’s Due] buffs and debuffs. Stacks refresh only when using [Heart Strike] in [Death and Decay], so make sure you are providing enough time for that to occur.

Remember, you’ll need enough time to cast both [Death and Decay] *and* [Heart Strike] before [Death’s Due] expires, additionally, you may want to leave yourself enough time to cast one [Death Strike] as well if it is needed for immediate survival or to prevent overcapping RP.

Why is this not ranked higher? Our immediate survivability, our primary form of proactive mitigation, and establishing initial threat are all more important than ramping our [Death’s Due] buffs/debuffs.

[8] (Night Fae) [Heart Strike] while in [Death and Decay] when the duration of the [Death’s Due] buff/debuff is about to expire or (the duration of our [Death and Decay] ground effect is about to expire and the [Death’s Due] buff/debuff won’t outlast the [Death and Decay] by at least 9 seconds).

Basis: This priority is meant to insure that our [Death’s Due] buff/debuff will maintain full uptime and that the [Death’s Due] buff/debuff duration will last long enough to be refreshed during our next cast of [Death and Decay] when our most recent cast of [Death and Decay] is about to fade.

Without a timely [Crimson Scourge] proc, [Death and Decay] will not be castable for 5 seconds after the previous cast fades. After taking the forced downtime, human reaction time, the cast time of [Death and Decay], and any extra time to support [Death Strike] usage, you’ll probably want your [Death’s Due] buff/debuff to last at least ~9 seconds after your current [Death and Decay] is about to fade.

Again, we don’t try to be overly restrictive of the timing or you can lose RP efficiency or the [Death’s Due] stacks themselves. At the same time, but we also don’t want to [Heart Strike] too excessively to the point where it hurts our [Bone Shield] management.

Why is this not ranked higher? Initial threat generation is more important and this does not directly conflict with the previous (Night Fae) priority.

[9] [Death Strike] when Runic Power is above 105 or above 85 when [Dancing Rune Weapon] is active.

Basis: While we want to maintain reserve RP (so we can use [Death Strike] in situations where it’s needed), if our RP levels get too high, then we won’t be getting full benefit out of our RP generating abilities. In 9.2, this situation becomes a bit more complicated because BDK’s have too much resource generation to fully utilize everything and, on more than two targets, we can gain damage and run a higher RP pool by prioritizing rune and [Blood Boil] usage over RP usage (which is seen as preferable to lower damage and a lower RP pool. As such, we are overly concerned about wasting some amount of runic power on multiple target pulls and the RP thresholds on this priority mostly exist for the sake of single target efficiency.

What RP threshold should we use to decide when we use [Death Strike] to avoid capping excessive amounts of RP?

In general, the higher the RP threshold on this priority, the more RP we will have pooled to use when we start taking significant amounts of incoming damage. However, once this threshold gets above 95, we begin making a tradeoff between a larger average RP pool (which can increase our overall survivability) and some level of lost RP. In 9.2, the tier set bonuses greatly increase the potential to generate large amounts of RP in a single global cooldown, but we are generally only concerned about excess RP waste on single targets in 9.2 due to excessive resource generation. In such a situation, the highest amount of RP that can be generated is 17 RP (27 RP during [Dancing Rune Weapon]), with some chance of a tier set proc to generate an additional 17 RP (27 RP during [Dancing Rune Weapon]). The use of the (Night Fae) legendary increases RP generation potential by 20%, but it also increases resource bloat in a way that allows for similar thresholds to be used regardless of covenant.

This guide recommends using [Death Strike] whenever we are above 105 RP (or 85 RP with [Dancing Rune Weapon] active), where the only way RP is only wasted in some capacity with tier set bonus procs or on multiple targets where the RP waste is less of a concern. This small loss in RP efficiency on single targets is thought to be worth the tradeoff for the extra RP pooling it provides. But, feel free to adjust this as you please.

Why is this not ranked higher? If anything, the global spent on [Blood Boil] (or [Shackle the Unworthy]) , buys you some time to potentially take more damage and make this non-essential [Death Strike] more valuable, which can improve your RP efficiency. Even if it doesn’t result in a stronger [Death Strike], [Blood Boil] generates no RP and isn’t causing you to overcap. In any case, establishing threat takes priority over a non-essential [Death Strike].

Also, even if RP is lost as a result, maintaining [Death’s Due] buffs/debuffs is more important.

[10] [Marrowrend] if below 8 stacks of [Bone Shield] (below 6 stacks when [Dancing Rune Weapon] (with [Crimson Rune Weapon]) has less than 60 seconds on its cooldown or when [Dancing Rune Weapon] is active or (Necrolords) when [Abomination Limb] has less than 20 on its cooldown).

Basis: With a high [Bone Shield] stack count, you will never be in any immediate need to [Marrowrend] to maintain 5+ stacks for the 5 RP cost reduction on [Death Strike] with [Ossuary]. However, it’s important that [Bone Shield] stacks aren’t wasted when you [Marrowrend], which would occur when we have more than 7 stacks. High [Bone Shield] stacks also give you more flexibility to use your runes without losing [Ossuary] and it maximizes the benefits of [Foul Bulwark] if using that talent, so keeping stacks high is the default recommendation. Even if you are only generating 5 bone shield stacks on a [Marrowrend] during [Dancing Rune Weapon], it is still generally more efficient than losing [Ossuary].

With [Crimson Rune Weapon] or (Necrolords), it is recommended to swap to a lower [Bone Shield] band when [Dancing Rune Weapon] or [Abomination Limb] are coming off of cooldown in the near future, so we can maximize the bone shield generation of these abilities while avoiding a significant loss of [Ossuary] uptime. Ideally, we want our [Bone Shield] stacks as low as they can possibly be when the corresponding abilities are coming off of cooldown *without* losing any [Ossuary] uptime in the process.

Why is this not ranked higher? There is no immediate urgency to top off [Bone Shield] stacks. While this priority does enable the 5 RP cost reduction on [Death Strike], that RP efficiency isn’t going to make up for the fact that we’re overcapping on RP generated by this [Marrowrend] if we’re above 105 RP, in most cases. An argument could be made to make a [Marrowrend] priority higher on this list for when 4 or less RP would be lost to overcapping and where the [Marrowrend] would enable the 5 RP reduction to become active, but there aren’t a lot of gains to be had for the added rotation complexity. We should almost never have such high levels of RP in combination with such low stacks of [Bone Shield].

[11] [Death and Decay] as (Night Fae) or with 3+ targets

Basis: With [Crimson Scourge] active, [Death and Decay] costs nothing and generates no RP on cast. This tradeoff is preferable since the saved rune can now be used on [Heart Strike] to generate even more RP. The value from this priority is that standing in [Death and Decay] allows us to generate an extra 2, 4, and 6, RP per cast of [Heart Strike] on 3, 4, and 5+ targets, respectively. It is reasonable to assume that, with proper ability usage, we can fit at least 3 [Heart Strike]‘s reliably during the duration of [Death and Decay], for 6, 12, or 18 extra RP per cast of [Death and Decay].

Without [Crimson Scourge] active, [Death and Decay] costs a rune and generates only +10 RP for that rune. This will reduce the amount we can [Heart Strike], but will increase our [Death and Decay] usage and uptime (which has a complicated impact on our overall throughput). Depending on current tuning and character considerations, it may or may not be worth using [Death and Decay] without a [Crimson Scourge] proc. However, in the current state of the gain, it is generally worth using [Death and Decay] regardless of [Crimson Scourge] status as long as you are using [Crimson Rune Weapon] (or the 4-piece tier set bonus) when 3+ targets are present. The math can change if you are using [Relish in Blood] or if you are not using [Heartbreaker], but neither of those actions are recommended for this guide.

Being (Night Fae) (particularly with [Grove Invigoration] or [Field of Blossoms] enabled) means that [Death and Decay] is providing additional benefits to the point that casting [Death and Decay] even without a [Crimson Scourge] proc or multiple targets simplifies the rotation significantly and doesn’t put the player in a situation where they are taking obvious performance losses.

For the sake of [Death’s Due] management and limiting the added rotation complexity associated with (Night Fae), it is recommended to aggressively use [Death and Decay] over [Heart Strike] even when there is not an immediate need to refresh [Death’s Due] stacks.

Why is this not ranked higher? It is generally better to get our [Marrowrend]s out of the way before using [Death and Decay], since we won’t be generating [Death’s Due] stacks, anyway, if we don’t have the freedom to [Heart Strike] and we want to avoid situations where we need to refresh [Bone Shield] and [Death’s Due] at the same time without enough time available to do both.

[12] [Heart Strike] with, or when 1.5 second away from, having more than 3 runes.

Basis: We can regenerate up to 3 runes at any given time, so we are missing out on rune regeneration if we have more than 3 Runes. This priority is making sure we are using all of our rune regeneration and, with [Bone Shield] topped off, [Heart Strike] is going to generate the most RP per rune.

Why is this not ranked higher? It is generally better to efficiently top off [Bone Shield] stacks than to frontload RP generation, unless you are in immediate danger of dying and desperately want to build RP to [Death Strike]. It can be argued that a higher priority of [Heart Strike] should be added to capture this situation, but this is not included because panic [Heart Strike]‘s are a very short-term band-aid that will quickly cause more problems than they solve with loss of RP efficiency from dropping below 5 stacks of [Bone Shield].

Unfortunately, the math starts to get very messy with this when comparing it against the above priority; it has been done, but it is not being written out since both priorities interact with each other on 3+ targets in a complicated way. But, as long as the player is using [Heartbreaker] while also either being (Night Fae) or, with 3+ attackable targets, using [Crimson Rune Weapon] or the 4-piece tier set bonus, it is generally seen as a net gain to [Death and Decay] above normal [Heart Strike] priorities regardless of whether or not we have a [Crimson Scourge] buff active.

[13] [Blood Boil] with 2 [Blood Boil] charges and less than 5 stacks of [Hemostasis].

Basis: With 2 charges of [Blood Boil], we are missing out on potential [Blood Boil] casts (and [Hemostasis] stack generation), by not using this ability. As long as we are not already capped on [Hemostasis] stacks, we want to generate more to increase the effectiveness of our next [Death Strike].

[Hemostasis] is the primary defensive benefit of using [Blood Boil]; the previous [Blood Boil] priority can be seen for this discussion.

Why is this not ranked higher? Casting [Heart Strike] will generate more effective RP than using [Blood Boil] when the maximum number of [Hemostasis] stacks are generated at all target counts.

[14] [Heart Strike] with 3 runes or (with 8+ stacks of [Bone Shield] and 7.5+ seconds left of [Bone Shield] duration and (Night Fae) while standing in the [Death and Decay] ground effect).

Basis: We want to go down to 2 runes to create some margin for our rune regeneration as our runes come off of cooldown. But, we usually want to preserve 2 runes so we can [Marrowrend], if needed.

In general, we can only go below 2 runes (without missing a crucial [Marrowrend]) If our [Bone Shield] is at sufficient stacks and has remaining duration. However, (Night Fae) has to manage their [Death’s Due] stacks in addition to [Bone Shield], so they have an added condition when using these final runes that they need to be standing in [Death and Decay] to guarantee that they won’t (also) risk losing their [Death’s Due] stacks.

Why the [Bone Shield] thresholds?

For duration, the longest amount of time you would be unable to use [Marrowrend] if you just spent on your runes on [Heart Strike] is ~6 seconds in a worst-case scenario, so the duration of [Bone Shield] needs to be sufficient to cover this window beyond the buff duration. This number is being increased to 7.5 for consistency with the next priority and to account for player reaction time.

For the number of stacks, the largest amount of stacks we can lose in 7.5 seconds is 4 stacks of [Bone Shield] (with really bad luck). So, we have no risk of falling below 5 stacks of [Bone Shield] if we have 8 or more stacks before using our reserve runes.

Why is this not ranked higher? At this point, we are not in immediate danger of losing out on long-term Rune generation, since 3 runes are already on cooldown. And, if we were to not use [Blood Boil] above when we can generate [Hemostasis] stacks, then we’d be missing out on long-term [Hemostasis] generation.

[15] [Blood Boil] with 1 [Blood Boil] charge and less than 5 stacks of [Hemostasis].

Basis: This will reduce the likelihood that we’ll miss out on long-term [Hemostasis] stack generation by sitting on 2 [Blood Boil] charges for any period of time. There is no particular reason to save 1 charge.

Why is this not ranked higher? Runes are more valuable than [Blood Boil] charges based on previous discussion, so preventing lost rune regeneration time takes priority.

[16] [Death and Decay] with a [Crimson Scourge] proc.

Basis: Under these conditions [Death and Decay] is a no cost damage ability with no defensive benefit. We’ve used everything else at this point, so we may as well take advantage of it.

Why is this not ranked higher? This ability provides no mitigation value at this priority; everything else above it *does* provide mitigation.

Non-Global Cooldown Abilities

[1] [Blood Tap] if you have 2 charges, or are less than 5 seconds away from 2 charges of [Blood Tap], and you have less than 3 runes.

Basis: If we are under these conditions, long term rune regeneration is, or is about to be, lost by not using [Blood Tap].

[2] [Blood Tap] if you have less than 3 runes and less than 63 RP (57 RP with [Bryndaor’s Might] equipped).

Basis: [Blood Tap] gives us the ability to pool an additional rune without losing long-term rune regeneration. So, the question is, under what circumstances do we want to use this additional rune?

  1. The lower this limit is, the higher likelihood it is that it will be used in a situation where it is needed.
  2. If the limit is too low, the time needed to convert the rune into RP and into a [Death Strike] may be too long to avoid a player death.

So, how low is too low? For the purposes of this guide, that is when the extra rune is not enough to enable an extra [Death Strike] on its own. Given that the minimum RP generated from a [Heart Strike] is 17 RP with [Heartbreaker], we don’t want to be below 23 RP while we are still holding onto a charge of [Blood Tap]. We get into this situation by using [Death Strike] when we are below 63 RP, so being below that is our trigger for using our last charge of [Blood Tap].

[3] Use [Raise Dead].

Basis: [Raise Dead] is a minor 1 minute duration single target damage cooldown. Either use it on cooldown or attempt to strategically use it when single target damage is more valuable, but, like all damage cooldowns, sitting on the cooldown can quickly become a net loss in most situations.

Potential Deviations from the Core Priorities

Does anything change in the opener?

In most ways not a whole lot changes during the opener, but, if you are beginning a pull with full or near-full runes available, one of two actions can become efficient to do if you specifically perform them on pull. Both of these actions speed up your immediate rune consumption, so you can start your rune regeneration faster. As such, the cost of these actions specifically on-pull are much lower than normal, making them worth doing under these circumstances.

  1. If you [Death and Decay] *at range* at the start of an encounter for increased damage and mitigation at any target count. (If you are in melee range during this GCD, it is not worth doing.)
  2. Use [Rune Tap] immediately on pull for a usage that is likely a mitigation increase. (This becomes less useful if other defensive cooldowns are available or if enemies are initially scattered.)

Especially in M+, the on-pull [Death and Decay], in particular, is especially useful and also aids in improving on-pull threat.

What if I’m outside melee range for an extended period of time?

At range, our only offensive ability options are [Death and Decay], [Death’s Caress], and [Chains of Ice] (also [Death Coil], but [Death Coil] is almost always not worth using). Without a [Crimson Scourge] proc, all of these abilities cost 1 rune and generate 10 RP compared to the 15-30 RP that we could get from a [Heart Strike]. So, if you think you’ll be out of melee range for more than ~4 seconds with more than 3 runes, it can be beneficial to use any of these ranged abilities to generate RP while maintaining full rune regeneration.

What if I’m at low health and I don’t have enough RP for a [Death Strike]?

We have a few options as follows:

  1. Kite (more details on this below)
  2. Use one of the following consumables as a substitute for [Death Strike]:
  3. Use remaining runes to [Heart Strike] for enough RP to [Death Strike]

The best options are going to depend on the exact situation you are in. Using [Heart Strike] desperately for [Death Strike] should only be used as a last resort; it’s going to quickly put you into an unrecoverable situation and there’s a time delay before it can help save you. 

When is kiting appropriate?

Kiting should ideally only be done when needed for immediate survivability or if a specific pull has an excessively high potential to kill the tank when they are not kiting.

Kiting is mainly used to mitigate damage or to drop stacks of a stacking debuff (such as [Necrotic Wound]). Knockbacks, snares, personal speed buffs, and other forms of CC (Crowd Control) are usually used to support kiting. Use of kiting generally results in targets moving and spreading out, both of which have a negative effect on group damage (in addition to the loss of tank damage). Additionally, [Death Strike] healing is based on the damage we have taken in the last 5 seconds, so any attempt to *stop* kiting will mean we initially have a weaker [Death Strike] for the initial 5 seconds of re-engagement. One more consideration is that some mobs have high-damaging abilities that normally target the tank, but will target anyone within melee range if the tank, themselves, is not in melee range, so this can result in non-tanks getting targeted with those, often lethal, abilities. As such, you generally want to minimize kiting, when possible.

If you are in a situation where you are forced to kite, it is still often preferable to stay in melee range with at least one or more enemy units if at all possible, since our resource generation is heavily penalized without the use of melee attacks. As far as incoming damage is concerned, you are still drastically lowering incoming damage by staying out of melee range of the majority of the enemies. [Death Grip] is particularly useful in moving an enemy slightly in front of a cluster of enemies so they can be targeted by melee abilities while you remain out of range of other enemies, while [Death Grip] is still capable of hitting enemies just outside of melee range. Kiting in a tight circle maximizes the time enemies remain in AoE effects and helps to keep them grouped, though this has the side effect of making it more difficult to keep only a few enemies within melee range. For the purposes of keeping some enemies in melee range, a straight line can be better.

What tools do we have to support kiting?

BDKs have a few kiting tools available:

  • [Death and Decay] with [Grip of the Dead] provides a powerful AoE snare on cast that quickly gets weaker. It has a short cooldown and can make it easy to get out of melee range of enemies. [Death and Decay] should be placed that enemies need to move through the entire area when it’s used for this purpose.
  • [Chains of Ice] provides a powerful single target snare on cast that has no cooldown. Its main drawbacks are that Runes are limited and that it won’t work too well against large groups of enemies.
  • [Gorefiend’s Grasp]  can be used defensively by grouping mobs onto a target that is outside of melee range from you; grouping targets also makes group damage more efficient while kiting.
  • (Night Fae) [Soulshape] provides an immediate disengage and sustained mobility for the duration.
  • (Venthyr) [Door of Shadows] provides an (almost) immediate disengage if you can survive the cast, but the fact that you can’t Dodge or Parry attacks when channeling it can be problematic.
  • [Death’s Advance] can help counter any snare effects we have and/or help to create a gap between us and what we are kiting. It’s not a strong effect on its own, but it has its uses.
  • [Potion of Psychopomp’s Speed] can get us out of range quickly by giving us a strong movement speed buff. This tool is more limited, but it can be useful at times.
  • [Death Grip] can separate an enemy so we can still efficiently use our abilities while kiting the bulk of an enemy trash pack (as previously explained).

Are there any tweaks we can make to improve our survivability in especially dangerous situations?

There are two methods that can be used to improve your survivability under the right conditions:

  1. If the high damage you are taking is consistent and predictable, you can wait until your health drops much lower than 60% before using [Death Strike] as long as you are confident the next attack will not outright kill you, which can increase the overall efficiency of your [Death Strike].
  2. If you are worried that you could potentially run out of resources in a drawn-out encounter, you could potentially stop using [Death Strike] to avoid overcapping RP and, in an extreme case, stop using runes when your RP is full. This is a VERY conservative strategy that loses long-term resource generation and damage, but it is something that can situationally improve survivability. Realistically, you generally only need to use this tactic for short periods of time to respond to multi-hit tank-buster mechanics that may require multiple [Death Strike]‘s in succession to survive. Even then, such actions should only be necessary in high-end content.

What’s the best way to use empty globals?

Depending on your exact setup, you often have cast time that is not being used when in combat. If you are caught up on spending your resources, sometimes there might be value in strategically delaying your [Death Strike] timings (by just a second or two max) if it allows you to make the [Death Strike] substantially more efficient. This can quickly become a net loss, however, if you are delaying casts enough to waste resources.

What if generating threat is the immediate priority?

The following options should be considered:

  • [Dark Command] or [Death Grip] can be used to instantly have the top threat on single targets and to increase threat generation by 400% over the 3 seconds after cast. The best way to capitalize on this short duration threat amplification is with [Blood Boil], [Death Strike], and (when appropriate) [Marrowrend], our hardest-hitting abilities on individual targets.
  • If we have no immediate solution to the problem, stunning a target we have lost threat on with [Asphyxiate] or snaring them with [Death and Decay] can buy some time.

Defensive Cooldown Priorities

While there is not a hard and fast priority list for defensive cooldowns, there are a few things we want to take into consideration:

  • We generally don’t want to stack defensive cooldowns unless we are trying to mitigate extremely large amounts of damage for very specific purposes, since most defensive cooldowns stack in a way that they mitigate less overall damage when used together. For example, if you use two 50% damage reduction cooldowns at the same time, the second cooldown is only effectively mitigating 25% of the original damage. Even for cooldowns that don’t stack in this way (such as [Anti-Magic Shell]), it is usually more beneficial to spread them out to even out the incoming damage profile.
  • If we have no other considerations, we generally want to prioritize the defensive cooldowns that mitigate the most damage, per second, while they are active; this maximizes overall mitigation. The effectiveness of all defensive cooldowns (while active) are as follows:
  1. [Anti-Magic Shell] – 100% of damage taken (if magic)
  2. [Vampiric Blood]* – ~50%+ of damage taken
  3. [Dancing Rune Weapon] – ~50% of damage taken (if it can be parried)
  4. [Icebound Fortitude] – 30% of damage taken
  5. [Rune Tap]** – 20% of damage taken
  6. [Anti-Magic Zone] – 20% of damage taken (if magic)
  7. [Swarming Mist]*** – ~10%+ of damage taken (it’s complicated)
  8. [Lichborne] – <10% of damage taken (with [Hardened Bones])

* – [Vampiric Blood] is a bit unique, while it functions as a ~26% damage reduction cooldown when we are *just* looking at the increase in healing coming from our healer, it actually does a lot more than that since it increases our own healing by 35%, which is a large portion of our mitigation. The exact strength of this defensive depends on both our character setup and the incoming damage profile.

** – [Rune Tap] should only be used situationally, as explained below.

*** – [Swarming Mist] has unique considerations which are explained in the core priority section. It also has the potential to mitigate much more than 10% of damage taken when the RP generation is taken into consideration.

  • It can be worth prioritizing situational defensive cooldowns if you aren’t always going to be able to make use of them. (This mostly just applies to [Anti-Magic Shell] and [Dancing Rune Weapon].)
  • One alternative perspective to take with your defensive cooldowns is to use them to refill your RP pool when it’s getting low, since defensive cooldowns should reduce the frequency that you need to [Death Strike]. This reactive style can be useful in very difficult content, but can potentially result in unnecessarily low defensive uptime if you aren’t being mindful of when you should proactively use your cooldowns.
  • Because of how [Vampiric Blood] works (it affects our healing received and it increases our health pool by a flat 30%), it’s our only defensive cooldown that can be used *reactively* to high damage taken at no penalty. As long as a situation won’t instantly kill us, it can be useful to save this cooldown for moments where you aren’t sure whether or not you’re *actually* going to take a lot of damage.

Also because of how [Vampiric Blood] works, your current health % will go down by up to 23% when the buff expires (So, if you have 23% health or lower when the buff goes away your health will be reduced to 1). It can be useful to be mindful of this.

  • [Dancing Rune Weapon] has extra value when our RP and/or [Bone Shield] stacks are low, since it increases the generation of these resources. It also increases our threat generation.
  • [Icebound Fortitude] is a stun break and gives stun immunity, which can be an *extremely* useful trait when it can be utilized.
  • [Anti-Magic Shell] can prevent the application of many debuffs, which can greatly increase its value.
  • [Rune Tap] is not used like our other defensive cooldowns, since the RP generation lost by using it would generally mitigate more than [Rune Tap], itself. Instead, it is a defensive cooldown of last resort when you explicitly need EHP.
  • [Anti-Magic Zone] can be used as a personal defensive cooldown, its obvious usage is as a group/part magic damage reduction cooldown.
  • [Lichborne] is a sleep/fear/charm break and immunity, which is extremely useful in some circumstances.

When is [Rune Tap] worth using?

The 5-25 RP lost from using [Rune Tap] over [Heart Strike] generally results in less overall mitigation due to a loss of [Death Strike] healing that outweighs the damage reduction effect of [Rune Tap]. As such, it is recommended that the ability only be used in the following ways:

  1. When you are likely to die without a short-term increase in EHP and more efficient options are not available.
  2. When you have excess runes, you don’t have an attackable melee target, and you are taking damage.
  3. When you are starting an encounter with full runes and an early usage of [Rune Tap] will allow you to maximize your full rune regeneration at least two global cooldowns faster than you would be able to without using [Rune Tap].
  4. With [Crimson Rune Weapon] equipped, using [Rune Tap] immediately after [Dancing Rune Weapon], while you are benefiting from the 40% increase in rune regeneration effect, can be a net gain if you are otherwise unable to keep 3 runes on cooldown and stay below 2 charges of [Blood Boil].
  5. When you have an extended period of time out of combat, the RP you generate with [Rune Tap] will carry over into the next encounter, and the rune regeneration has no other productive use ([Death and Decay] can also be used in this manner).


About the Author


Notify of
Inline Feedbacks
View all comments
Scroll to Top