- Author: Kyrasis
- Date: February 19, 2022
- Updated: January 20, 2024
- Expansion: World of Warcraft
Welcome to the Advanced Blood Death Knight Guide for M+ in Patch 10.2.5! The purpose of this guide is to provide accurate, in-depth BDK information specifically geared towards players progressing in M+. All information provided herein is based on a combination of M+ log data analysis, simulationcraft analysis, BDK mitigation model analysis, in-game testing, and expert opinion. While this guide was primarily made with +18 and higher key levels in mind, the recommendations (particularly those concerned with playstyle) can also be used for success in key levels below this point, though a small number of optimization recommendations may be different if particularly low gear level, poor ability usage, or low key levels were taken into consideration. It is a comprehensive guide that gives readers a chance to understand not only *what* decisions can improve their play, but *why* these decisions are made, so feel free to skip to whatever sections are relevant for you!
This guide was created for BDKs (Blood Death Knights) progressing in challenging M+ (Mythic+) dungeons (+18’s and above); some recommendations presented herein were not made for use in raid or PvP scenarios, which can have different priorities and considerations. Furthermore, it assumes the user is capable of playing a BDK reasonably well and that they are able to obtain reasonably competitive gear (so, items of +18 end-of-dungeon reward ilvl and up; not questing greens). Furthermore, the use of all tier set bonuses is generally assumed.
All recommendations reflect the current state of 10.2.5 as accurately as possible and if significant emergent changes occur the guide will be updated to reflect those changes as soon as possible.
If you have any specific questions, comments, feedback, or concerns regarding this guide, please message discord tag kyrasis as your first line of support. I’m always happy to answer questions about BDKs in M+ and I’m generally available to perform log reviews!
About the Author
I’m Kyrasis, I’ve been primarily doing a massive amount of the math-heavy theorycrafting on BDKs since Legion; I was also the #1 World Ranked M+ BDK for both Season 2 of Dragonflight and Season 4 of Battle for Azeroth on raider.io. Additionally, I personally maintain the BDK mitigation model that is primarily used to evaluate mitigation on BDK in various situations along with maintaining this advanced guide and occasional Youtube content when I have time to spare. Again, I can be contacted through the discord tag kyrasis or through the discord server if you have any feedback or questions!
Specifically, I’d like to thank the following:
- Thorlefulz for continuous theorycrafting support activities as well as providing some independent statistical analysis and general guide review.
- Arma for his work on the BDK simcraft module, his independent testing/theorycrafting work, and review comments.
- Bicepspump for playing an important role in the development of a BDK Death Strike log analysis tool and providing a methodology for doing large-scale log analysis.
- Mythie for helping to independently validate core portions of the BDK mitigation model.
- Ellychan, Cele, Naed, Yoda, Terra, Dreams, Baconmumbles, and Brewseph for review comments and/or playtesting feedback.
- Druchaon, Dorki, Trell, FloorFruits, and Darkmech for historical contributions.
- Bees, Obadyahu, Cambam, Горсун, Riles, MrPottiez, Zivvy, HiddenoO and LGhost for contributing valuable feedback that led to modifications or additions.
- All other users who provided support!
The initial sections overview some more basic information before moving onto specific recommendations in future sections. The Tank Performance Measures section, in particular, lays the groundwork for how character optimization recommendations are approached later in the guide.
- Maximum Runes: 6 Runes
- Base Rune Cooldown Time: 10 seconds
- Base Runic Power Generated Per Rune Used: 10 RP (Runic Power)
Our primary ability resource that passively regenerates. All runes have individual cooldowns, but only 3 runes can progress their cooldown at a time. The base rune cooldown is reduced by haste. In general, 10 RP (Runic Power) is generated per each rune used, but there are exceptions.
Base Maximum: 125
RP is our secondary ability resource that is primarily generated by spending runes. The main use of this resource is to spend it on [Death Strike] to generate health. Alternatively, it can be spent on [Bonestorm] (when talented), which also generates health. RP is effectively a second health bar and needs to be mindfully managed for BDK success. Note, 1.25 RP is lost every second when out of combat and RP is lowered to 20 when an M+ dungeon starts.
Maximum Stacks: 10
[Bone Shield] is a stack-based mitigation buff where stacks are generated with [Marrowrend] and lost when damaged by melee attacks (at max you can lose one charge every 2.5 seconds). A significant amount of passive mitigation is lost when this buff falls off completely and [Ossuary] provides a [Death Strike] RP cost reduction when you have at least 5 stacks; other than that, the number of [Bone Shield] stacks do not provide any additional bonuses outside of talent interactions like [Foul Bulwark]. Generating new stacks of [Bone Shield] is done at the cost of RP generation, since [Marrowrend] does not generate RP as efficiently as [Heart Strike]. [Bone Shield] stacks are reset when a M+ dungeon starts, so there is no value in generating stacks ahead of starting a key.
[Bone Shield] is a stack-based mitigation buff where stacks are generated with [Marrowrend] (or [Death’s Caress]) and lost when damaged by melee attacks (at max you can lose one charge every 2.5 seconds). A significant amount of passive mitigation is lost when this buff falls off completely and [Ossuary] provides a [Death Strike] RP cost reduction when you have at least 5 stacks; other than that, the number of [Bone Shield] stacks do not provide any additional bonuses outside of talent interactions like [Foul Bulwark]. Generating new stacks of [Bone Shield] is done at the cost of RP generation, since [Marrowrend] (and [Death’s Caress]) do not generate RP as efficiently as [Heart Strike]. [Bone Shield] stacks do not reset when a M+ dungeon starts, so there is potentially some value in generating stacks ahead of starting a key.
Class Design Implications
- BDKs take significantly more up-front damage than other tanks, since we can mitigate significant amounts of this damage using reactive healing from [Death Strike]. As a consequence, BDKs are usually more vulnerable to dying from burst damage than other tanks, since a large portion of our mitigation happens *after* we take the damage.
- More so than any other tank, a BDK’s health, alone, is not an indication of how vulnerable they are to dying, since our RP functions as a second health bar. This generally means that healers have a more difficult time efficiently healing BDKs, since healers need to track an extra resource to heal BDKs effectively and there is potential for both players to overlap their healing.
- BDKs are the most vulnerable at the start of pulls, when their [Bone Shield] may be at low or no stacks, when their RP may be low or empty, and where their [Death Strike] will not be as impactful during the first few seconds of a pull (since healing is proportional to damage taken in the last 5 seconds). Attempt to refresh [Bone Shield] and to top off RP at the end of any pull, so you can have these resources available at the start of the next one, which can help with this vulnerability. For much the same reasons, BDKs have a lot of difficulty recovering after they have died and have been resurrected.
Tank Performance Measures
While damage roles are almost exclusively focused on their damage output, tanks have several measures that can affect them. The main four are as follows:
- Threat Generation – A tank cannot do their job if enemies are attacking other people. Without enough threat, you risk one of your group members dying to whatever you are not tanking. Threat generation is related to the amount of damage you deal, though taunt effects can be used to instantly gain the highest amount of threat on a target and to heavily amplify threat generation during the taunt duration.
- Effective Hit Points – EHP (Effective Hit Points) as used in this guide is your normal health corrected for any reliable, or controllable, upfront mitigation. The primary concern with EHP is that it is high enough so that you or your healer have time to reactively heal incoming damage you are taking before you die. Particularly on more difficult content, more EHP will reduce the likelihood of dying to burst damage, while also allowing tanks to mitigate damage more efficiently (and allow healers to heal the tank more efficiently) due to the tank having a larger range of health where they are safe from imminent death.
- Mitigation – As it is used in this guide, mitigation refers to the amount of raw damage that a tank can reduce with proactive mitigation (such as armor) *and* reactive mitigation (such as healing from [Death Strike]). Simply put, more mitigation means that less healing is required by healers for the tank to survive.
- Damage – While tank damage output is lower than that of damage dealing roles, tank damage still has some impact in dungeons. Strict damage requirements in M+ dungeons aren’t as commonly seen as they are in raid encounters, but more personal damage will result in faster dungeon times assuming nothing else changes.
What should BDKs be looking to maximize when progressing into higher M+ Dungeons?
A large part of BDK optimization in M+ comes down to finding the best blend of performance stats. We are not necessarily maximizing any one trait. Instead, we are trying to find the setup that will maximize our chances of timing a key, which each of the above metrics contribute to in some capacity, though how they weigh against each other can change given the situation. Furthermore, some tradeoffs can be more one-sided than others, where the *amount* of benefit being provided can outweigh the *type* of benefit being provided.
That being said, you should generally be placing a relatively high value on Mitigation as a starting point, unless a tradeoff is particularly favorable to EHP or Damage. As content starts to get more difficult and incoming damage becomes more bursty, EHP becomes a secondary consideration in addition to Mitigation to help improve our overall survivability. Personal damage can always provide increased clear speed potential for a given route, though time generally only becomes limiting near the maximum key level your group composition is capable of doing with a given route and making routing improvements is often the more efficient method of improving dungeon timers outside of very specific situations.
Tyrannical keys place relatively less of an emphasis on Mitigation and EHP than Fortified keys, since trash is often more threatening to the tank’s personal survivability than bosses. Furthermore, Tyrannical timers are usually (but not always) harder to beat. As such, Tyrannical keys can sometimes place a greater priority on personal damage output than what is normally the case. Likewise, specific dungeon characteristics can also be different enough, at times, to justify different options on a per-dungeon basis.
Threat problems, when they arise, should be considered when they are a noticeable cause of group deaths and/or group damage throttling. In particular, threat is usually more of an issue on high target count pulls and near the start of those pulls. That being said, options that noticeably improve threat generation in these situations are extremely limited and threat issues have been noticeably less prevalent in the Dragonflight expansion (probably because of [Shattering Bone]).
A partial solution to threat generation issues involves pooling abilities that provide significant upfront AoE (Area of Effect) damage for the start of any problematic pull, such as [Dancing Rune Weapon], [Abomination Limb], [Tombstone] with [Shattering Bone], or making sure to use [Death and Decay] while walking into a multiple target pull. Heavier-handed solutions include using [Bonestorm], though this is generally less desirable for other reasons.
What’s the basis for this?
Mitigation is the primary way to increase our overall survivability, which increases our group’s pull potential (which can decrease the total number of pulls in an instance) and provides increased group damage from our healer and damage dealers, since we need less healing and are less likely to need to resort to kiting (so enemies are more grouped up and take more damage from AoE abilities). Until incoming damage profiles become extremely bursty and group damage requirements become very demanding (bringing in the need for more EHP), mitigation is basically the only metric that’s going to have a significant impact on your M+ success rates (from an upfront character optimization perspective). Even in higher keys when other metrics become considerations, mitigation will still be very influential to your success.
As for the other measures:
- Threat generation past the point of solving threat issues has no value, though it is very important to address significant threat issues should they arise and if a reasonable solution is available.
- EHP becomes increasingly valuable as you move into mid-to-high level keys, where your health pool size starts to limit your [Death Strike] efficiency and snap deaths become a concern. If you think you are at the point where EHP begins to have value, you probably are. Just be careful about overcommitment to EHP at the cost of too much mitigation, because EHP won’t help you survive deaths caused by a lack of mitigation throughput. Because of its inherent value, EHP options are generally recommended when the tradeoff is reasonably efficient. The Build Modifications section provides additional information if you would like to deviate from standard recommendations regarding EHP levels.
- Damage is always nice and can potentially lead to faster dungeon completion times, but sacrificing significant Mitigation or EHP for damage in progression keys can also have the potential to lead to slower dungeon completion times by hindering the rest of your group’s damage through increased healer attention or more frequent kiting or by increasing the chance of group deaths. Usually, higher cost damage-oriented trade-offs should be made only when you genuinely believe you are in a situation where it will result in higher overall key success rates, such as when more tank damage can make a high-end key possible to time in the first place or if you are worried about a boss with specific types of damage check mechanics. Overemphasizing personal damage output is a common mistake for new tank players in M+ and, given these circumstances, damage options are generally recommended only when the tradeoff is exceptionally efficient (with some extra leeway on Tyrannical weeks). The Build Modifications section provides additional information if you would like to deviate from standard recommendations regarding damage levels.
What’s the best way to estimate these performance measures?
Disclaimer: Again, tank optimization is often more complicated than simply maximizing a single performance measure, so take care when leaning too heavily on any single measure. For example, for a large number of previous patches, simulating damage to determine gear upgrades for tanks would lead to worse outcomes than randomly selecting your highest ilvl piece of gear for a given slot. With that in mind, here is how you can measure the individual measures.
Today, damage measurements are easier than ever with tools such as raidbots.com and simulationcraft. When performing damage evaluations for dungeons, it is suggested to look at the results of long duration (~8 minute) sims with fixed target counts and continuously melee attacking enemies (this requires a manual override) to determine what works best for single target and multiple target situations. Even if the default action priority list in simulationcraft for BDK isn’t perfect, it should generally be good enough for basic damage comparisons. Hectic add cleave or dungeon slice variants are not recommended *regardless* of action priority list quality since overall damage done in that scenario is not equivalent to “effective damage done” (which reflects time saved in a dungeon).
Note: Damage Measurements in Dungeons
There is a big difference between total damage and and total *effective* damage done in a dungeon, since the *real* thing we care about with damage is how much damage done affects the *overall dungeon completion time*. So, keep the following in mind:
First, damage meters are biased towards multi-target pulls. When making damage decisions, how much faster you clear a dungeon is determined by how much you can increase your group’s damage relative to the damage they are already doing in any given circumstance. So, you need to provide a *much* greater increase in raw DPS (damage per second) in a 10-target encounter to reduce dungeon completion time by one second than you would need to do in a similar length 1-target encounter. Damage meters are not taking this into consideration
Second, large gaps in health pools among enemies can mean that damage on low-health targets is *meaningless*, or even counterproductive, to overall dungeon completion time if they are going to die to nothing but passive cleave damage. So, damage to these targets will not contribute to faster dungeon completion times, at all, even though the damage dealt to them shows up on a damage meter.
All in all, this generally means that it’s easy to undervalue single target damage relative to AoE damage in M+.
Dungeon route sims have the potential to side-step a lot of the above issues, though, given the multiple metrics we need to keep in mind as a tank, that level of refinement is hard to make use of and adds a number of complications.
Mitigation is more complicated. At this time, the best known tool for BDK’s to measure Mitigation is the BDK mitigation model I maintain. It can be difficult for new users to determine what inputs are best to use for this tool and the output is going to depend heavily on the inputs, but it is recommended to model higher than average damage intakes that are representative of the most dangerous situations you encounter in M+ Dungeons (the ones that are most likely to potentially kill you). Using too low of an incoming damage input is a bigger problem than using unrealistically high numbers, since most significant forms of mitigation fully scale with incoming damage and you are unlikely to die during low damage intake periods.
EHP can be measured by the above model or by simply applying passive damage reduction effects to your health total.
That being said, the most optimal setups rely on all three of these metrics in some proportion, and because comprehensive evaluation is burdensome I have taken the effort to develop recommendations and rules of thumb that can lead players to optimal outcomes without them having to personally go through the grunt work involved. Granted, more players digging into the theorycrafting is always welcome!