Dragonflight Retrospectives & War Within Wishlists
Restoration Druid
Augmentation Evoker
BM Hunter
Fire Mage
Protection Paladin
Assassination Rogue
Restoration Shaman
Affliction Warlock
Frost Retrospective & War Within Wishlist
Dragonflight created an upheaval of class mechanics unlike anything seen since Legion. The nature of the changes meant every class and spec saw some level of change. For specializations such as Frost, that had remained stagnant since Legion, this was an opportunity to lay a foundation for the spec’s gameplay for the next era. As Dragonflight draws to a close and significant changes come to a halt, it is time to take a critical look at the current state of Frost and how it can be improved in The War Within. Translating a fixed set of spells into a tree presented unique challenges; some were resolved, while others caused setbacks for the spec.
Fitting two playstyles in one tree
The first challenge for the talent tree design was accommodating two disparate playstyles from previous expansions. Breath of Sindragosa, with a unique playstyle focused on resource management, and Obliteration, concentrated on generating Killing Machine inside Pillar of Frost. On top of this, it also needed to support dual-wield and two-hand weapon types after two-hand Frost returned in Shadowlands. The new talent tree largely met these goals to support multiple build types. Throughout the expansion, both builds shone in particular niches, allowing each to show off its strengths. Obliteration rose to be the dominant build for Mythic+, offering relatively fast burst damage and the ability to drop into any key and route without fuss. Breath of Sindragosa led in raid content, offering sustained cleave without sacrificing boss damage. The niches for each build encouraged those wishing to get the most out of the spec to learn how to play with both builds. Distinct builds also ensured Frost had an alternate build available if one struggled with particular content. However, this design incurred a casualty. After changes in 10.0.7 switched Might of the Frozen Wastes from a talent to a baseline ability, the dual-wield Obliteration builds died off. With two-hand builds no longer paying a talent tax, Shattering Blade was insufficient to keep dual-wield competitive outside of Breath of Sindragosa.
Too many talents and not enough points
The changes in 10.0.7 to move talents to the spec’s baseline abilities highlight a particular problem that plagued Frost throughout Dragonflight: the lack of flexible talent points. The build-defining talents, Breath of Sindragosa and Obliteration, are expensive to reach and leave few points for experimentation. After pathing to one, only five points remain for flexible picks. Frost’s tier sets in Season 2 and Season 3, Lingering Phantom’s Encasement and Risen Nightmare’s Gravemantle, heavily incentivized taking Frostwyrm’s Fury + Absolute Zero and Chill Streak + Enduring Chill respectively. With tier sets taking another two points, Frost only has three talent points to consider swapping to suit the situation. This lack of flexible points often left Frost with stagnant builds, as there were apparent disparities in talent balance within those three remaining points. For example, while Obliteration builds are not required to path through Fatal Fixation to unlock the bottom of the tree, it offers too much throughput to ignore. This results in the trees for Breath of Sindragosa and Obliteration feeling static with little room for experimentation.
While Frost has two distinct builds, finding meaningful variation within them is challenging. The capstones are expensive to unlock, and some of the talent choices are not choices when looking at their throughput. A prime example of this is Frostscythe. While this talent should find a niche in Mythic+ instead of relying on Cleaving Strikes, it simply did not keep up. Those who pick it feel like they are intentionally degrading the effectiveness of their builds. This feeling was not helped by Frostscythe being bugged for most of Dragonflight by not working with Bonegrinder. A similar issue exists on the class tree; the final two rows of the class tree offer significant throughput. The pathing required to take the desired capstones left the class tree static, with only a few points changing to suit the content at best. Needing to sink two points into any of the nodes above a capstone exacerbated this feeling that the tree has too many nodes and too few points.
Talent Taxes
One of the primary contributors to this lack of flexible points is the presence of two-point nodes directly above Breath of Sindragosa and Obliteration. Cold-Blooded Rage is an almost unnoticeable talent. It contributes so little Killing Machine procs that it would go unnoticed if the talent ceased to work. It is even more unnoticed in AoE content, where Obliteration builds will use Glacial Advance as the primary Runic Power spender and not interact with Cold-Blooded Rage. Counterintuitively, a mandatory, two-point talent in the goto build for Mythic+ is deadweight for most of that content. If the interaction also worked for Glacial Advance, it would feel better, but it would still be underwhelming due to the core design being a chance of a chance.
Invigorating Freeze is even worse. It adds a 10% chance to an event that has a chance to happen on a critical strike from one spell. Against a single target, with a 35% critical strike chance, this increases the chance for Frost Fever to grant Runic Power from 10% to 11%. A node with such a low impact feels terrible for a talent node so far down the tree, and it feels like this node is simply a tax to stop Frost from taking too many strong talents. While Cold-Blooded Rage can at least claim to offer some change in rotation by feeding into effects like Murderous Efficiency and Icecap, the same claim does not apply to Invigorating Freeze. These talents feel inadequate and need to feel more impactful for their position on the tree as required nodes to unlock the build-defining capstones.
A third talent tax comes from Avalanche. This one is well worth the point this time. It is so good that it is almost impossible to leave out. The damage it does is dependable, but the real power comes from the free application of Rune of Razorice. Death Knight Runeforges are much more potent than typical weapon enchants, and obtaining one without a weapon to hold it is something Death Knights will do whenever given a chance. Because Runeforges are so powerful, adding Rune of Razorice to Avalanche was likely done to offset the penalty two-hand builds incur by not getting a second Runeforge. However, the result is that any build that forgoes Avalanche gives up an extra Runeforge. This means Frost will always have less flexibility to spend points below the 20 gate.
Obliteration of Tomorrow
Going into The War Within, one badly needed change stands out for Obliteration. Cleaving Strikes needs to go away for Frost. Tying Obliterate, the primary damage-dealing ability, to a ground effect with positioning requirements is simply not fun. DPS cannot control where mobs stand in group content, and Mythic+ players are all too familiar with the pain of placing Death and Decay and then watching the tank decide to drag the pack far away. Enabling Obliterate to cleave also overshadows the rest of the AoE kit, making many feel that other buttons are not impactful.
Historically, Frost relied on its spells in the frost school to deal AoE damage, buffed by Mastery: Frozen Heart. However, Dragonflight introduced Frostreaper and stacked more multipliers for Obliterate with Frigid Executioner, Improved Obliterate, and Might of the Frozen Wastes. Applying all these modifiers to a guaranteed critical hit amplifies Obliterate so much that it feels like the only button that matters, even in AoE. By comparison, Howling Blast only has Improved Rime and Avalanche in most situations, with Icebreaker often out of reach for Obliteration. The tree overloading Obliterate with modifiers pushes down the importance of the rest of the core spells. Many Frost players want their other abilities to feel more impactful when used correctly.
Another common complaint is that Icecap makes the spec much less resilient against downtime. Granting cooldown reduction to Pillar of Frost, which then generates more cooldown reduction with Obliteration, introduces high consequences for forced downtime. A choice node that turns Icecap into a flat amount of passive cooldown reduction is one way to address the penalty for forced downtime. And, of course, forced downtime means needing to restack Unleashed Frenzy and Icy Talons. With Breath of Sindragosa demanding no downtime, Frost should have an alternative build that offers better resilience against downtime.
Don’t Hold Your Breath
Breath of Sindragosa is currently defined by a single problem: absurdly high uptime requirements. In past expansions, maintaining Breath of Sindragosa for 60 seconds was fantastic. In Dragonflight, 60 seconds is considered poor. Resource-generating talents used to be mutually exclusive; now, Frost can take as many of them as it wants. Where Frost previously had to pick one out of Runic Attenuation, Murderous Efficiency, and Horn of Winter, it can now have them all. Piling on more resources, Anti-Magic Barrier is a new talent that makes Anti-Magic Shell a potent resource cooldown. It is now expected to see 80-90% uptime on Breath of Sindragosa in raid content. Sustaining this uptime makes Breath of Sindragosa a high-risk spec, but the reward does not match.
Being tuned around this high uptime also made Breath of Sindragosa drop out of favor for Mythic+. Achieving good uptime requires a coordinated group, and the result is not impressive when accomplished. It does not deal enough damage to warrant the risk involved with Breath of Sindragosa. Most Frost players are better off not trying Breath of Sindragosa in Mythic+.
In 10.0.7, an attempt was made to curb Breath of Sindragosa by raising the Runic Power cost from 16 Runic Power per tick to 18, and the spell damage was buffed to compensate. Unfortunately, this ended up failing to have a significant impact on curbing uptime and ultimately only made the build more punishing. With Breath of Sindragosa now doing a larger share of damage, lost uptime hurts more. The same changes also enabled Frost to spend more Killing Machine procs with the new talent, Fatal Fixation. This introduced more Runes through Murderous Efficiency, and the Season 3 set bonus added more resources through direct Runes and more uses of Empower Rune Weapon. Any initial disruption to Frost’s resource economy is long gone. If Breath of Sindragosa will recover from falling to niche spec status in The War Within, it must revert to its previous cost, and the resource economy requires rebalancing.
What is a Frost Death Knight?
Looking at each build, both share the resource issue in common. Obliteration spams Obliterate because it constantly has Runes available. The abundance of resources and the myriad of effects tied to Obliterate crowds out the Runic Power spenders and makes the spec feel one-dimensional. Breath of Sindragosa lasts for far too long, and being tuned around very high uptime causes it to fall far behind in content that does not support the constant uptime. Resource management has long been a trademark of Death Knight gameplay, but in Dragonflight, it feels non-existent. Frost has all the resources it could ever want, so there is no need to manage them. Frost has no identity to stand out from other builder/spender melee specs. The core gameplay for each spec can be accurately described as just using Obliterate and Remorseless Winter as much as possible.
The lack of meaningful utility is on the other side of the identity problem. It is often repeated that because Death Knight has a monopoly on Death Grip, it should not have other forms of utility like dispels, cleanses, or a raid buff. Death Grip is how Death Knight will find its place in groups. However, grips are proving to be forgettable. Fyrakk’s second phase demands controlling enemy positions so the friendly adds can walk through safely. However, when guilds first got to Fyrakk, they used old Engineering items to gain additional knockbacks rather than bring more Death Knights. While it is true that no other class can precisely relocate a mob like a Death Knight, what good is that when the content does not require it? We are seeing guilds realize that getting the mobs moved matters more than precisely controlling where they land. In Mythic+, grips are rightfully even less necessary, as five-player content needs to allow a wide range of compositions. What good is it to be the sole provider of Death Grip if the content does not let Death Knights shine in that role?
The lack of utility is felt even more in Mythic+, where Death Knights can often not meaningfully help with affixes. Although Death Grip with Death’s Echo gives Death Knights more opportunities to stop casts, there are some affixes Death Knights cannot handle and instead must rely on the rest of the group to solve them. There is nothing to help with Raging or to dispel Afflicted without spending 3 GCDs on Death Coil. The most unique utilities are Raise Ally, something many classes can do, and grips out of Sanguine. However, this is another case where precise positioning matters less than relocating the mobs, which multiple classes can do.
Closing
Without a compelling gameplay hook, either through how damage is dealt or through how Death Knights impact the group, Frost’s population is shrinking. This is unfortunate because the spec has a solid foundation. It successfully offers two viable playstyles, but the lines are blurred when examining the differences in the gameplay hooks, and most of the distinction comes from how the spec punishes players for mistakes. Some key changes to pull apart the playstyles and decisively separate them will breathe new life into the spec. Obliteration as the build with a sustained profile and an emphasis on Rune management. Breath of Sindragosa as the build with a decisive burst phase that rewards carefully planned cooldowns. The pieces are there for Frost to be a well-designed spec in The War Within, but those pieces need to be picked up and polished instead of left to languish.
About the Author
This guide is written by Khazak, a Death Knight player since 5.4. Besides writing guides, I am a moderator in Acherus, the Death Knight class Discord and maintainer on SimC, contributing code and APL for Death Knights. I raid in Agency on Zul’jin US and sometimes stream raids on Twitch. I am always happy to take questions or comments about the guide through Discord.
