Dragonflight Retrospectives & War Within Wishlists
Frost DK
Restoration Druid
Augmentation Evoker
BM Hunter
Fire Mage
Protection Paladin
Assassination Rogue
Elemental ShamanRestoration Shaman
Affliction Warlock
Subtlety Rogue Dragonflight Retrospective
Intro
Dragonflight set out with a difficult mission to revamp the talent tree design and change each specialization in the game. Now that Dragonflight is slowly phasing out, and in preparation of The War Within, I am happy to sit down and take a closer look at the progress made this expansion, talk about things that improved, and discuss what went wrong. So lets start!
What happened during Dragonflight?
Overview
If you like Subtlety Rogue, Dragonflight was a good time. The spec ended up in a favorable position in all 3 raid tiers. While Subtlety was usually tuned to be weaker than the other two specializations, it managed to usually get ahead after the other two received nerfs. The list below briefly summarizes all three seasons.
- Season 1: The talent tree design allowed for combinations with low trade-offs between single target and aoe talents. Players could take full advantage of fights like Raszageth to do relevant damage at all stages of the fight.
- Season 2: Season 2 started with changes to talents empowering magic damage sources. These would stop working on item effects. Subtlety being one of the dominant users of borrowed power systems like Onyx Annulet got a small compensation buff. The compensation increased the baseline performance and in combination with a beneficial bug, managed to make the spec one of the strongest performers for the first weeks of the tier. The spec fell off long-term due to the weak T30 Four-Set Bonus and the low benefits from haste. Haste was a key element of the Scalecommander Sarkareth encounter (final boss of that tier).
- Season 3: Season 3 shipped with a talent tree rework. More damage was put into cooldowns, while the damage outside of cooldowns became almost negligible. The redesign let us reach certain talents more easily. The most noteworthy one wass Invigorating Shadowdust. This talent allows Subtlety to adjust more dynamically to encounters and is the key element for the outstanding performance in the Smolderon encounter. The raid design almost seems to work too well with this redesign, with 3 bosses offering damage amplification.
The Evolution of Talents
Talents
Rogue was one of the earlier classes to preview the new talent tree. Getting your tree early is great but also comes with some problems.
The design direction on the trees was less streamlined and more unique ideas ended up being explored. Both the Rogue and Subtlety trees had talent points with decisions between utility and damage nodes. Pathing was also more restrictive, and made specific points feel taxing to reach because you were forced to take talents you might not need.
Season 1 had, as a consequence of this, just one talent build that people played; it was the talent build usually referenced with the capstone talent Dark Brew.
Season 2 tried to change talent choice with the new tier set. The Two-Set Bonus empowered weak talents on the left side of the talent tree, which ended up leading to a selection of three unique talent options, all within a low range of around 2% dps difference.
Talent Builds – Season 1 and 2:
Season 3 reworked the talent tree to fix some of the issues. Decisions between utility and damage were reduced or removed, and traversing in the tree allowed for more fine control over specific talent options.
Talent Builds – Season 3:
Overall, I am very happy with the direction set in the update and think the development team did a great job of evolving the specialization. That said, there are still some areas with rough edges. I will cover some of the issues in the following sections.
Talent Diversity (or the Lack Thereof)
The first iteration of the talent tree had a lot of talents not specifically designed with Subtlety in mind. Many talents ended up being Azerite traits, old Legendary effects, or Conduits, often without interesting hooks or very niche use cases. To make matters worse, the change in spec design in the rework made some previously good options bad. This lead to a situation in which many of the talents ended up always being picked, not because they are absolutely mandatory but because of a lack of good alternatives.
Talent choice:
I will only highlight some talents I consider outdated to keep this section short.
Sepsis – Previously a Covenant ability, even with Dark Brew, it’s fairly weak and complicated to play. It adds another cooldown to play around and track for little reward, and in general does not fit well in the gameplay rotation.
Lingering Shadow,The Rotten,Perforated Veins and Gloomblade – All talents that empower combo point builders. Most of them have special requirements to trigger, which does not work well with most of the specs focused on finisher damage. Most of the options also have additional conditions, making them hard to align well. Gloomblade also stands out as only a minimal improvement over Backstab, with both buttons designed as weak filler spells.
Shadow Focus, Master of Shadows and The First Dance – There are too many resource-generating talents in the tree currently, while resource management is already solved by different nodes.
Shuriken Tornado – This spell was always a bit problematic; it has no direct feedback but is delayed and is very time sensitive, so easy to play incorrectly. It additionally does not work well with the reworked Shadow Blades and the new talent, Shadowcraft, because it can lead to lost combo points.
To end this section on a more positive node, Danse Macabre, Shadow Techniques, Shadowcraft, and Invigorating Shadowdust are all well-designed talents adding room for creativity.
Gameplay
Complex by design
To start this off, a simple question: how difficult should a class or specialization be? There is no correct answer to this question but Subtlety definitely takes the crown in terms of difficulty this season.
The rework added even more interactions to spells and almost every spell on your action bars needs explanation to be understood. We are at the point where learning the spec became more than just understanding the core gameplay loop with small adjustments. Each encounter can have very different cooldown uses, which requires that we use spreadsheets and external websites as part of the learning process to give new people the option to get to reasonable performance levels quickly. A lot of this complexity comes from one single talent: Invigorating Shadowdust. Without a Trace adds a second charge to Vanish and enables even more freedom to decide on the use.
Why this is, can be easily explained by looking at differences in cooldown timings.
It becomes obvious, looking at the table, that cooldowns have different timers which don’t line up well naturally. This wouldn’t be an issue on its own, but becomes complicated when you start to get cooldowns ready by using Vanish, and can adjust to encounter specifics. Smolderon is the perfect example for this, the fight has very specific timing, and playing your cooldowns correctly allows Subtlety to be the only spec in the game to have cooldowns perfectly aligned with all vulnerability phases. It allows for the spec to stand out on encounters with vulnerability phases.
Overall, this is exciting. It gives Subtlety a unique identity, but comes with trade-off of a higher difficulty curve. Tuning and raid design allowed this to happen this tier, but difficulty could quickly become frustrating when tuning is worse.
Weaknesses
Reworking the damage profile to be more bursty came with some trade-offs. The already only mediocre low-target cleave damage became even worse, and lost its strong sustained AoE damage. Both only small niches in raiding, with low target cleave being the more common scenario; typically one fight per raid (Council of Dreams, Forgotten Experiments, Prime Council) this expansion. This being said, i think these trade-offs are fair, despite the weak-ish AoE. The unusual AoE design, in which using the AoE finisher Black Powder essentially spreads damage equally among targets, and using Eviscerate instead becomes a massive loss in AoE output.
Bugs
It was probably obvious, but the spec had positive and also negative/annoying gameplay bugs for long periods in Dragonflight. Some of the bugs ended up being turned into gameplay mechanics. Cold Blood initially worked on the Shadow Clone parts of Secret Technique, without consuming them. The consequence was a gameplay loop in which Cold Blood was used between Secret Technique and the next spell attack. The development team saw this and changed Cold Blood to make the entirety of Secret Technique critical strike, simplifying the gameplay around it without taking away the benefits.
We ended up having some nasty bugs too. Secret Technique ended up losing more than 2/3 of its damage when standing too far away from the target (bug report) which got finally fixed in 10.2.
There are still around twenty smaller bugs and minor annoyances left on the bug tracker, but to give credit to the dev team, a lot of inconsistencies and major bugs ended up being fixed in the 10.2 cycle.
Tier Sets
Tier sets ended up very hit-and-miss this expansion. The table below summarizes my thoughts quickly.
Tier set | Notes |
Season 1 – 2-Set, 4-Set | The season 1 tier set was boring but this was perfect for the first raid tier. |
Season 2 – 2-Set, 4-Set | The season 2 tier focused too much on making unwanted talents relevant. The two piece bonus did not use the base Symbols of Death duration for pandemic timers, but had its own. This made using Symbols of Death less intuitive.
A more intuitive solution would have been to simply extend Symbols of Death instead. The four piece bonus was also heavily undertuned and only used because of higher gear levels. |
Season 3 – 2-Set, 4-Set | The season 3 tier set ended up strong. It was a big step up from the previous tier sets, even though it has some unintuitive elements. It emphasized the strengths of the spec and mostly empowered finishing moves. The additional Combo Point generation made up for the lower resource income from the talent tree. It will be interesting to see how well the spec plays once it’s removed. |
Conclusion
Subtlety got better over time but kept some of its problems. The talent tree still has a fairly high number of weak or unwanted talent nodes, resulting in a reduced amount of choice.
The season 3 rework added a lot of complexity to Subtlety which does not always seem justified. Cooldowns also don’t align well naturally, and things get complicated if you add Invigorating Shadowdust, resulting in a higher learning curve. The package is still unique, and allows for a fun gameplay experience. One of the biggest open questions is how much the weaknesses matter in the long term because tuning currently makes this trade-off reasonable.
Wishlist for The War Within
Invigorating Shadowdust is currently a fairly dominant talent choice, while it is possible to skip it, the trade-off is rarely worth it. The wishlist for the new expansion would be to iterate more on the talent tree. There are a lot of outdated or weak talent choices that could be updated or replaced.
It would be nice to lean more into the shadow damage fantasy, and maybe return some thematic effects like Banshee’s Blight. Gameplay could also be improved in multi-target situations, especially to make the trade-offs between Eviscerate and Secret Technique worth more. Cooldown timings could also be simplified. One idea would be to put Flagellation and Shadow Blades on the same cooldown. Subtlety is currently very hard so simplifying some small parts of it could be beneficial in the long term.
About the Author
This article is currently updated and maintained by Fuu. A Rogue Theorycrafter since Legion with the main focus on Subtlety. I am also part of the Rogue Discord as a moderator and you will usually find me discussing rotation, strategies, or talking about theories to improve the simulation.
If you want to chat about rogue feel free to contact me on discord under .fuuu.