Unholy Leads the 95th Percentile; Arms Leads the Broader Plot
The leading spec changes with the statistical view. Unholy is first in both 95th-percentile charts, while Arms begins the all-percentile overall distribution. This is the historical Dragonflight Season 4 Week 7 snapshot for Awakened Mythic Vault of the Incarnates during the week of June 4, 2024.
Anshlun’s June 11 report also carries a material logging warning: player attribution for Akaari’s Soul, the shadow-damage component of Secret Technique, was missing. The source estimates that component at roughly 15% of Subtlety’s damage. Its last-place result needs that caveat; the estimate does not supply a universally corrected rank.
The Full Table and the Separate Boss-Damage Order
The original captures select one week, Awakened, Mythic, all bosses and all item levels, with normalized scores. Overall position, population and movement preserve the source’s typed 95th-percentile table. Boss position is read independently from its boss-only image.
The movement comparison is the previous week’s Amirdrassil order. It is not a same-raid tuning experiment. Population is the source’s recorded sample rather than a verified number of unique raiders, and rank change counts positions rather than a percentage of damage.
| Overall position | Specialization | Source population | Change vs prior Amirdrassil | Boss-damage position |
|---|---|---|---|---|
| 1 | Unholy Death Knight | 6,483 | +2 | 1 |
| 2 | Devastation Evoker | 9,136 | 0 | 3 |
| 3 | Marksmanship Hunter | 2,377 | +8 | 19 |
| 4 | Arms Warrior | 11,442 | 0 | 4 |
| 5 | Frost Death Knight | 1,383 | +5 | 7 |
| 6 | Assassination Rogue | 5,062 | −5 | 2 |
| 7 | Shadow Priest | 14,208 | −2 | 14 |
| 8 | Windwalker Monk | 6,008 | −1 | 21 |
| 9 | Retribution Paladin | 22,276 | −3 | 5 |
| 10 | Havoc Demon Hunter | 12,656 | −2 | 20 |
| 11 | Arcane Mage | 1,910 | −2 | 6 |
| 12 | Frost Mage | 9,703 | 0 | 10 |
| 13 | Demonology Warlock | 8,863 | +4 | 22 |
| 14 | Enhancement Shaman | 6,947 | +6 | 12 |
| 15 | Beast Mastery Hunter | 19,275 | +3 | 8 |
| 16 | Fury Warrior | 4,202 | +3 | 24 |
| 17 | Balance Druid | 8,253 | −3 | 11 |
| 18 | Outlaw Rogue | 3,891 | +3 | 23 |
| 19 | Destruction Warlock | 8,514 | +3 | 18 |
| 20 | Elemental Shaman | 3,314 | −5 | 17 |
| 21 | Fire Mage | 6,172 | +2 | 13 |
| 22 | Feral Druid | 3,249 | −6 | 9 |
| 23 | Affliction Warlock | 1,151 | −10 | 16 |
| 24 | Survival Hunter | 672 | +1 | 15 |
| 25 | Augmentation Evoker | 7,284 | −1 | 25 |
| 26 | Subtlety Rogue | 2,119 | 0 | 26 |
Marksmanship gains eight places, Enhancement six and Frost Death Knight five. Affliction loses ten, Feral six and Assassination five. Devastation, Arms, Frost Mage and Subtlety keep their numbered positions despite the change of raid.
Retribution has the largest reported population at 22,276, followed by Beast Mastery at 19,275 and Shadow at 14,208. Unholy’s sample is 6,483; Devastation’s is 9,136. A population column does not measure the share of logs optimized through funneling, so it cannot establish that one result is artificial and another broadly representative.
Marksmanship’s Third Overall Is Nineteenth on Bosses
The boss-only top five are Unholy, Assassination, Devastation, Arms and Retribution. Marksmanship is nineteenth, compared with third overall. Feral is ninth on bosses and twenty-second overall; Arcane is sixth on bosses and eleventh overall.

Those differences identify a question for encounter review: which targets mattered to the group’s success? They do not identify a talent interaction or a player’s assignment without inspecting the underlying fight. An overall lead and a boss-damage lead answer different questions.
Arms Moves First When the Percentile View Changes
The broader overall plot begins with Arms, Unholy, Devastation, Marksmanship and Assassination. Frost Mage appears seventeenth in its displayed order, compared with twelfth in the 95th-percentile table. Fury is thirteenth in the broader plot and sixteenth in the table.

The score ranges differ between these figures, and their distributions overlap. A 95th-percentile aggregate is not a measured ceiling, while the all-percentile plot is not an individual player’s expected DPS. Preserve the same target and percentile when comparing dates.
Read the Logging Warning Without Inventing a Repair Formula
The original report describes a recent attribution failure. It does not establish an exact introduction date, identify a fixed damage correction for every build, or provide a replacement position for Subtlety. Adding fifteen percentage points to a normalized aggregate would not reconstruct the missing event-level contribution.
Other explanations need their own evidence too. The table does not demonstrate that all groups had completed tier sets, that Unholy’s sample was dominated by coordinated parse runs, or that Devastation’s position came from a particular tier-scaling interaction.
For an earlier visit to the same raid, see the Awakened Vault Week 4 reference. That comparison keeps the raid constant, but still changes the date, sampled logs and progression context.




