The Blinding Vale

Arbiter charges Shadow Priest 5,124,626 of "over-exposure" -- damage comparable players did not take -- and that is most of why his mechanics score is 54 and his grade a D. The damage is real. The reading is not.

Bloodthorn Roots carries two spell ids: 1259365 is the debuff, 1237330 is the damage. Split them apart and the applications are almost identical across the party:

  player         roots applied   damage ticks     damage
  Shadow Priest                      6             57  6,085,493
  Devourer DH                      6             44  3,836,645
  Prot Warrior                   6             35  1,644,886
  Ele Shaman                6             13  1,486,446
  Hyporock                   5              5    572,912

Of the 154 damage ticks in the key, 154 landed on a player who was carrying the debuff at that instant and 0 landed on anyone who was not. So the tick count is not a measure of where somebody stood -- it is a measure of how long the root stayed on them. Shadow Priest was rooted the same six times as everyone else and was cleansed once. Hyporock was rooted five times, cleansed himself twice, and ate five ticks.

Read the mechanics score with that in front of you. The axis is doing exactly what it is designed to do and it is still the wrong lens here: the cohort comparison is valid, but the mechanism behind the gap is a dispellable debuff nobody removed, which is a party finding filed under one player's name.

The dispel axis reads 100 -- 37 of the 37 that one dispeller could physically reach, given that Cleanse recharges in about nine seconds and up to five cleansable debuffs can be live at once. That number is honest, and it is a ceiling on count. It says nothing about allocation, and allocation is where this key was decided.

The 33 cleanses went: Spore Spines 22, Bloodthorn Roots 9, Toxic Spew 2. By target, 24 of the 33 went to Prot Warrior.

Rank the three by what they cost per second they are up, which is the number that should decide a contested cooldown:

  debuff             uptime      damage    per second up
  Toxic Spew            16s   4,666,805          291,986
  Bloodthorn Roots      64s  13,626,382          213,764
  Spore Spines         205s   7,307,486           35,644

Spore Spines is six times cheaper per second than the roots and eight times cheaper than Toxic Spew, and it took two thirds of the cleanses. It is up almost constantly on the tank, which is exactly why it feels urgent and exactly why it should lose the race: it will still be there in nine seconds, and the tank was never in danger -- zero deaths, 97% mitigation uptime.

There is a second dispeller available. Shadow Priest' loadout carries Mass Dispel, which strips a Magic effect from up to five allies at once and is the right shape for a root that lands on the whole party. He pressed it zero times. Worth one line of confirmation with him first -- a loadout entry is read out of COMBATANT_INFO and is not proof the spec grants the button.

The Belcher started casting at 19:34 and finished at 19:35. The damage did not arrive until 19:38.9, and then again at 19:40.4 -- one cast, two waves, a second and a half apart.

Going in, Spouting Floret had been ticking about 200k on everyone every two seconds and the party sat at 40-70%. First wave: Devourer DH to 218k, Shadow Priest to 121k, Ele Shaman to 315k, Hyporock to 301k. Shadow Priest pressed Dispersion half a second later and took 162k on the second wave instead of 500k. Hyporock pressed Aura Mastery 0.04 seconds before it and finished on 42,455 health out of 792,740. Devourer DH pressed Blur in the same instant the hit landed. Ele Shaman pressed nothing and took Reincarnation afterwards.

The healer half of that window is not fixable. Even with the party-wide reduction applied to both waves, Devourer DH's 1,138,486 still exceeds what he had plus everything that reached him. The fixable half is upstream: the same spell was cast ten times in this key and reached players on only three of them, so it is dodgeable, and there were roughly four seconds between the cast finishing and the first wave landing.

Three things looked like leaks in the tables and are not, and they are written down here so they do not get re-investigated.

Shield of the Righteous took 270 of the 630 Holy Power spent -- 43%, against 360 into Light of Dawn and Eternal Flame together. That reads as healing diverted into a personal button. It is not: only 6% of its 141 casts fell in the worst-damage quarter of the key, against 25% for a button pressed at random and 66% for Light of Dawn. It is a downtime dump, which is what it should be, and it returned 231,415 mana on the way.

The old holding habit is gone. Healing output in the worst quarter of the key ran 3.34x the rate everywhere else. Nineteen gaps over four seconds add up to 172 seconds, and exactly one of them -- five seconds -- sits inside that worst quarter. Idle placement, not idle total, is the tell, and this is as clean as it gets.

Beacons are fine. Virtue was pressed 61 times and reached all five players; the Savior beacon moved as it should. No placement change.

The one genuine item: 155 Holy Power overcapped of 790 generated, and 57 of those 155 fell inside the worst windows rather than the 39 a flat spread would put there. At 260k effective per Light of Dawn that is about nineteen casts, roughly 5M -- small next to the 9M sitting in the roots, but it is the only output-side number in the key with anything under it.