Entombed Sentinels, 34 pulls,
two bosses to keep level.

The night, pull by pull

Boss health remaining at the wipe, the two Sentinels averaged, as Warcraft Logs reports it. Bar colour is how many intermissions the pull reached. Pull 2 and pull 22 lived to the berserk at 7:00, pull 22 with 9.3% still to go.

Seven minutes

Vitriolic Stasis opens at 46 seconds on every pull, and each one after that opens exactly 91 seconds after the previous one ends. The damage phases are a fixed length. The only part of the clock the raid controls is how long each intermission lasts, and the berserk lands at 7:00 regardless.

Pull 22 spent 69 seconds in four intermissions. The six kills below spent 28 to 50 seconds in three or four and were finished by 6:07–6:17. Across the night's 62 finished intermissions the median was 13 seconds, against 12 for the kills, so a typical one is not the problem. The long ones are: 6 ran past twenty seconds, when no kill's took longer than 16.

What Stasis hands back

During Stasis the lower Sentinel is healed, a tick a second, until it matches the higher one. Any imbalance going in has to be done again. The heal is in the log as Vitriolic Stasis healing on the boss, so this is measured rather than inferred from the health bars.

Across the night it handed back 727M, 1.7 times the 435M health of one Sentinel. On pull 22 it was 71M, 16.3% of one Sentinel. That is 8.2 points of the two Sentinels' combined health, on a pull that ended 9.3 points short. The kills handed back 9–59M across a whole kill, 22M at the median. Going into an intermission the bosses were 3.1 points apart at the median this night, against 1.9 for the kills, and 31 of the 62 intermissions opened more than three points apart.

Two teams, and where the gap comes from

Mark of Acid only lands within 40 yards of Breath, and Mark of Blood within 40 yards of Blood, so the marks show where everyone stood without any coordinates. Read that way, the raid is two teams that trade bosses after every intermission, read pull by pull rather than once for the night. Team A (1 tank, 2 healers, 7 damage on a typical pull, Paladindevil tanking) opened every pull on Blood. Team B (1 tank, 2 healers, 7 damage on a typical pull, Vesemirr tanking) opened on Breath. The teams changed during the night: Emeralld (Holy) from team B to team A at pull 16, Shazàm (Holy) from team A to team B at pull 16, Ørnin (Demonology) from team B to team A at pull 29 and Arozax (Arcane) from team A to team B at pull 31.

Going into an intermission with the bosses more than half a point apart, the lower boss was team B's in 47 of 55. So the heal is not bad luck in one phase: the same team's boss is low almost every time. The gap can be broken into three parts, read at the median phase. In phase one (over 32 pulls), team A did 0.9M more on its own boss than team B did on theirs and cross-room damage put a net 6.1M more into B's boss. In phase two (over 23), with team A on Breath and the slime, team B did 10.6M more on its own boss than team A did on theirs and cross-room damage put a net 7.8M more into B's boss.

Cross-room damage is concentrated in two players: Niydra (Destruction, 351M, 37% of their boss damage) and Arozax (Arcane, 73M, 18% of their boss damage). No other damage dealer put more than 4% of their boss damage into the other team's boss. On pulls where a damage dealer could not be placed with either team, standing between the bosses or collecting both marks, their boss damage counts as its own part. Arozax was unplaced on 12 pulls and put 230M into Breath against 82M into Blood on them.

The gap moved with those changes. Over pulls 16–28 the bosses went into Stasis 4.3 points apart at the median and each intermission handed back 16.4M; over pulls 29–34 it was 0.6 points and 2.2M, level with the kills' 1.9. That last stretch is only 9 intermissions, a short run rather than a settled answer. Pulls 16–28 opened with the healers trading teams, and healers put next to nothing into the bosses, so that change is unlikely to be the cause.

Damage onto the bosses

Damage onto each Sentinel and the slime across every pull, start to finish, with pets credited to their owners. The pull selector narrows it to one attempt.

Against the kills

Six public Mythic kills measured by the same code: the two earliest in the rankings, at item level 320, and four from the last two weeks at 327, the same gear as this roster (326). Boss damage per second is counted inside the damage phases only, so a long intermission does not dilute it.

Phase one carries the lust (lust went out 1 to 5 seconds into every pull) and is close to the kills: 4.04M a second at the median this night, 4.16M over the second half of it, against 4.16–4.68M. Phase two is where it falls behind: 2.09M at the median and 2.32M at best, against 2.36–2.57M, about 17% short. Pull 22 averaged 2.35M through its damage phases, against 2.59–2.77M. With 351 seconds of damage phase before the berserk, pull 22 needed 940M (both Sentinels plus what Stasis gave back), which is 2.68M a second.

The puzzle

Every player gets Helical Toxins with one to three orbs. Two players whose orbs make exactly four clear each other on touch; any other touch merges their stacks without clearing, and spends orbs somebody else needed. A 1 and a 1 become two 2s, a 1 and a 2 two 3s, so every stack increase is a wrong pairing even when it stays under four. Stasis does not wait for the puzzle: the debuff lasts about 28 seconds, runs on after Stasis ends, and whoever still holds it when it runs out takes Cultivated Burst. So a solve is every player clearing before the debuff expires, and the solve time is when the last of them did.

Of the night's 62 intermissions, 40 were solved and 11 failed on a wrong pairing (pull 3 #1, pull 3 #2, pull 5 #1, pull 6 #1, pull 8 #2, pull 9 #1, pull 11 #1, pull 21 #1, pull 23 #1, pull 29 #2 and pull 32 #1). The other 11 were lost to deaths or a called wipe: 8 opened with more than two of the raid already dead, and the rest had a player die holding the debuff, never to be resolved, before anyone paired wrong. A corpse keeps its counter and can still be bumped and cleared, so a death only counts here when the debuff was never resolved. 11 of those pulls were over within twenty seconds of the debuff running out. Counting only the 51 the puzzle decided, 22% failed. The kills solved every one the puzzle decided, 19 of their 20 intermissions, with 1 wrong pairing among them all.

Wrong pairings are counted two ways. One broke the puzzle when it was the first in an intermission that could still be solved: two or fewer dead going in, and no unresolved death before it. Every other merge is counted as after it was lost. That covers a called wipe, where pairing wrong on purpose ends Stasis so the bosses can be kited out and reset, and a puzzle cascading on its own. The log cannot tell those two apart, so neither is held against anyone. 46 merges across the night fall in that second group.

When it was solved, the solve took 13.2 seconds at the median and 22.6 at the longest, against 8.1–15.5 for the kills (12.0 median). The first pair cleared at 2.0 seconds and half the raid by 4.0. Slowest to clear at the median: Preparedguy (7.3s), Arozax (6.1s) and Aelyssar (5.5s). Most often in the last pair of a solved puzzle: Vesemirr (7), Arozax (7) and Preparedguy (7). Most often in the wrong pairing that broke a puzzle: Solmarion (4), Valanarr (3) and Emeralld (3). A wrong pairing can fail somebody else: when two players spend orbs that a third needed, it is the third who is left without a partner and takes the burst, even if the two who touched later clear each other. Most often left without a partner that way: Ørnin (2) and Emberforge (2).

Toxic Droplets

Breath sprays droplets 148 times across the night. One that nobody runs over goes off as Noxious Blast 14 to 15 seconds after the cast and hits the whole raid for about 272k each. 30 sets left at least one behind, 46 blasts in all, on 20 pulls; the kills missed 0, 1, 1, 1, 2 and 1. 18 deaths went to the soak itself.

Noxious Blast is the killing blow on 97 deaths. 16 of the 19 sets that killed anyone went off with 70% or more of the raid still up, and those killed 83 between them. In 11 of them the pull was over within thirty seconds. These are wipes the droplets started, not a raid already falling over. The damage comes from sets that leave more than one droplet: 12 sets had two or three go off together and they killed 71.

The set that matters is the last one before Stasis, cast about ten seconds before the intermission opens, so it has to be soaked while the raid moves to the middle. 15 of those 30 sets (50%) left a droplet behind and they killed 59, against 15 of the other 118 (13%) and 38 deaths.

Who soaks

Every set is twenty droplets, and every droplet is accounted for: one player runs it over (one Toxic Droplets hit, never shared) or it goes off. Soaks plus blasts come to 20 on 128 of the night's 135 complete sets, so the counts below are droplets, not estimates. With twenty players that is one droplet each per set if the load were even.

It is not even. Aelyssar (261), Alfi (255), Indris (250) and Solmarion (243) took 38% of the night's 2642 soaked droplets between them, 2.0, 2.0, 1.9 and 1.9 a set while alive. The damage dealers taking the fewest per set were Mysticake (0.3), Arozax (0.5), Evonux (0.5) and Atheniás (0.6). The droplets land all over the room, not just around Breath: the team on Breath soaked 64% of them, the team on Blood 36% and players not placed with a team that pull 0%.

The set before Stasis is soaked late. Droplets from that set were run over 5.9 seconds after the cast at the median, against 4.7 for every other set, and 13% of them after ten seconds against 4%. They blow at about fourteen and a half seconds, so a late soak in that set is one step from a missed one.

Living Venom

Breath throws out venom that flies back to the boss four seconds later, hitting anyone in its path. 306 hits landed across the night, counting ticks on one player within a second as one hit. They killed 17, and 7 pulls began their death log with one. 86% of the hits on a player whose side could be read came while they stood on Breath, which is where the venom flies. Per minute of phase, phase one took 3.4 hits and phase two 3.2, about the same rate either way, so it is not the slime phase that gets hit more.

Set against time alive in damage phases, the non-tanks hit most often per five minutes were Emeralld (2.9), Valanarr (1.8) and Indris (1.4); the median non-tank took 0.7. A hit was followed by that player's death within five seconds 27 times, most often for Emeralld (5), Sunnymikej (4) and Aelyssar (4).

Shifting Protovenom

100 sets of marked players went out. Two marked players meeting clear each other; a marked player touching a clean one sets off Protovenom Eruption on everyone within ten yards. Full sets of eight cleared 3.2 pairs out of four on average, against 3.4 for the kills, and 33 sets cleared with no eruption. There were 89 eruptions in all (the kills had 0, 0, 4, 2, 0 and 0), and 24 deaths went to Protovenom or its eruption.

Each eruption is pinned to one carrier and the clean player they touched, using where everyone stood when it went off: 85 of the 89 could be pinned, the carrier a median 3.7 yards from the player they touched. The median eruption came 1.6 seconds after the carrier was marked, and 80% within three seconds. Most eruptions happen in the first seconds after the marks go out, before anyone has had to walk anywhere: a marked player starting the mechanic next to someone clean.

Most eruptions caused: Alfi (8), Solmarion (8) and Arozax (7). As a share of times marked, the highest were Solmarion (8 of 36 times marked), Alfi (8 of 38 times marked) and Evonux (6 of 35 times marked). Most often the clean player touched: Valanarr (10), Evonux (8) and Shazàm (7). Pairs that happened three times or more: Alfi into Evonux (4), Arozax into Shazàm (3), Sunnymikej into Valanarr (3), Solmarion into Alfi (3) and Paladindevil into Valanarr (3).

The rest of the floor

Blood Venom pools were stepped in 1276 times and killed 23. The most frequent were Alfi (97), Indris (91) and Mysticake (80), tanks aside. Blighted Blood went out 263 times and was dispelled after 4.2 seconds at the median; 82 took longer than eight seconds and 14 were never dispelled. Unstable Miasma went off 105 times, shared by 8 at the median; 10 had fewer than six under it and it killed 10.

The marks stack every five seconds for as long as a player stays in range, so they build through a phase to a median peak of 17 stacks. They are 48% of all damage the raid took and the killing blow on 97 deaths, 28 of them solo. Only 14 came in the last twenty seconds before an intermission, so they do not mostly kill at the end of a phase, when stacks are highest. They kill players who are already low.

Marks from the other boss

Stasis does not clear the marks. Each one runs out forty seconds after it was last applied, so after every swap a player carries their old boss's stacks for twenty-odd seconds into the next phase. That is the fight and is not counted here. What is counted is a new stack from the boss a player's team was not on that phase, which only lands within forty yards of it. It restarts the old mark's timer and adds a stack, or starts a second mark alongside the first.

849 stacks went on from the wrong boss across the night: 228 in the first ten seconds after Stasis, while everyone crosses over; 450 later in a phase, extending an old mark that was still running; and 171 fresh marks picked up from the other boss. The extra damage they did, over what the old stacks would have done if left to run out, came to 245M, 5.7% of all mark damage. The kills lost 2.2–12.1% the same way; per minute of damage phase it was 2.9M this night against 1.3–8.4M for them. So as a raid this night was no worse than the kills: the marks from the other boss are a handful of players, not a spacing problem across the room.

Most extra damage: Arozax (33M, 15% of their mark damage), Valanarr (32M, 12% of their mark damage), Niydra (17M, 7% of their mark damage) and Aeonara (17M, 7% of their mark damage). Per minute alive in damage phases that is Arozax 0.42M, Valanarr 0.40M, Niydra 0.22M and Aeonara 0.20M. The top two broke down as Arozax: 21 on the walk-out, 94 extending an old mark, 24 fresh and Valanarr: 15 on the walk-out, 51 extending an old mark, 22 fresh. If Arozax and Valanarr were placed to reach both bosses on purpose, this is the price of that job; if not, it is the most avoidable mark damage on the night. Overall, the most mark damage taken was Valanarr (268M), Ørnin (242M) and Emberforge (236M), against a median of 221M.

Deaths

402 deaths across the night, 96 of them solo, meaning the raid was still standing (70% or more alive) and fewer than three people went down within three seconds. The first death of a pull went to Living Venom 7, Droplets 6 and Marks 5. Among solo deaths the leaders are Marks 28, Droplets 14 and Protovenom 12. The solo rule leaves out anything that kills several people in the same instant, however healthy the raid was, so a missed droplet barely registers there; the Toxic Droplets section counts those blasts set by set.

Damage taken, second by second

Incoming damage on the raid across a pull, tanks excluded, since their intake is the job and would bury everyone else. The axis is fixed at 435 seconds, the longest attempt, so every pull reads against the same clock, with each Stasis shaded. Drag the handles to a stretch of the fight and everything below recalculates for it: damage taken, how much was mitigated, and every cooldown, stone and potion spent inside it.

The two sides take damage differently: Breath's side runs over droplets, kills the slime and dodges Living Venom; Blood's side shares Unstable Miasma, dispels Blighted Blood and steps around its pools. The view switch splits the damage by the boss each player stood beside, from the marks, phase by phase. Teams change bosses after every intermission, so a side is a place, not a group of people. Intermissions belong to neither side. The ticks along the top are Breath's droplet casts and each Unstable Miasma going off, so a spike can be matched to its mechanic. In a side view the table counts each player only for the seconds they stood on that side.

Defensives in the heavy seconds

137 spans across the night ran at twice the pull's median damage taken or more. They are 10% of the time on the boss and 19% of the damage. This is who pressed something in them.

Healthstones & health potions

231 healthstones and 143 health potions went down across 34 pulls: 205 of the 374 at under 40% health, 69 of those under 20%, 21 above 80%. Average 40%, median 37%. Two of the roster used no healthstone at all (Aeonara and Emeralld). Potions: 98 Concentrated Silvermoon Health Potion, 36 Silvermoon Health Potion and 9 Potent Healing Potion. Stones: 180 Healthstone and 51 Demonic Healthstone.

How this was counted