Icecrown Citadel: reading your logs wing by wing to fix your wipes

Three hours of raid, twelve wipes on Putricide, and at the debrief nobody can say why. Good news: your combat log knows. Here is how to read an Icecrown Citadel log (ICC 10/25) on Skaldlogs, wing by wing, with the concrete indicators for each boss that turn a blurry wipe into a precise fix for the next pull.

The method before the bosses

Start by uploading your WoWCombatLog.txt at /upload — free, no account needed, parsing runs right in your browser. If you use the Skald addon, the combat log enables itself the moment you enter a 10/25 raid, so no more forgotten pulls. Then, for each boss, cross-check four views:

• The death recap: the last 10 seconds before each death — damage taken, healing received, defensive CDs used (or not).
• Damage taken by source: who eats what, and above all who eats avoidable damage.
• The DPS timeline: where throughput collapses, and whether it is a whole phase or one specific player.
• Food/flask compliance (4-level colour code) and the 60-second prepull check: on progression pulls, a raid without flasks is a self-inflicted wipe.

Keep /guides open for the per-boss checklists, and /top to compare your kill times with everyone else's.

The Lower Spire

Lord Marrowgar

First boss, and already two eternal traps:

• Damage taken from Coldflame. If a death recap shows Coldflame ticks, the player stood in the fire on the ground. One isolated tick happens to everyone; a player stacking them every pull has a positioning problem, not bad luck.
• Bone Spike Graveyard freed too slowly. Look at the DPS timeline of impaled players: someone sitting at zero DPS for 10-15 seconds after a Bone Spike means a raid that does not switch. The impaled target takes continuous damage — every second of delay is paid for in healing stress.
• Damage taken during Bone Storm. The whirlwind telegraphs itself from miles away: those who eat a lot of it simply are not moving.

Lady Deathwhisper

• Add control. Compare damage done to the Cult Fanatics and Adherents with damage put into the mana shield. DPS tunnelling the boss while adds roam free shows up instantly in the damage breakdown — and in the healer deaths.
• Damage taken from Death and Decay. 100% avoidable. A player who eats it regularly is camping the zone to finish a cast: show them their numbers, it works better than a lecture.
• Mind-Controlled players. On 25, Dominate Mind turns your own raiders against the raid. The log shows the damage dealt BY your players to their teammates: if it is high, your CC on the MCs is landing too late.

Gunship Battle

Let's be honest: it is a loot pinata. If someone dies here the death recap will tell you how, but the real use of the log is elsewhere: use this quiet fight to verify food/flask compliance before Saurfang. A serious raid re-buffs during the Gunship — does yours?

Deathbringer Saurfang

• The number of Marks of the Fallen Champion is THE metric of the fight. Every Mark is born from Blood Power you let leak. 0-1 Marks at the kill: clean execution. 3 Marks or more: your healers are drowning and the fight becomes a race lost in advance. Count the Marks pull by pull — it is the most reliable trend indicator in the whole wing.
• Blood Beasts reaching the ranged. Look for Blood Beast melee hits in the damage taken of your casters and healers: every hit landed feeds the boss's Blood Power. If it keeps happening, your slows and kiting start too late.

The Plagueworks

Festergut

• A pure DPS check, almost chemically pure. The raid against the berserk (roughly 5 minutes), full stop. Look at the raid's effective DPS — Skald excludes AFK time and inactive phases, so no excuses — and compare your kill time on /top. If you wipe to the berserk at 5%, there is no mechanic to fix, it is throughput: flask/food compliance goes from desirable to mandatory.
• Inoculated stacks. A player who died to Pungent Blight without their 3 stacks of Inoculated missed their spores. The death recap shows it in black and white — no debate possible.
• Vile Gas. Vile Gas damage hitting several players at once = badly spaced ranged.

Rotface

• Mutated Infection handling. Filter the deaths with Mutated Infection active: an infected player who dies far from the kiter left too late or panicked. Also check the healing received by infected players — the disease reduces healing taken, those players need dedicated attention.
• Unstable Ooze Explosion. Entirely avoidable damage. If several players take it on every explosion, your raid is not moving away from the Big Ooze in time — or nobody is calling the merge.
• Kiting quality. The damage taken by your kiter tells the whole story: a kiter who takes too many hits is turning too tight or moving too slowly.

Professor Putricide

• Malleable Goo hits. Every goo taken reduces cast and attack speed: a dead loss of DPS and HPS. Count the hits per player — three players hit per phase, and your P3 wipe is already written in P2.
• Ooze and gas target management. The Volatile Ooze target dying before the add does, the Gas Cloud catching its prey: it is all in the death recaps. Also measure the add kill times — an ooze that survives 30 seconds is that much boss DPS evaporated and a berserk drawing closer.
• P3 stacking discipline. If the timeline shows raid DPS collapsing while damage taken explodes, your raid is scattering instead of holding the pack behind the boss. P3 is won on discipline, not on throughput.

The Crimson Hall

Blood Prince Council

• Kinetic Bomb control. Look for Kinetic Bomb in the damage taken: every explosion means a bomb hit the ground with nobody keeping it in the air. Assign a dedicated ranged — and verify in the log that they actually do the job, pull after pull.
• Chained Empowered Shock Vortex. Several players hit by the same vortex = players stacked at the wrong moment. The number of targets hit per cast is your spread indicator: it should trend towards one.

Blood-Queen Lana'thel

• Bite chain efficiency. Essence of the Blood Queen is a huge DPS multiplier: on the timeline, a bitten player's DPS must take off. A missed bite = a player in Uncontrollable Frenzy, and most of the time a wipe. Reconstruct the bite order pull by pull and identify who broke the chain.
• Pact of the Darkfallen. Pact damage keeps stacking up until the linked players have come together. Big Pact totals in the damage taken = players searching for each other instead of converging.
• Swarming Shadows. Swarming Shadows damage = a player who did not leave the zone — or worse, who dropped it inside the raid.

The Frostwing Halls

Valithria Dreamwalker

• An HPS race, not a DPS race. The central indicator is the healing sent TO the boss. Compare each healer's effective HPS against the others and from one pull to the next: this is the fight in all of ICC where the gaps between healers show the most.
• Portal usage. Healers who take the dream portals massively boost their output. A healer whose HPS does not climb over the course of the fight is not exploiting the portals — the timeline gives them away.
• Add pressure on the raid. While your healers heal the boss, the raid takes a beating. Watch the deaths on the DPS side and the Blazing Skeletons' damage: if your players drop, the priority adds — Suppressers first, who reduce healing received by the boss while channelling — are not dying fast enough.

Sindragosa

• Unchained Magic: who keeps casting? Marked casters who keep spamming stack up Instability and explode — themselves and everyone around them. Instability damage in the log names exactly who cannot stop.
• Deaths to Blistering Cold. After Icy Grip, everyone sprints outward. Every death to Blistering Cold = a player too slow or who left too late. It is the most binary indicator in all of ICC: the target is zero, no exceptions.
• Ice Tomb placement. In phase 3, Ice Tombs are used to break line of sight on Frost Bomb. Frost Bomb damage on the raid = players not behind the tomb in time; a player dead inside their own Ice Tomb = a tomb broken too late by the raid.

The Frozen Throne

The Lich King

• Defile, the number one raid killer. Defile grows with every tick taken. In the log, a series of swelling Defile ticks followed by a cluster of raid-wide deaths = somebody fed it. Identify who takes the first ticks, pull after pull — it is almost always the same people.
• Val'kyrs. Measure the time between the pickup and the moment the Val'kyr drops her target: too long = late stuns and slows, or a sluggish DPS switch. A raider thrown off the edge of the throne is a very short death recap and a pull finished under-strength.
• Necrotic Plague (Heroic). On Heroic mode, watch the Necrotic Plague damage on the raid's side: the plague should jump cleanly towards the adds, not wander between your players. Details vary with your server's tuning — check on your own pull rather than on the forums.
• The Frostmourne phase (Harvest Soul). Players pulled inside Frostmourne have to come back out alive: a death in there reads in the death recap and costs you a raider for the entire rest of the fight. Also keep an eye on the Vile Spirits exploding in the pack.

Make it a routine, not an event

The log reading that actually drives progress is the regular kind: after every raid night, upload, sort deaths by ability, compare with the week before. Three questions are enough: who dies to avoidable damage? where does the DPS collapse? is the preparation (flask, food, prepot) up to standard on the pulls that matter? With the per-boss checklists on /guides, the kill times on /top and the class/spec percentiles on /stats, you have everything you need to turn a night of wipes into a concrete action list. The Lich King is waiting — and now you know exactly what to look at.