DREAMBREATH

How to ramp

A ramp is healing you cast before anyone is hurt. Every dangerous pull in Mythic+ has one moment that does most of its damage, and you want your heals already landing when it arrives, so the party bars dip instead of dropping. If you cast your first heal when you see a health bar move, you are already too late.

One thing to know before you read the numbers

We only recorded the last 30 seconds before each spike. So everything below tells you what gets pressed inside that window, and none of it tells you when a ramp really starts. Yours may well start earlier. Do not read these as start times.

Cast data exists for 1335 of the 1336 logged runs, and holds 355,869 casts.

A quarter of your cast log is not a decision

Log-verifiedn=1335 runs with cast data
22% of the casts recorded in the window are procs: 79,054 of 355,869 rows that nobody pressed.
A proc is logged exactly like a press, so a raw cast count reads as though you pushed a button for it. Anything built on that count, including a ramp order, inherits the error. Every figure below this point is against the 276,815 casts that were chosen.

Twin Flame is 55,526 of those. It fires when you consume Essence Burst. It is the single largest row in the whole dataset and it is not a button.

Merithra's Blessing is another 23,528. It is a passive that upgrades your next Reversion, so Merithra's Blessing and Reversion are one button logged under two names, and their 23,528 and 6,309 casts were never independent numbers. You cannot put Merithra's Blessing in a sequence, because you cannot press it. Any guide listing it as its own ramp step is wrong.

Against the 276,815 casts you actually chose, ramp abilities are 47%.

The set of abilities is consistent. The order is not.

You can only see this by comparing runs against each other. Take Animated Guardian in Kings' Rest, below. The abilities repeat. The opener does not.

175 runs with ramp casts · a typical ramp is 6 abilities

No ability reliably goes first here. The most common opener was Temporal Anomaly, in 33 of 175 runs. Treat this as a set to get out, not an order.

Ramp abilities for Animated Guardian, by how often each appeared and opened.
AbilityAppeared inOpened the rampTypical position
Dream Breath99% (174 of 175)22 of 1753
Temporal Anomaly93% (162 of 175)33 of 1753
Verdant Embrace74% (129 of 175)22 of 1753
Echo60% (105 of 175)25 of 1753
Emerald Blossom78% (136 of 175)28 of 1754
Log-verifiedn=15 Kings' Rest pulls with a ramp table
On 15 of 15 Kings' Rest pulls, no single ability opened even half the runs.
These are good players ramping the same pack, and they do not agree on what to press first. If you have been trying to learn a fixed opener off a guide and it keeps falling apart in the key, that is because there is no fixed opener to learn.
Thin sample: 15 Kings' Rest pulls with a ramp table.

So memorise which abilities go into a pull. Do not memorise an order.

What goes into almost every ramp

Across the 15 Kings' Rest pulls with a usable ramp table, 2 abilities turn up on most of them. The middle column is the strict one: it counts the pulls where the ability was in every single recorded run, not most of them.

Out of 15 pulls. An ability is only listed on a pull if it appeared in at least 60 percent of that pull's runs, so an ability can be missing from a pull here and still have been cast on it.

Ramp abilities in Kings' Rest, by how many pulls they appear on.
AbilityPulls it appears onPulls where every run used itPulls it ever openedTypical position
Dream Breath151152 to 3
Temporal Anomaly150152 to 3
Emerald Blossom110113 to 4
Verdant Embrace100102 to 4
Echo6063 to 4

Dream Breath and Temporal Anomaly are the core. Treat Emerald Blossom, Verdant Embrace and Echo as pull-specific rather than standard.

Dream Breath is the clearest case of a cast everybody uses and almost nobody leads with. It is on 15 of the 15 pulls, it was in every recorded run on 1 of those, and its typical position sits between 2 and 3 everywhere it appears. Get it out, and get it out in the middle.

A typical ramp runs from 4 to 6 distinct abilities depending on the pull, and it is 4 or 5 on most of them.

Keep dealing damage while you ramp

Damage is 18% of the casts you actually press, and 14% of everything recorded. What gets pressed is Disintegrate, Fire Breath, Deep Breath or Azure Strike, and Disintegrate and Fire Breath are 94% of it.

Living Flame is not on that list, despite being one of your largest rows at 39,871 casts in 1314 of the 1335 runs with cast data. For Preservation it is a heal that can also damage: the tooltip leads with the healing and lists the Fire damage second, and a cast record cannot tell you which one happened. Treat it as your filler heal.

No damage ability appears in the ramp tables above, and that is a fact about the query rather than a finding: those tables only consider abilities filed as ramp, Stasis or hold-for-it casts. Press damage in whatever moment is free. Do not slot it as a fixed step.

Know when the hit lands

The pull timings are the part of this data you can put a stopwatch on. Every figure here is a median across the runs containing that pull, measured from the moment the pull starts.

Spike time is measured from the beginning of the pull, not the beginning of the run. Leading cause is the largest single non-melee source inside that window. Learn the number for the pulls you keep dying on: it tells you when you need to be finished.

Kings' Rest pulls with a ramp table, and how far into each pull the damage spike lands.
PullRunsSpike landsLeading cause
Guard Captain Atu17712sForked Lightning
King Timalji17715sBladestorm
Royal Berserker12523sBloodthirsty Axe
Queen Patlaa17724sShoot
King A'akul17729sSudden Rupture
Shadow-Borne Champion12632sShadow Whirlwind
Interment Construct11835sHeavy Slams
Ghostly Brute17737sBloodthirsty Axe
Shadow of Zul17742sShadow Barrage
Mchimba the Embalmer17743sDrain Fluids
Animated Guardian17644sHeavy Slams
Purification Construct17744sPurification Strike
King Dazar1771m 20sGilded Destruction
The Golden Serpent1771m 22sSerpentine Gust
The Council of Tribes1772m 12sArc Lightning

The full sheet for every pull, including the ones without a ramp table, is on the Kings' Rest page.

When the pull comes early

You cannot shorten a sequence you do not have, so cut by importance instead. Keep the abilities that appeared in every run of the pull, which is the middle column of the table above. Drop the ones that only turn up on some pulls, because these players were already dropping them.

Then change how you release Dream Breath. Its total healing is the same at every empower rank, but the lowest rank spreads it over a long window while the highest front-loads it. If you are short on time, release early at a higher rank. The lead time is what a low rank wants.

The logs do not record mistimed pulls, so the cutting order here is reasoning from the appearance counts rather than a measurement.

When the pull comes late

Do not start over. Fill the extra time with damage, then re-press whatever came off cooldown while you waited.

Stasis is the one measured tool here and it is used unevenly. Stasis was stored 3,991 times across 1211 runs and released only 1,972 times across 945 runs, so most stores made inside a ramp window are not spent inside one.

What Stasis banked

Stasis duplicates your next three helpful spells and stores them. Dream Breath appears in 15 of the 15 banks recorded on a Kings' Rest pull. Read that as an observation rather than a rate: it rests on 1324 uses.

Every row is one or two runs. We worked out each bank from the next casts after the store, then stripped anything Stasis cannot hold, so a row with fewer than three spells is one we could not fully recover rather than a bank that went out short.

What Stasis held going into Kings' Rest pulls, and how long it was held.
PullRunsHeldBanked
Animated Guardian12420sDream Breath + Temporal Anomaly + Emerald Blossom
Shadow-Borne Champion8024sDream Breath + Temporal Anomaly + Emerald Blossom
The Golden Serpent14421sDream Breath + Temporal Anomaly + Emerald Blossom
Guard Captain Atu3840sDream Breath + Temporal Anomaly
King A'akul6723sDream Breath + Temporal Anomaly
King Timalji2344sDream Breath + Temporal Anomaly
Queen Patlaa2929sDream Breath + Emerald Blossom + Emerald Blossom
Purification Construct11120sDream Breath + Temporal Anomaly + Emerald Blossom
Interment Construct5234sDream Breath + Temporal Anomaly
Mchimba the Embalmer10731sDream Breath + Temporal Anomaly + Emerald Blossom
Royal Berserker5823sDream Breath + Temporal Anomaly + Emerald Blossom
Ghostly Brute10924sDream Breath + Temporal Anomaly + Emerald Blossom
The Council of Tribes14024sDream Breath + Temporal Anomaly
Shadow of Zul8021sDream Breath + Emerald Blossom + Emerald Blossom
King Dazar16220sDream Breath + Temporal Anomaly + Emerald Blossom

The long holds are somebody banking a ramp early in a dungeon and carrying it to a boss. The reactivation clock starts once the third spell is stored, not when you cast Stasis, so a long hold is a slot left unfilled rather than a bug in the data.

How these numbers were measured

Runs come from Warcraft Logs. Preservation Evoker only, Mythic+ only.

Only one report counts per key. Two players in the same group each upload the same run, so the raw data holds that key twice. 0 of 1336 logged rows are a second copy of this kind, and only one of each pair is counted. That is where 1336 comes from.

Runs whose event stream was cut short are excluded, because their totals would be low by an amount nobody can measure. None have hit that yet.

A pull is listed only if it appears in at least 60 percent of that dungeon's runs. What you see is the dungeon's spine, not one group's exact route. If your group skips a pack listed here, or takes one that is not, nothing is wrong.

An ability is called part of the ramp only if it appears in at least 60 percent of the runs with ramp casts for that pull. One player's habit on one key does not reach the page.

Casts are only collected from the 30 seconds before a damage spike. That is a setting, not a finding, so nothing here can tell you whether a ramp starts earlier than that. Where you see a window, it is the window we looked in.

Ramps are worked out one run at a time and then compared, rather than by taking each ability's average timing separately. Doing it the second way produced ramps of six casts inside four seconds, which is faster than the game allows anyone to press buttons.

Damage figures are the whole party added together. Nothing here records who took the damage, and most autoattack damage lands on the tank. Use these numbers to compare one pull against another. They are not an amount you have to heal.