[Resto] LHW Spamming for Fun and Profit
 

The new talents introduced for Wrath of the Lich King and implemented in the 3.0.2 patch are designed to round out shaman healing from straight Chain Heal (CH) spam to a more rounded mix of single target heals and CH spam, though blue comments suggest that they consider AoE healing our niche (just as tank healing is the pally's niche). The new talents revolve around crit, affecting both Lesser Healing Wave (LHW) and Healing Wave (HW) but the fact that imp. water shield returns a fixed amount of mana means that LHW becomes our heal of choice. Through crit stacking, it's possible to buff the HPS and HPM of LHW to levels comparable to 2-3 target CH spam.

Since the build for this heal style is the cookie cutter 0/13/54+4 with required points in AA and Imp. WS, this is more of a "how it works in practice" article and assumes you're stacking crit by wearing eleshaman gear. My theorycraft backing for this is located in this spreadsheet (google docs), from which you can draw your own conclusions and which I'd appreciate vetting. Thanks to Daidalos for providing the intial spreadsheet.

LHW in depth

The LK LHW offers:

• Good efficiency - Talented LHW costs 626 mana. Talented water shield returns 492.2 per globe, or 78.6% of the spell cost on crit. With the 2t7 bonus plus full talents, water shield returns 535 mana, or 85.4% of the spell cost. These are both significantly better than illumination, but are offset by the need to regularly refresh water shield.
• Medium Throughput - The throughput of LHW spam is good but will never approach the amount a pally bombing 2 second Holy Lights. It is, however, considerably higher on a glyphed target than any other 1.5 second cast, putting it squarely in the medium heal range priests have been requesting since the downranking nerf.
• Fast reaction time - With tidal waves up and WoA down, the heal hits in < .9 seconds. You have to wait out the GCD but the heal is delivered very fast.
• Crits proc Inspriation - It's called Ancestral Fortitude but the real name is inspiration. 25% armor bonus 100% of the time is simple to hit even with basic gear due to all the crit talents. The warrior/pally tanks I was running heroics with were reporting about a 5% increase in the reduction from armor on their paper doll tooltip.
• 195% crits - Crits are traditionally not counted in healing mechanics, but I believe that this is a mistake in the case of LHW. I've been healing reactively rather than proactively so my overheal is in the 10-20% range and my Ancestral Awakening overheal is under 5%. HW crits still seem to be mostly overheal.
• Nice Totem - At the time of writing, totems at 80 aren't really itemized. The [Hateful Gladiator's Totem of the Third Wind] changed from supplying resilience on target (s1-s4) to providing 204, 236, and 263 SP. With Naxx levels of SP in the 2.1-2.3k range, this offers a ~10% SP boost rather than the CH's ~10% discount.
• High Earthliving uptime on the tank.

All in all, much improved from the BC state and a solid general purpose spell.

Ancestral Awakening
AA is 30% of the amount actually healed by your crit. The heal arrives approxiamently 1 GCD after the crit lands and heals the lowest health member --determined by the lowest percentage of total health-- of your group or raid regardless of whether the crit target is in your party or not. This does include pets, so you can feel good about saving the ghouls, felguards, and kitties of the world when you're tank is getting pounded.

The delay is the crucial thing for making this ability good and further favors reactive LHWs. When timed naturally (i.e. you wait for a hit to land with tidal waves up and cast when it does), the LHW will land, top off the tank, the next hit comes in the gap, and the AA+ES will hit to fill the gap. The two together are roughly equal to a non-crit LHW at 2k SP, so crits that top off the tank provide a buffer for the water shield refresh.

I have seen multiple people commenting that AA isn't worth taking because its a small portion of a shaman's healing output. This is true if you're primarily using CH, but the amount depends entirely on your crit rate and amount of LHW. I am running 40-45% LHW crit in raids and if I am just LHW spamming, it's normal for me to see 15% of my overall output from AA.

Other heals
Riptide
Riptide is a nice way to proc TW on a single target. When the tank is on an avoidance streak, I'll RT when he gets hit. This is fairly safe because an ES goes off at the same time and I have a fast LHW following. If the boss hits particularly hard and the tank is on an avoidance streak, I'll RT just for the buff+hot and be ready for the hit. I use it just before I refresh my water shield since it gives me a fast LHW to catch up from the missed GCD.

For raid healing, it heals for more than a non-crit LHW does, so I use it on slightly more damaged raid members when I'm raid healing and let the hot+AA take care of them. I also use it as the first heal on a melee dps since it provides TW for immediately taking care of the tank and sets up a big chain heal within range of the tank for later.

Since this benefits from all the single target talents, RT is quite a bit less expensive with a crit heavy build and you can freely use it in conjunction with LHW spam. For pure throughput on a single target, it's not that useful and I normally won't hit it just for the haste when I'm tank healing.

Healing Wave
Healing Wave is primarily useful in combo with a tidal waves+NS macro for emergency saves. HW is poor for general use. With Healing Way completely stacked, HW spam only offers only slightly (~15%) more HPS over glyphed LHW with comparable/lower efficiency. If you can't keep Healing Way up, it's considerably worse on both counts. Note that the Healing Wave glyph only works off effective healng.

Because Tidal Waves only reduces the cast time of LHW and HW and doesn't affect the GCD, HW with tidal waves up is the highest spammed HPS spell we have since LHW is GCD capped. For single target healing, CH and RT's initial heal offer comparable HPS. If RT is already on the target, then CH offers higher HPS. So for tank healing where you're truly spamming, RT->HW->HW->CH->HW->HW is the most HPS you can put out.

Chain Heal
Chain Heal is still a good spell. If you can hit 3+ targets with CH, then CH is generally the most useful spell to cast. Building up a healing set with 'normal' healing stats (i.e. Haste/mp5) is recommended since some raid fights are simply better handled with CH spam.

Stats

The HEP calculations here are done off 2080 SP, 42.3% raid crit, and 15% haste (my current stats) and are based on pure LHW spam.

Intellect
Int is the best all around stat for any shaman healer. You want to gem for int more or less regardless of your playstyle as long as you have any concerns about mana. 100 Int grants (assuming BoK):

• 1815 increase in your mana pool
• 20.4 mp5 from replenishment with a 90% uptime
• 7.25 mp5 from Mana Tide on cooldown (435 mana per tide)
• 36.3 SP to heals
• .726% crit

In comparison to other stats, Int provides ~1/4 the SP, ~1/3 the crit, and ~3/4 the mp5 that an equivalent budget of SP, Crit Rating, and mp5 would provide, respectively.

Crit Rating
For LHW spam, crit is primarily a regen stat and because we get such high returns from imp. WS on LHW spam, it's a pretty good regen stat that scales with haste in addition to being a decent throughput stat. Every 2% crit takes about 10 mana off the effective cost of LHW. At no haste, this would be 33.3 mp5 for spammed LHW, at 15% haste, that ups to 38.3 mp5. In terms of itemization budget, 100 crit rating (2.17%) is equal to 40 mp5. The 2t7 bonus is additive, so these numbers are exactly 10% higher with that bonus. For throughput, 198 Crit Rating provides the same increase as 120 SP or about half the effectiveness.

Haste Rating
Haste provides an excellent increase in HPS per itemization point but no increase in HPM. For throughput, 109 haste provides the same HPS increase as 120 SP, so slightly worse per itemization point for pure HPS at this level of SP. The one thing particularly nice about haste is that it lowers the GCD time of all the water shield refreshes high crit LHW causes. It tends to be itemized along with mp5 on mail because it's useful for increasing CH HPS without generating a lot of overheal on the first target.

Spell Power
With the downranking nerf, SP is no longer the best stat, but it is a very good stat for throughput. You'll get plenty on your gear and the best enchants are all SP, so you don't particularly need to go out of your way to get it. If you decide you have more mana than you need, SP would be the next stat to start gemming.

Mana per 5 seconds
The primary use case for this stat is in speed runs. Individual fights aren't generally long enough for mp5 to deliver more mana than int per point and crit provides comparable longevity on LHW spam along with all its other benefits.

Glyphs
Obviously, the [Glyph of Lesser Healing Wave] is essential to this style. The [Glyph of Water Mastery] (30 mp5) is also very nice because you will have very little mp5 on your gear and a bit of mp5 from another source goes a long way. The third glyph is more interesting. I prefer [Glyph of Chain Heal] for raids and [Glyph of Healing Stream Totem] for heroics. There are very few heroic fights that favor straight CH spam (Loken, the 3rd boss in Oculus if you have melee) while I find a use for Healing Stream in almost every instance.

Metas
The [Revitalizing Skyflare Diamond] provides only a 3% increase in crit healing. If you're familiar with the [Chaotic Skyflare Diamond] family, you might expect more, but that's only the case when you have talents that directly increase the value of the crit bonus. This meta is still the strongest bonus for throughput given a healthy crit rate, but its advantage over the [Ember Skyflare Diamond] is minor. The [Insightful Earthsiege Diamond] would be the best for longevity.

This post brought to you by the letters L & W, the number 9, and the joy of holiday travel.

Update 11/29: I'd initially thought that gheal spam outdid LHW spam, but I'll now say that they're comparable when you ignore WS refreshes.
Update 11/30: Finally got a parse with 2t7, post updated accordingly.
Update 12/06: AA seems to be percent based targeted rather than absolute
Last updated 2008-12-06, patch 3.0.3

Very nice post! I'm personally very happy to hear about AA, as some "sources" have reported that it only does mostly over-healing, which sounded absurd in my ears. (I don't know, maybe if no one else takes damage and they keep their tank topped at 100% all the time, but then they don't really need the extra healing for the moment heh).

I haven't tried LHW spam out in a raid setting (nor am I likely to with our Paladin attendance) so can anyone who has answer a couple of questions:

• How does a LHW spamming shaman compare to a Beacon of Light spamming paladin doing the same job (for instance, on Patchwerk our paladins were over 5000 HPS)?
• How do LHW specced shaman perform compared to CH specced shaman in a pure CH raid healing situation (think Instructor Razuvius for a current example)?
• How do LHW spamming shaman compare to Druids or Priests (disc or holy) in the MT healing role?

I just can't get over the feeling that while we'll be better tank healers as LHW spammers than trying to HW with a raid healing spec, we'll still be the least optimal choice of the 4 healing classes.

I went through the WWS logs of all my WotLK raids so far, and AA has an average of 7.8% overheal for me, 10 man and 25 man naxx counted (No wws for malygos or sath)

I don't believe this is the case.

The clearly worst single target healer in WotLK is the Holy Priest. One thing the OP got wrong was in mentioning that LHW spam couldn't compare to Holy Light/Greater Heal. While LHW spam doesn't stack up against Holy Light, the entire structure of Earth Shield/Earthliving/LHW beats Greater Heal spam for throughput.

Discipline Priests aren't much better off. They can slightly beat the LHW-spam in terms of throughput, and have a decent ability to react to damage with instant healing. However, the entire constellation of Resto Shaman abilities - including HoT, reactive healing, and buffs - means that a Resto Shaman would likely be chosen over a Discipline Priest for inclusion in your raid.

Comparing to a Resto Druid is a bit tough. However, we can probably assert that the two are 'equivalent' in some vague way for single-target healing in a raid. I should also note that Discipline Priests, Resto Druids and Resto Shaman are not 'self-stackable'. Having a pair of any of those specs on the same target significantly downgrades the value of the second healer.

Holy Paladins are, on paper, the best single target healers. But in practive, their healing structure just doesn't "work" that well for 25-mans. Their efficient healing is extremely slow - you might as well just have your Druid tossing HoT around. Holy Light itself is just too big and the casting time reductions it gains are too random. Additionally, your raid doesn't gain much of anything from having the Holy Paladin heal the tank. No proc buffs, nothing setup for future healing, etc.

So in terms of 25-man raid healing, I'd actually place the Resto Shaman between #1 and #3 depending on what considerations you weigh most heavily.

Looking at the numbers LHW with imp water shield and Earth Shield on the target is very efficient even not counting AA. Also it looks like the HPS of LHW with ES on the target is higher than the HPS of HW assuming no tidal waves and no healing way. Also on the 5k HPS note HW is almost 6k HPS with tidal waves however the difficulty of achieving this in a raid setting is not lost on me.

Things get more interesting if you factor in ELW healing into other heals but as with most hots they tend to get overwritten and not really used to their fullest. I have things fairly well broken down in resto sheet.

2100 Healing power 20k mana 19% spell haste 40% crit

Heal name	Mana cost	Cast Time	Avg.Heal	Heal/mana	Heal/s	Mp5 used
Chain Heal 7 (1 target)	714	2.10	4,478	6.27	2131.96	1699.83
Chain Heal 7 (2 target)	714	2.10	6,716	9.41	3197.94	1699.83
Chain Heal 7 (3 target)	714	2.10	7,836	10.97	3730.93	1699.83
Chain Heal 7 (4 target)	714	2.10	8,395	11.76	3997.43	1699.83
Earth Shield 5 (8 charges)*	659	1.26	19,379	28.31	15378.55	1785.53
Healing Wave 14	860	2.10	7,693	8.95	3663.15	2047.41
Healing Wave 14 (w tidal waves)	860	1.35	9,078	10.56	6723.54	3184.68
Lesser Healing Wave 9	442	1.26	4,727	10.69	3751.35	1753.79
LHW 9 w. Earth Shield	442	1.26	5,673	12.83	4501.62	1753.79
Riptide 4*	567	1.26	7,330	12.93	5816.49	2249.77
Riptide 4 w. ELW*	567	1.26	7,904	13.94	6272.08	2249.77

*Note the HPS for hots is calculated as total healing / cast time

• I solo healed (minus a renew+PoM bounces which I traded for Riptides) the hateful tank on our first 5:45 kill and potted because I heard he hit harder at 20%, but that seems to be only the MT so I finished at half mana (had around 19k mana raid buffed and 31 mp5 from gear). I believe the overall output was in the 3.1k range but I didn't save a parse. I healed a number of farm/alt runs on my pally and LHW spam is much better than BC FoL spam (I've heard the coefficient is higher in LK).
• While I haven't done Razuvius (nobody in my guild has done naxx, so we're learning one wing per week doing spider then abom), CH vs LHW for raid healing depends on whether the raid is dense enough for CH to jump 3-4 times consistently. I have a haste+mp5 set for CH spam but I've only put it on and spammed CH for Grobbulus.
• Haven't ran into comparably geared disc priests or druids. Compared to Holy Priests, LHW spam keeps the tank at higher health due to heal frequency, the tank takes less damage due to higher inspiration uptime, and the spam throws off 2k raid heals on a regular basis. The downside is that going over the LHW throughput isn't great. I haven't had this bite me yet, but I decided that ~40% crit was enough for me after the Patchwerk fight and plan on stacking haste from here out and picking up more crit from int.

As for LHW spam vs HL/GH, I've seen both holy priests and pallys drop a serious amount of healing when they really have to and I don't feel I have the ability to do the same. As for the cast time reductions for HL, you just keep LG hot by dropping HLs on a regular basis and the only time variance is on HS crits, which is no more random than LHW crits.

I guess what I'm getting at is if a single paladin on patchwerk can do over 10k HPS with full use of Beacon (Wow Web Stats) and still achieve over 5k HPS in a single tank beacon-unfriendly fight like Sapphiron (Wow Web Stats) how would a LHW shaman compare? Does the (presumably) lower overheal help bring the classes towards parity?

I think you can see from the chart above there is no reasonable way to get that kind of HPS in a typical raid setting. The theoretical max is still below 5k HPS (with LHW spam) so taking into account overheal its not really conceivable that we can reach that kind of HPS with our current gear. I did leave out ELW procs and AA procs but I don't really see that closing the gap. If people want I can include those numbers as well. The ELW numbers are already in my sheet I just didn't include them for comparison reasons.

As I mentioned, we can't compete with HL spam for throughput. Direct comparisons are about as useful as comparing CH spam vs FoL+beacon for raid heals. If you take out the overheal in those parses, the effective HPS is around 4k, which is roughly where LHW spam is. I anticipate that we will see fights beyond the range of comfortable LHW healing where you will have a real pally advantage over shaman but I haven't seen any evidence of that in Naxx/Archavon/Sartharion.

From Previous Resto Thread:
Vuldunobetra posted:

HPS
1. 2941 Druid Nourish w/HoT
2. 2684 Shaman LHW+G
3. 2603 Priest Flash Heal
4. 2603 Priest Flash Heal+G
5. 2509 Druid Nourish
6. 2271 Shaman LHW
7. 1921 Paladin Flash of Light

HPM
1. 9.37 Paladin Flash of Light
2. 7.17 Druid Nourish w/HoT
3. 6.43 Shaman LHW+G
4. 6.24 Priest Flash Heal+G
5. 6.12 Druid Nourish
6. 5.61 Priest Flash Heal
7. 5.44 Shaman LHW



link to the post:
http://elitistjerks.com/f79/t27030-r...47/#post960153

Hmm I don't know how this person got 5.44 HPM but as you can see in the chart above LHW is quite abit above that factoring in imp WS. Even lowering it down to 1500 spell power it still about 9 HPM and if the target has ES then it raises up to almost 11 HPM. I think to have an apples to apples comparison you really need to include cit and all other relevant talents. I'm assuming that these talents didn't include crit for the shaman and pally numbers in particular are off. I think this is a useful comparison but only if done properly. I can supply any of the shaman numbers but I don't know how the other classes work well enough to run numbers for them.

It's really, really, really worth mentioning somewhere: if you don't get to have Earth Shield up on the target, things change a lot. This is not at all included in the original post and will end up causing confusion. If you're the second shaman healing a target, your strategy is going to be different than what is posted here; that needs to be pointed out in some fashion.

I was originally going to pan the entire post because LHW spamming has traditionally been a very horrible way to heal, but the synergy between water shield charges being consumed and the glyph is going to make me take a second look at incorporating it more fully in to my usual cycle.

I'd happily include whatever, but I don't understand what's going to be different. The efficiency of LHW is driven by crit and the glyph is just a (really nice) bonus. My spreadsheet including AA is showing efficiency between 2 and 3 target CH even without the glyph, though CH provides higher throughput on 2+ targets and this will be exaggerated when I can get another 120-150 SP on LHW (I don't pvp so I don't expect to get the deadly) from the s5 totems. My question is: When would this happen? You'd have to have more resto shaman than tanks AND not have the excess shaman on raid heals, both of which seem unlikely to me. Should I just make it more clear that you should have a backup haste+mp5 set for CH friendly fights?

I'm not sure I'd trust those old numbers. My understanding of the formulas at that time was weak.

The one big difference between mine and Daidalos's numbers, I think he is including the IWS mana return into the cost of the spell, whereas I didn't. I think including it is probably the right way to do it.

But the big problem I see is refreshing the water shield. With 40% crit, the shield will proc all 4 charges every 10 GCDs. Does refreshing the shield cut 10% out of our effectiveness? If we don't refresh it, the mana efficiency turns horrible again.