Life in Group 5 – A Resto Shaman Blog
A resto shaman perspective on raiding

Shaman - General

November 23, 2009

The Haste Quandary

More articles by »
Written by: Vixsin
Tags: ,
Effect of Haste on Chain Heal and Lesser Healing Wave Cast Time

Effect of Haste on Chain Heal and Lesser Healing Wave Cast Time

A recent post on MMO-Champion’s Shaman forums got me thinking about Resto Shaman’s current stat of choice, and if there was such a thing as too much haste. After the geek in me quieted down about the chance to make an excel spreadsheet, I set about finding the answer. A quick interweb query netted me the formula used to calculate hasted cast time and armed with the base times of CH  and Lesser Healing Wave, I set out plotting the results.

The result: when it comes to Chain Heal (within the confines of the existing game) you cannot have too much haste. Lesser Healing Wave is a different story, since you will bump up against the GCD at a little less than 1300, exclusive of Bloodlust. (The minds at EJ corroborate this finding.) But in the absence of another Tier 8 styled set bonus, even factoring in Bloodlust and class haste buffs (+5% haste from WoA and +3% from Imp Boomkin/Ret Aura) we won’t be approaching the GCD with Chain Heal any time soon.*

All was well, the excel geek in me was satisfied at his chance to graph WoW-related information, but then the financial geek piped up and asked, “But if haste has a depreciating effect, at what point does the cost outweigh the benefit?” (In other words: at what point should a shaman stop stacking haste because some other stat becomes more valuable?) Time for another graph! This one shows the relative benefit of every point of haste rating in terms of the cast time decrease of Chain Heal. Note that the vertical scale is shown in hundredths of a second per 1 haste rating.

The Cast Time Benefit per Haste Rating

The Cast Time Benefit per Haste Rating

Thus we see that the gain from haste rating decreases significantly as your total haste rating accumulates. But how does this compare to other stats? The valuation of haste relative to other stats is something that both Daidalos and Stassart both address in their respective Shaman spreadsheets. And although they have been able to calculate the relative effect of stats in terms of your playstyle, the fact remains that it would be incredibly arduous to unilaterally demonstrate a quantifiable correlation between haste and increased HPS. Yes, trending will tell us that if HEP (Healing Equivalency Points) fall along a consistent and predictable path then we can plot the point at which another HEP value will overtake Haste. But, this begs the question, is a faster CH always the answer?

It’s a question I’m not sure I have a good answer for, because every time I try to answer it I wind up with a handful of asterisks on the statement. Part of me has believed that when we start talking about non-progression content, a faster CH simply allows me to beat other healers to the punch. If I can heal someone up to full before they can, then I get to see a nice increase in my HPS. Hooray me! These days, I’m lucky to be part of an amazing healing team, one that is not only consistent but also incredibly reliable in terms of output. So, this brings up some other interesting points about healing and HPS:

  • As healers we are capped in terms of HPS by incoming damage and the output capacity of our healing team. Shouldn’t I be gearing based on the damage qualities of the fight and the healing characteristics of my teammates? (EG: If I can count on our holy priests to be able to instantly dish out a CoH after Gormok’s stomp, shouldn’t I be looking to cover the differnence between the total damage taken and the amount of healing from the priest’s CoH, instead of simply sniping their heal?)
  • Instant casts will always be faster than a time-based cast, so when trying to measure up against CoH/WG, is stacking haste a futile effort? (See above example.)
  • Lastly, is stacking haste really dealing with damage in the appropriate way? Excluding mana considerations, is a faster heal always better, regardless of power and sustainability?

Unfortunately, it looks like the end of this post will only yield more questions and no answers to go with them. But I guess for me, the ultimate conclusion is—there’s a difference between mindlessly stacking haste and understanding why to mindlessly stack haste.

* For those of you who are interested, CH will not reach the 1sec GCD until a shaman has about a 4300 haste rating. No, I’m not going to calculate the GCD reduction at that point as well; I’m leaving that to some other nerd!


  1. If I stack SP, a lot of SP gets wasted in my overheal.

    If I stack haste, I snipe other peoples heal, thus reducing my overheal but not at all improving the overall efficiency of my healing team.

    Both statements have some truth in it and some flaws.

    SP is extremely valuable whenever I’m not overhealing. Just think about all the RTs we throw out at targets suddenly dropping low. In that situation SP is so much more valuable than haste.

    Yes haste will snipe some heals. But in the content I’m raiding, I’m thankful for every tiny bit of cast reduction on my chain heal. Especially in movement heavy fights or in fights were 1-2 healers are incapacitated for some seconds, I value haste a lot more than SP.

    I think the biggest argument for haste, is that you don’t have to make too much compromises in the department of spellpower. Even when stacking haste, we’re still making huge steps in SP as well by getting loot of higher iLvL. Choosing haste isn’t mainly a decision against SP, but against mp5 and crit, also to mana regen stats, if mana isn’t an issue.

    The only thing we have to keep in mind is the GCD-haste cap.

    Boomkin Aura + WoA get us 8% haste, to cap the GCD to 1 second we would need 1275 haste rating (something like that, too tired for math). At this point RT (=GCD) and LHW (casttime=GCD) would stop to profit form haste (LHW would of course still land faster).

  2. Monsieur

    But haste feels so good!! And I see your point, and I would really like to know when haste stops being the best stat and sp takes over. Was haste going away for cataclysm by the way? That will be really weird to go back to, if that means chain heal will take the full cast no matter what, and riptide and the t10 bonus will rock the house for all eternity.

    Seriously, the t10 bonus has to be a huge hps bonus, and riptide will never ever be of cooldown again. Has there been done any theorycrafting on it yet?

  3. I agree; haste is a particularly tasty stat. And based on what I’ve read thus far, it will endure when Blizzard revamps things in Cataclysm. (Melee are rumored to even be getting a slight buff with regards to the benefits of haste.) Personally, I couldn’t imagine life with a base-time CH; I think I’d go loony.

    As to the T10 set bonus, I am simply drooling in anticipation for the graphics but somewhat reserved about the bonuses. The 2piece really discourages blindly spamming CH, which I’m all for, but I’m not sure the 4piece will be as powerful as everyone is thinking. Although I haven’t seen much theorycrafting about it yet, I would anticipate that the hot will have a fairly high overheal percentage, similar to Earthliving.

    But hopefully, next Tuesday will be an early Christmas and before long I’ll have some solid data to either reinforce or disprove my hypotheses.

  4. Monsieur

    Ah, ok so haste stacking continues in cataclysm then I guess. Agree on the 4 piece bonus tho. Unless shit hits the fan, you won’t let people slowly tick their way to a full healthbar. Exceptions would include raid damage auras. And maybe if the damage on the tank would be less spiky and more consistant, and punching a ch into the tank to spread onto the melee would let the hot part actually heal something, as opposed to the way it usually is now where your tank takes a 30k hit every 2 seconds. That would actually be pretty cool if there was a boss that just put some serious dots on the tanks, and the healthbar decreased fast and steady every second, and the tank healing would favor lots of hots and smaller heals.

  5. The second graph is a bit misleading because the cast time benefit from haste is relative to the cast time of the spell. A 1% decrease in cast time will obviously net a larger amount of time when casting a slow spell as opposed to a fast spell and that is essentially what your chart shows. What it doesn’t show is that the absolute time gain does not depend on the cast time of the spell (barring GCD cap) simply because you can cast more of the fast spells in the same time as a slow spell.

    Example (number exaggerated to make the math easier): 1 heal for 10k with 4s cast time and one for 5k with 2s cast time (identical HPS of 2500). 10% haste makes that 10k/3.6s and 5k/1.8s respectively. On the larger heal we gain .4s while only gaining .2s on the slower heal – the throughput gain is absolutely identical, however.

    I would guess that tracking HPS instead of cast time decrease on the Y axis would show you that there are no diminishing returns on haste before the cap.

    Nice blog so far though :)

  6. Excellent point, Scrusi. You are absolutely correct in your supposition–if I were to instead plot theoretical HPS against cumulative haste value we would in fact see a linear relationship. But, I think that graph would be misleading too. For example, my current haste rating sits at around 1100 unbuffed. Using calculations from Daidalos’ spreadsheet to generate an average CH value based on my current gear (19,753, per 1.87sec) this nets me an impressive, and at this point unrealized, 10,563 HPS. Do consider, though: this value ignores the fight’s limiting factors (incoming raid damage and healer capacity), while assuming that overhealing is a constant percentage (technically 0 in this case.) So it doesn’t truly represent the effects of haste either.

    I appreciate the counterpoint, though. Keep ’em coming!

  7. Rhom

    One of the things I like about a haste heavy build rather than a SP heavy build is that it gives you finer control over where your spells go. Often before I started loading the haste, I found I had to interrupt my casts more often as another class with an instant cast spell had already healed my target while I was waiting for my spell to go off.

    Obviously this depends on the fight and your healing role (works much better for non-predictable raid healing than predictable tank healing). However, even on fights like twins where the raid damage is very predictable, I find it works better with a haste build since I can change my main CH target more frequently to the person with the lowest health.

    Focusing on haste makes it feel like you have a more surgical strike, but again, it really depends on the fight and what your healing role is to be able to say whether this is a benefit. (I’m also only around 900 haste, so I don’t know first hand what higher haste values play like.)

    Love the blog so far, looking forward to more posts :)

  8. […] One conclusion I draw from the graph is that when it comes down to a fight where I’m spending a majority of my time spamming CH, haste is invaluable; short of trading a ridiculous amount of spellpower for it, I’ll continue to stack it. In the cases where I’m predominately casting other spells, the decision gets a little more dicey. Vixsin, from LifeInGroup5, recently made this post about HW which is more than just tangentially a post about haste (though, he also made a post specifically about haste here). […]

Leave a Reply

Your email address will not be published. Required fields are marked *

CommentLuv badge