Mistweaver Minutiae: A Love-Haste Relationship

I’ve been doing a lot of thinking about Mistweavers lately, as I am determined to learn the spec, gear my Monk, and get Heroic-raid ready. And as part of the thinking process, I’ve also been talking to a lot of Mistweavers – a few who are learning the spec alongside me, a few who are experienced, skilled, and highly-progressed Mistweavers, and a bunch of in-betweeners. Surprisingly, one thing keeps coming up in all these discussions – Haste, Haste, Haste.

Renewing Mist

Renewing Mist is most definitely the weirdest healing spell in the game.

Haste has an interesting interaction with Mistweavers. It’s the only class for which Haste past a breakpoint is actively negative. I find this absolutely fascinating.

Many of the people I have talked to seem to know that being right on the breakpoint is key. Not enough seem to understand why. Quite a few were haphazardly geared and sitting between breakpoints, or worse, right beneath one. It became very easy for me to predict which Mistweavers were suboptimally Hasting, based on their complaints. If a Mistweaver says, “I have trouble getting enough ReM out for Uplift to seem worth using,” my first response is, “Switch to Stance of the Fierce Tiger. How much Haste do you have?” Every time, it was just a bit shy of a Haste breakpoint.

In this article I’ll explore the fairly basic, but somewhat counter-intuitive, link between Haste, Renewing Mist, and Uplift. Given that these two spells constitute the vast majority of raiding Mistweavers’ healing breakdown, they’re the only spells I’m going to be focusing on.

I’ll be the first to admit that this whole post is mostly academic – if you are already sitting at a Haste breakpoint, nothing I’m going to say here is going to improve your gameplay experience. However, it’ll help you understand why you sit at that breakpoint, and I hope this reaches more people who, like some of the friends I made this week, had no idea that this strange “love-haste” relationship exists.

Haste Basics

The benefits of Haste are easy to see on any cast-time, non-Heal-over-Time type spell. As your Haste increases, the cast time T gets shorter, like so:

T(new) = T(base)/(1+Haste)

This is a linear relationship, and it’s very simple. The more Haste you have, the faster you cast the spell, and the throughput of the spell increases (because you can deliver the healing faster, and begin on your next cast sooner).

For many classes, Haste lowers the global cooldown – the forced delay between one cast and the next – in the same way that it lowers cast-times of spells. However, there is an artificial 1-second “floor” on the GCD, beyond which the GCD can no longer be reduced. Even if you lower the cast time of a spell to below 1 second, you’re still going to have to wait out that second in order to cast your next one. This happens at 50% Haste for many classes, which is probably reachable in t16 Heroic gear with raid buffs, but for Mistweavers this is not a concern since most of our spells are instant and our GCD is already 1 second.

For HoT effects, things get a little bit more complicated. This simple calculation now applies to the tick interval, t – the time between the individual healing events that comprise the HoT:

t(new) = t(base)/(1+Haste)

This also decreases the duration, D, of the HoT effect, by cramming the same number of ticks into a smaller period of time:

D(new) = # of ticks * t(new)

Here’s where it starts to get a little bit mind-bendy. When the duration of your HoT effect gets low enough that half a tick interval is missing, and then you add a wee bit more Haste, you get a new tick! This occurs at events we call “Haste breakpoints”. The value of Haste as a stat jumps up at these breakpoints, because now Haste isn’t only delivering the same amount of healing at a faster rate – it’s actually creating more healing! Going from, say, 5 ticks of a HoT to 6 ticks of a HoT is a 20% increase in the power of a spell.

Renewing Mist Haste Breakpoints (with 5% Haste Buff)
Haste % Extra Ticks Required Haste Ratings
Fierce Tiger Stance Wise Serpent Stance
5.57% 1 154 231
16.65% 2 3145 4718
27.75% 3 6141 9212
38.94% 4 9158 13737
49.98% 5 12138 18207
61.10% 6 15137 22706
72.19% 7 18131 27197

The below graph illustrates the value of increasing Haste on several common types of spells – cast-time spells, cast-time spells with a HoT component, and instant-cast HoT effects. You can see that for spells with a HoT component, the value of Haste increases significantly at certain points in the chart – these are the Haste breakpoints.

As Haste increases, spells with periodic effects occasionally gain additional benefit. The vertical leaps in the blue and green line are increases in your spell's throughput due to reaching these "haste breakpoints".

As Haste increases, spells with periodic effects occasionally gain additional benefit. The vertical leaps in the blue and green line are increases in your spell’s throughput due to reaching these “haste breakpoints”.

You can also see that, between Haste breakpoints, adding more Haste still provides a marginal increase in your throughput. This is why it’s usually not such a big deal if, say, you’re aiming for 3,043 Haste, but end up sitting at 3,200 Haste, and don’t have enough wiggle room in your gear to get to the next step on the Haste ladder. So long as you cast anything that has a cast time or a channeled duration, your overall healing throughput will increase as your Haste increases, even beyond a breakpoint. Unless….

The Renewing Mistconception

… Unless your class mechanics are weird. And for Mistweavers, they are!

Okay, so this part isn’t weird. I’ve made for you a graph of Haste’s effect on the duration of Renewing Mist – which would be equally valid for any HoT spell. This will help me illustrate the rest of my points.

Haste's effects on the duration of Renewing Mist. Red lines represent breakpoints; the yellow line highlights the fact that the maximum duration of ReM decreases with each successive breakpoint you reach.

Haste’s effects on the duration of Renewing Mist. Red lines represent breakpoints; the yellow line highlights the fact that the maximum duration of ReM decreases with each successive breakpoint you reach.

So what I want you to take away from this is that at each Haste breakpoint, your Renewing Mist is at what we call a “local maximum” for duration – e.g., a peak in the graph. The longest the spell’s going to get, for a while at least. And between breakpoints, Renewing Mist’s duration is shorter. And as you reach higher and higher breakpoints, your Renewing Mist’s maximum duration is also shorter.

Now, this is important for a couple of reasons. First of all, the longer an ReM lasts, the easier it is to squeeze in Uplifts. But we’ll get back to that in a moment, because there’s a little more to ReM than meets the eye here!

A longer ReM duration also means longer windows of “overlap” between the end of your 1st cast’s ticks and the start of your 3rd cast’s ticks. Because Renewing Mist has a hard cooldown – 8 seconds – your 3rd cast is always going to be 16 seconds after your first cast. So the longer your ReM duration, the wider the window will be where you have 7+ ReMs active on the raid. If you had no Haste at all, the window would be 2 seconds (18-second duration of ReM(1) – 16-second interval between the casts). At a Haste breakpoint this can jump up to almost 3 seconds. And just before a Haste breakpoint, it can get as low as only 1 second! But this window, which I’ll call the “7+ window”, narrows as you reach higher and higher breakpoints.

There’s a second effect of Haste, though: The tick-interval-decreasing function. This is important for Renewing Mist because ReM jumps to a new target only once it has delivered a tick of healing to the original target. The same rule applies to jumping from the 2nd to 3rd targets. So the shorter your tick interval, the faster Renewing Mist will spread.

What does this mean? First of all, it may raise the average number of ReMs out on the raid over the course of the encounter. 6 ReM at all times is very easily achievable and should be your first goal to attain when you’re learning to heal in a raid, but it depends upon your attention to the ReM cooldown more than anything else.

Once you’re casting the spell on cooldown – or as close as possible – increasing your Haste will decrease the average number of ReMs out on the raid, because the 7+ window gets shorter. However, you do see longer 8-ReM windows, and at extreme haste, you’ll note the emergence of 9-ReM windows.

Haste’s Effect on ReM Overlap
Haste % # of Ticks Overlapping ReMs (sec.) Average ReMs
7 8 9
5.57% 10 4.633 2.101 6.807
16.65% 11 3.998 2.291 6.790
27.75% 12 3.476 2.442 6.775
38.94% 13 3.045 2.547 6.760
49.98% 14 2.667 2.667 0.002 6.751
61.10% 15 2.483 2.483 0.139 6.741
72.19% 16 2.323 2.323 0.261 6.732

Although you have slightly fewer ReM HoTs active on the raid on average, the extra tick you gain at each breakpoint will more than compensate for this effect, and your ReM healing will increase as you jump up from one breakpoint to the next:

Haste's effect on ReM healing

Haste’s effect on ReM healing

Note that if you are between Haste breakpoints, you would be better off dropping back down to an earlier one than staying at your current Haste rating. In fact, it’s even possible that if you’re above the 3rd breakpoint, you could be doing less healing with ReM than if you were sitting on the 2nd breakpoint. Seriously, guys, make damn sure you’re as close to that red line as you can be, without going under. It’s srs bzns.

This effect is so dire because being between those Haste breakpoints can drop your ReM duration low enough that ReM(1) runs out before ReM(7) can tick and spread, and you lose the 8+ window entirely.

Hurry Up(lift)

Uplift

Uplift your HPS and make your raid’s Resto Shaman cry.

As I started to discuss before, the number of Renewing Mist HoTs on the raid also determines the number of targets you’ll heal with Uplift. By increasing your average ReMs on the raid, you’ll be increasing the average number of targets you can heal with Uplift.

(In reality, you’re likely to skew your average Uplift targets slightly higher, as you’ll be trying to ensure that you have the most ReMs possible when heavy raid damage occurs, but modeling that is a little more difficult than I am capable of achieving whilst heavily caffeinated and sleep-deprived.)

Interestingly, average Uplift healing decreases slightly as you reach higher breakpoints. That’s because at each breakpoint, there are actually fewer ReMs on the raid on average. So if you really and truly cast Uplift at complete random – with no attempt to maximise ReM targets, other than keeping ReM on CD – you’ll notice your Uplift healing gets less significant as you increase from one haste breakpoint to the next. It is a very slight amount, but it is interesting, right? An increase in Haste making your healing less effective?

The graph below illustrates this effect. You’ll note that you quite drastically gain a lot of Uplift throughput by reaching a breakpoint, compared to the throughput you get between breakpoints. However, each breakpoint has a slightly lower maximum Uplift throughput than the previous one.

Haste's effect on Uplift average healing. Note how awful it is to be between breakpoints!

Haste’s effect on Uplift average healing. Note how awful it is to be between breakpoints!

But as I keep hinting, this isn’t the whole story. This is the story for a beginner, or perhaps an early intermediate Mistweaver who hasn’t figured out how to optimise ReM spreading and holding Uplift for the right time. (I, by the way, am one of these types of Mistweavers.) Going up to the next Haste breakpoint makes it a little more feasible to get Uplifts at 8 ReMs, as you can see in the table below.

Haste’s Effect on Maximising Uplift Targets
Haste % Overlapping ReMs (sec.) Uplifts at 8 ReM Uplifts at 9 ReM
8 9 0ms 200ms
5.57% 2.101 3 2 0
16.65% 2.291 3 2 0
27.75% 2.442 3 3 0
38.94% 2.547 3 3 0
49.98% 2.667 0.002 3 3 0
61.10% 2.483 0.139 3 3 0
72.19% 2.323 0.261 3 2 1

While in each case you can theoretically get 3 Uplifts in during the 8-ReM window (if you have perfect reflexes and somehow exist inside the servers so the speed of light isn’t an issue), I like to consider the idea of at least 200ms of latency in your own human reaction and in the game’s response to your action. In that case, you’ll see that you actually can squeeze more Uplifts in during the 8-ReM window if you aim for higher breakpoints, until of course you reach the breakpoint of absurdity, 72.19% Haste, where you can no longer squeeze in 3 Uplifts during the 8-ReM window, because your last one will be on 9 ReM targets.

Now, of course, you’re unlikely to be able to actually get 3 Uplifts in during the 8-ReM window, because you’d need to generate 1-2 Chi (depending on whether you have Ascension or not) during that brief window as well, which will take up at least one GCD and push you out of the 8-ReM window. But even if you’re only planning to fit two Uplifts in, it’s nice to have the extra cushion to get those casts in. Latency, game mechanics, and your mobility cooldown can all interfere with being able to Uplift at exactly the right time, so any “padding” will help! A skilled Mistweaver should find their Uplift heals more targets as they reach higher Haste breakpoints and they are able to utilise the 8-ReM window a little more often. (I don’t think the 9-ReM window is terribly feasible.)

Synthesis

So, your ReM healing increases, and your Uplift healing decreases. What’s the end result?

Haste's effect on your combined ReM and Uplift healing

Haste’s effect on your combined ReM and Uplift healing

Since ReM increases by a heck of a lot more than your Uplift decreases, if we add the two healing effects together you’ll see that overall, increasing your Haste to a higher breakpoint will still result in greater healing from the ReM/Uplift combination. Yay! But being between breakpoints is still incredibly bad, m’kay?

And there’s another factor …

The F@*$ing Thunder Focus Tea Factor ((TF)3)

Pardon my language, but holy cats, this analysis took forever. I’m not going to go too deep into it, because for the most part, it’s very similar to the above analyses. As your Haste increases, and as we factor in Thunder Focus Tea, there are a few effects that make everything a little easier to do.

Thunder Focus Tea

Thunder Focus Tea: Caffeinate your raid!

First of all, when you refresh Renewing Mist with TFT/Uplift, it extends the duration of your existing Renewing Mists not just to their usual maximum duration, but also by a little more. The extra duration – equivalent to the length of time you had left ’til your next tick happened when you pressed the Uplift button – will be more significant at medium Haste levels, where the tick interval is still long enough that you’re like to refresh ReM(1-3) just after a tick has occurred, rather than just before.

Faster spreading of ReM makes it easier to ensure you have at least 7 ReMs active on the raid when you refresh them with Thunder Focus Tea.

While this is not vital – whether you cast ReM(7) then hit TFT/Uplift immediately, or you TFT/Uplift right before ReM(7) comes off CD and hit ReM immediately, it comes out to pretty much the same effect – this will mean that the Uplift you use with TFT will deal more healing. No point in throwing away healing – so long as you can make sure you don’t lose a ReM by waiting too long!

For 10-player raiders, the next points will mean very little. Sorry 🙂 For you, it’s not that difficult to keep 10 ReMs going with TFT/Uplift, and additional Haste will have very little effect on the length of your 10-ReM window.

For 25-player raiders, this also impacts your maximum ReM count during the heavy damage phases of an encounter. Higher Haste breakpoints allow you to squeeze a little more time out of the 14-ReM and 15-ReM windows, while occasionally reducing the 13+ window. It should always be possible to squeeze at least 3 of your 4 available Uplifts (assuming 4 Chi, 2 x Uplift, Chi Brew, 2 x Uplift rotation) into the 13+ window, but with higher Haste, more of these will land in 14- or 15-ReM windows.

Haste's effect on ReM and Uplift healing, with on-CD TFT usage.

Haste’s effect on ReM and Uplift healing, with on-CD TFT usage.

Being between breakpoints will often remove the 15-ReM window entirely, and can even remove the 14-ReM windows, which leaves you less powerful for healing up during these heavy damaging events. There are a few localised oddities where you get a small spike in healing between breakpoints – still vastly beneath what you’ll spike up to if you’re at a breakpoint! – that are attributable to fluctuations in exactly which ReM buffs drop off at what times. There’s very little point in aiming for these anomalies, since it is always clearly better to be on the breakpoint.

This Article is in Execute Range! Pop Heroism!

Finally, just a brief note about Heroism/Bloodlust (and it would apply as well to the Herbalism cooldown or Berzerking).

Effects that increase your Haste can actually have a weird effect on your TFT/Uplift rotation. If you end up not near a breakpoint, you can lose a lot of your 13+-ReM window uptime. Your 15-ReM windows can disappear. Your 13- and 14-ReM windows can be under a second, meaning that you’re unlikely to get more than 2 Uplifts in at maximum capacity. Now, those ReMs will have a lot more ticks – if you cast them after Heroism was popped, of course – but you’re probably more interested in Uplift’s healing if you have lined up your TFT/Uplift to produce results during a heavy AoE phase.

With that in mind, I’ve prepared a table of the achievable-in-the-near-future Haste breakpoints and the Haste you’ll sit at under Heroism:

Haste Under Heroism/Bloodlust
Haste % Haste % Under Hero Breakpoint Exceeded ReM Duration (sec)
5.57% 37.24% 27.75% 17.487
16.65% 51.65 49.98 18.464
27.75% 66.07% 61.10% 18.064
38.94% 80.62% 72.19% 17.717
49.98% 94.97% 94.27% 18.464

Note that the 16.65% and 49.98% breakpoints produce a Heroism haste that is only a little over an existing Haste breakpoint. I’m not saying that you should necessarily stay at 16.65% Haste just because Heroism, but it’s possible that aiming for 49.98% once you get reasonably geared in t16 is going to be your best bet since you will be operating at near-maximum efficiency even during Heroism phases.

(The Haste breakpoint beyond 49.98% is probably going to be reachable with the +Haste% trinket in t16, or in upgraded t16 Heroic gear, but honestly, I’d consider sticking with the 49.98% breakpoint and funneling the extra stats into Crit or maybe even Mastery. Most of the benefit of Hasting up your ReM is going to be lost to overheal – except when your Jina-Kang procs, I guess? – but not gimping your Uplift healing during Heroism might actually be worth sticking around for.)

Conclusions

For the love of cats, please make sure you are at a breakpoint and not below it!

While I’m here, I’m also going to note – but not elaborate much – that pushing your TFT/Uplift back a second, so that you have 8 ReMs out when you use it, actually increases your 13+-ReM window durations significantly, and will allow you to get more of your 2-Uplift/Chi Brew/2-Uplift rotation in during 14- and 15-ReM windows. However, this relies on high haste to extend the 8-ReM window enough that you can reliably achieve this – and you may have to sit around waiting for it to happen, wasting GCDs. I’d call this “slim but possible” at the 27.75% Haste breakpoint, “achievable” at the 38.94% breakpoint, and “reliable” around the 49.98% breakpoint.

Interestingly, there isn’t really a Haste breakpoint past which additional Haste becomes irrelevant. Even looking solely at Uplift healing – so, removing the ReM additional ticks that bias this analysis so – you’re always gaining extra chances to use Uplift skilfully with more-than-the-average number of ReM HoTs out on your raid. I really thought it would plateau or even perhaps decline, but it’s looking pretty solid all the way up through 116.57% Haste! Haha.

And again: Be on a breakpoint! Remember that most sites give breakpoint information assuming that you are not in Serpent Stance (since mods like ReforgeLite and sites like Wow-Reforge/the Armory do not necessarily know how to handle Serpent Stance’s Haste modifier), and make sure that you are in the appropriate stance when you are doing your reforging. I highly recommend using a mod or website to optimise your reforging and gemming so you do not end up unintentionally lowering your throughput.

Acknowledgements

Many thanks (only partially sarcastically!) to Mist of Simple Math@Andorhal-US for prodding me into looking at this, and sending me down this rabbit-hole of what Haste really does to ReM. I have probably now lost 1/3 of my remaining sanity thanks to you 😛

The guides written by Elitist Jerk’s Hamlet on Haste breakpoints were invaluable for this project, as were the Haste breakpoints & calculations thereof provided by Binkenstein over at TotemSpot. Thanks, you two, for paving the way so I didn’t have to fail to figure it out on my own 🙂

Cheers to Frozenorange of Simple Math@Andorhal-US and Bloöm of Plan B@Korgath-US for helping me cling to the just-barely-salvaged 2/3 of my sanity by listening to me complain about this project for hours on end! ❤

Advertisements

About Dedralie

Stuff about me!
This entry was posted in Monk, Theorycrafting. Bookmark the permalink.

10 Responses to Mistweaver Minutiae: A Love-Haste Relationship

  1. Alexandra says:

    I love your articles; I am so happy that there are people out there that likes to do calculations so I don’t have to! With this in mind I’d love to see a Conclusion/Summary section, a “Dayani’s Healing Brilliance for Dummies” if you will, added to the articles.

    You see, I do many things very well, however, numbers are not one of them. Maths makes my brain hurt, I kid you not. “For the love of cats, please make sure you are at a breakpoint and not below it!” is an excellent example. Sometimes all I’m looking for is simple advice from people I trust. I don’t need to know why I ought to do something as long as a reliable source tells me that they understand how it works.

    Don’t get me wrong, I love the analysis and theorycrafting too (otherwise I would not be reading your blog!), This was just an idea that popped into my poor math-disabled brain whilst reading the blog post 🙂

    • Dedralie says:

      I know that I have a problem with succinctness (or a lack thereof), and I love talking about how to interpret graphs and numbers. Put those two together and you end up with overly wordy and impenetrable posts!

      Your suggestion is a good one and I will try to remember to add a clear and concise conclusion to all future posts. I am really wary of making black-and-white proclamations, though; I worry that my knowledge is not complete enough, my experience broad enough, to be a solid basis for universal truths. So I’ll probably always use too many words 😉

  2. Astraios says:

    I haven’t had much time this summer to keep theorycrafting for t16 (gnorrior on MMOC, which is a site you should visit if you want to do some reading with what is generally the other MW theorycrafters), so thanks for the analysis on haste break points.

    What I’m interested in (if I ever get the time to run numbers or if I can convince you to do it for me >:D) is whether haste or crit will scale better in t16 when we are running 2 14% secondary boosting trinkets. With 40% crit (probably closer to 45% in t16 gear, since im already at 39% raid buffed) and crits doing 260% healing, crit is going to be extremely powerful, but haste being almost doubled between the trinkets and our stance will be extremely strong as well. I’ve been testing on PTR of course, but you seem to have solid numbers so I have no shame in asking you ;D

    PS I like theorycrafting so if you ever want to chat, Id be up for it. And I might be able to help your MW’ing too. PM me on MMOC (I’m gnorrior there) if you want to follow up at all. If not then thanks for the breakpoints here, it’s interesting food for thought for my t16 gearing.

    • Dedralie says:

      Buh I know who you are from forums 😛

      I’m always up for talking to other people about this stuff so I’ll seek you out on MMOC (I don’t really post there much but I do have an account). Your question is interesting and someone else has asked me to do a crit v mastery analysis too, so I should tooootally get around to doing that, but I’ve never done stat weights before (except for a small dabble with disc priests) so I’m reluctant. I’ll definitely think about it and see what I can come up with. There’s of course value in having multiple people evaluate the same questions and then peer-reviewing our conclusions 😉

      And hmm I may well have you go through my mw’s logs and tear me apart! But only if you promise to be truly brutal. 🙂

  3. Pingback: Healing Theory, Part 5: Haste | It's Dangerous to Go Alone

  4. Pingback: | The Not-So-Layman’s Guide to Haste in 5.4

  5. Luq says:

    Hey Dedralie,

    I thoroughly enjoyed this read. I had already looked at haste’s negative implications in mw healing previously but you seem to provide very solid arguments with evidence to support it resulting in a pleasant and well-written page that I would recommend for both beginners and experts in the monk specialisation. I was mainly interested on how uplift healing gets affected by ReM’s duration and how your graphs expressed this relationship. Obviously we’re talking about milliseconds which may or not impact gameplay in a practical world but your post clarifies how poor itemization or TfT timing may result in less average ReMs on the raid which leads to less uplift healing.

    Finally, I confess that I admire your passion for all the healing specs because part of understanding how to improve on your own healing spec includes knowing and being able to understand how the other specs play. 😛

    • Dedralie says:

      Hi Luq, thanks for your kind words 🙂

      Definitely all of the points I covered in this post are very, very tiny factors compared to mastering the basic mistweaving competencies, but I’m glad you found it interesting to read, because I certainly found it interesting to think about! I have to confess that I found a bug in my model that doesn’t change my conclusions, but will change the graphs somewhat, when I get around to finally reconstructing the data I lost with a careless “No I don’t want to save this file grr go away!” click 😦

      Hmm I want to accept your admiration but I have to admit that I have little passion for Holy Paladin in this expansion so I feel kind of dishonest! Haha 🙂 I do agree that one excellent way of improving your healing is in trying another class; sometimes learning how another healer’s mechanics help them overcome certain obstacles can give you new ideas on how you can do it with your own class! I also think that playing DPS and tanking roles is helpful as well – perhaps even moreso – but I just don’t have the energy to keep up with that in this expansion. So many dailies…

      I’m thinking about doing another post all about the intricacies of timing TFT, I dunno if it would be interesting to anyone other than myself, but once I get an idea in my head it usually manifests as a huge spreadsheet and 5,000 words of rambling 😉

  6. Pingback: Blizz Master | Kung Fu Healing: Understanding Haste

  7. Pingback: Warlords of Draenor Mini-Analysis: Mistweaver Edition | healiocentric

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s