In watching this argument over the past few pages, I feel a little bit that you've gotten so caught up in wanting to argue that haste is a superior stat that you are no longer objectively analyzing the math that has been provided to support why giving up mastery, and to a lesser extent spirit, in favor of reaching the higher breakpoint is not necessarily ideal.

At the risk of sounding uppity - if people in your raid are dying regularly, it's unlikely that the main reason for the death is because they didn't have one more tick of Rejuv or Wild Growth to save them, or that .004 seconds off your Regrwoth cast would have been the difference between survival and death. Rather than continue to push an argument that several people here have advised you is mathmatically unsound, and offered you a lot of feedback on (including one of the most respected mathmatical members/theorycrafters of the WoW community), I might humbly suggest you re-evaluate why your raid team is meeting its demise and go from there.

I think that debate and dissent are fantastic things that can build excellent discussions, but I also think that you have to be careful not to fall into the trap of becoming so engrossed in your argument that you become obtuse to others.

On the haste vs mastery/spirit argument, I'm yet to see any modeling, math and/or logs, regarding the impact on overhealing. 10% less overheal on Rejuvenation alone (from 35% to 25% for example), or a 5% less overheal overall, would make a big difference.

Well reforging to reach 6652 is arguable, but I can't see why you consider this breakpoint being not useful. WG has always been a significant part of our healing. Care to elaborate?

It has been posted a few different times already. The amount of haste needed to go from 3043 to 6652 only gives you a ~4% increase in healing whereas that same amount of secondary stats put into mastery gives a bit over 6% increase. Changing your gear just for WG is never going to be worth it unless our class is changed. It will take getting extra ticks of tranquility and RJ (very close together) to make gearing for haste worthwhile, hence 13163 where you pick up a block of extra ticks.

Tecton might be able to confirm, but I think the sheet's main calcs will work in 64bit, but the armory import macros won't. So if you put in your gear you should still be able to play with it.

I think we want to be clear about terminology here. Mastery and Int are both linear in that your gain from either should be constant as long as nothing else changes (putting aside the small amount of crit you get from Int). But since the effects of mastery and Int multiply together, getting more of either makes the other stronger. So the amount of Int or mastery at which they cross will change as your gear level goes up. If you're at some optimal point where the values are equal, then increasing Int/mastery roughly in proportion will keep things that way.

Correct. Depending on what year's version you have, you may need to change a line of code to stop a popup, but it'll work. PM me for details or have a quick scan in the Wrathcalcs thread.

The raw spellpower gain from weapons also affects the INT/mastery ratio, and I can get TreeCalcs to suggest mastery for me instead of INT by just granting myself a 509 weapon, even one without a Sha-Touched slot. A tier difference in weapon would explain a lot of difference in stat weighting.

Biggest problem is probably that that much plain haste on a trinket starts making it hard to reforge down to 3043 once you're in epics, I would think.

I can confirm that it already requires some creative reforging to not be over budget on haste once you start adding in epics, heroic pieces and VP "upgrading". I would also be hesitant about utilizing a trinket lacking any regen at this point, even with a fairly solid set of gear I'm still finding that I struggle to have the longevity for many of the longer or high damage encounters - which, admittedly, may not be as significant an issue in 10s as it is for many in 25s, but I can't imagine it's completely absent to the point that you want to start considering non-regen based trinkets.

I raid 10 man and am 9/16 heroic currently. I've already begun shedding spirit, using some items like tier legs which have no spirit at all. So far the toughest fight for longevity is heroic emperor due to the constant damage and long fight length. It is the only fight where I truly feel like I'm scraping the bottom of the barrel. I typically run about 7-7.5k spirit and raid with a resto shaman so I do get tides.

First, id like to introduce myself, since its my first time posting here since TBC as a HPally, i'm playing a resto druid, on a irl friends guild (thats why im low on progress), but i did my fair share of hardcore raiding in the past, and i know one thing or another about theorycrafting, im by no means an expert, but i think i might contribute a little in due time

I will start by disagree with the said lower gain on HPS for Haste.
Its exactly the opposite, Haste is more efficient then Mastery or even Int (2 to 1 gem budget).

The lower amount of haste per 1% is a good start to look at it, wile the fact that the more haste you get the better it becomes makes it scaling even better then of mastery, which btw has lessened valor the more of it you stack.
Why? Simple, to calculate your additional heal gained from a +1% mastery you need to do this:

HPS * (Mastery%+1%) / HPS * (Mastery%)

The more mastery you have, the lower 1% of mastery will increase your HPS.

Instead, if you do the calculations for Haste it is:

Heal / (100% - [haste%+1%] ) / Heal / (100% - [haste%] )

In this case the more haste you have, the greater haste will be. Another way to show this is this example:

Lets say something heals for 1000 and have 1 sec cast time.

If you have 20% Haste, you will have an HPS of 1000/1*0.8 = 1000/0.8sec = 1250 HPS
If then you add +10% Haste you end with an HPS of 1000/1*0.7 = 1000/0.7sec = ~1428,57
So 10% of Haste gives you 1428.57/1250 = ~1.1428 of your old HPS, or 114,28% or a increase of 14,28%

Now lets change it to a 50% initial Haste = 1000/1*0.5 = 2000 HPS
Now again add 10%: 1000/1*0.4 = 2500 HPS
So 10% of Haste when you already have 50% gives you 2500/2500 HPS = 1.25 of your old HPS, or a increase of 25%

So Haste ends up being better the more you have different from Mastery, that is worse when stacked.

Still the calculations i have made shows that right now for T14, a balance between Mastery and Haste is the best possible RAW HPS. I lack the programming ability to make a excel sheet that can calculate the perfect balance between both of then, but im sure someone here can do this job.


The problem of haste as everyone here already knows is that we lose mana efficiency when we stack haste.
For a pure HPS situation haste is great but when mana is involved its the worst stats because it contributes 0 to your mana consumption.
The idea that its mana negative is wrong, it will keep your spell having the same HP/M, and it will not make you waste more mana because you are casting faster as some people might argue, you will still heal the amount necessary to deal with the present damage, haste in fact will help you LOWER your overhealing not increase it.
The only real HP/M Reduction comes from the fact that each point expended in Haste could be expended in Int/Mastery to increase the HP healed per spell so increasing the HP/M.

The problem with this kind of math is the fact that its very rare for a heal to be enough to top someone exactly, so to make you less mana efficient because your heal is healing by little less, all your heals would need to heal fully, every time a HoT continue healing someone at full HP the INT/Mastery increased HP/M will be lost and haste would had been better.

The problem is that we cannot safely make a guess, or a spreedsheet for how much we will heal fully, and how much is overhealing, and things like that. We can try to work with some WoL numbers to see if we can find a pattern that can be calculated, but i feel that this will be a major effort if at all possible.


Right now, im not advocating on Haste, up until the last few days when i started doing some math on haste i was gemming full spirit/mastery like most of people here. This week i will try raiding with the next Haste cap to see how it goes, and how it will hurt mana wise.

The problem for me right now is that i needed to drop ~2k of spirit to get the next cap. This is a real loss of number of possible casts/fight, and i dont know if this will make going haste a problem, i need to test more.

Outside of the loss of spirit if your gear cannot reach the next cap, i found that reaching a new cap of haste even 6652 worth it because you can keep your mastery and haste more balanced at around ~15%, and math showed that this is the best course of acting right now, and at the same time get another tick to WG.

People here tryed to imply that the only gain from haste was the WG gain, when in fact all your HoT's healing faster is an obvious gain of HPS. When the RAW HPS gained from HoTs ticking faster shows to be more HPS increase then Mastery, and this is not even calculating the extra tick of WG, i think that it proves that haste is way better then people were implying.

If my math is wrong somewhere please en-light me so i can avoid wasting more time researching this next raid, but the fact that 425 Haste gives 1% haste, wile 480 Mastery is necessary for +1% Mastery is already enough to notice something is wrong. When you add the fact that haste scales positively and mastery negatively it became obvious to me that i needed to do the math to see how it worked. And when i did, i ended discovering that haste is more beneficial then mastery for full epic character. (Not tried doing the math for 463 heroic)

Now again, i know that this is somewhat a loss in HP/M because of the loss of Healed/Spell, but the reduced overhealing might compensate for some of that, and since we are having the most problems when dealing with AoE Burst damage, atlast thats what i noticed reading and playing, having haste will help top everyone up.

Another thing i'd like to note is that some people think that haste is less efficient because it will not reduce the GCD of Rejuvenation, and i'd like to disagree with that. Think of rejuvenation as X amount of Heal divided between the "Time of the Hot" + 1seg of cast.
Since the hot time is WAY bigger then the 1 sec of GCD, and this whole portion will be reduced, haste keeps being more effective then mastery.

Ex: A 12sec duration Rejuvenation (using a low number to show that even in the worst case its still better to use haste for HPS), with 1sec cast time, the effect of haste over Reju is 12/13 of the normal overall value, so 92,3%.
If you do the math of Haste against Mastery you will see that it varies between 1.1 to 1.4 more efficient then matery.
So even in the worst case possible its still better to increase Haste then mastery for pure HPS.

Now, im not saying that everyone should just go full haste and all, but that we need to consider what is really better, and how much mana starved we are.
The less mana starved the more we can invest in haste instead of Mastery.

I still need to run some calculations of how much MP5 you lose going for haste, and how much spirit it would need to compensate for the loss of mana, and if that amount of spirit + haste is still better then mastery at some points or not.
But like i said before, it all depends on the amount of overhealing, because you just really "lose" mana with haste, if you heal someone, and its not enough because you heal for less then if you had used mastery/int, and you need to use another heal to complete. Since for small numbers i found that this situation is not real, and that we will just let someone with very small lost health just stay that way until it takes another damage so we can overheal less and avoid burning the mana pool, that in fact we will lose way less mana then if we were just doing math of "Healing dummy" that is a infinite health pool that we need to heal as fast as possible to see how much HPS we can get.

In the end my hypothesis is:
" If we use a healing dummy, forget Haste because of the lack of mana efficiency, but in a real raid situation where the real number gains from mastery/int is lower enough that no more number of healing spells will be needed for going haste or mastery/int, and since Haste offers better HP/S, going haste might be more beneficial, it will help avoiding deaths, it will increase your possible output of healing on necessary moments, and will give you a better answer against raid aoe burst damage that is a hole in our toolkit right now. "

Lets see if my hypothesis is sustainable, or if anyone here can show me what im doing wrong, and why haste outside of 3043 cap is weak.

Many tanks to all theorycrafters here that i been reading since MoP launch, and for the patience to read this long and boring post with this bad English of mine. (Not main language, my English is mostly self taught)