It just went live. Check out the Theorycrafting Think Tank.
Druid:Balance is our little corner there.

Greatly appreciated. I can only image the complexity of the system you're gunna need to solve for the break point for haste. I'd offer my help, but to be honest I think you surpass me in what I know about math.

The thing is, you can't have a threat problem AND be at the bottom of the dps charts at the same time. Or at least, not after the first 20 seconds of the fight. Try to show us a WWS or combat log of your fights, that might help because at the moment i don't see where your problem can come from.

[Starkiller's Bauble] is significantly better than the [Terokkar Tablet of Vim], and can be obtained solo. Heck, next daily run you make to Netherstorm for the Sunfury plans, take a guildmate with you and kill the big blob that drops the quest item. It's worth it.

I have lately been utilizing this good-awesome spreadsheet to great effect, once I decided to try and gear up for balance instead of my usual resto raid role.

I have (at least) one question to this, though: Everyone tells me to get hit-capped ASAP, but when I input 152 hit rating instead of my current 136, the calculated increase in dps is very marginal (~5dps). Is this due to some fact I am not aware of, possibly covered somewhere in the comments (I so far wasn't able to find anything on this in the comments, so I will apologize if I missed it somehow) or is it more a reflection on the fact I am only just gearing up (i.e. sup-par gear)?

In my trinket slots I am currently using bangle and darkmon card: blue dragon, because my current biggest issue seems to be lack of mana , at least according to the spreadsheet

Also, if it is ok to ask this question here (haven't so far been able to find a better place) I would greatly appreciate some advice on what parts of my gear I should improve next. I know at least my horrible shoulders needs a fix, as well as "real" caster gloves, and possibly a better healer belt so I can gem my current for damage instead of my resto role. (I do realize as well, that I should not gem greens, I will change once I find out if balance is soemthing I am going to stick with, and when I get more money of course).
My guild is a family sized thing, meaning I will most likely never get into 25man stuff with it.

Armory: Bloodthorn

Thank you for your time

I've been getting some odd results using the v0.80b spreadsheet in OpenOffice in Vista to try to gauge the relative values of hit/haste/damage:

At 21% crit, 11% hit, 1300 damage in the projected character sheet with 250 haste rating and no set bonuses, I'm getting approximate values of 1 haste rating = 1.31 damage and 1 hit rating = 1.24 damage

It seems rather implausible to me that haste rating is better per point than hit rating... so what's going on? Is it an OpenOffice issue, or do people in Excel get similar results? Or is haste actually better than hit even for some reason I don't understand?

(I computed these manually by adding to the 'For "DPS Comparison" page' row on the "Gear Select" tab and then looking at the "Character Sheet _ Buffs" tab; using the "Calculate DPS Ratings" button on the "DPS Comparison" tab gave a script error and didn't update the DPS ratings. I get similar results with just Starfire as with MFx2,SFx15)

Really wish I wasn't sick right now or else I would go into detail, but you don't want to stack crit until you get an absolutely retarded amount of spell damage. Sprll crit itself isn't bad, but it's just very expensive item point wise.
for 1 % crit, you can get about 1.4% haste or 26.52 spell damage (and remember we get a 1.32 starfire modifier on top of this)

Bloodthorn: Your spells may just not be hitting hard enough for Hit to be the best stat for you. You're inside, or very close to, the grey area for Starfire, where Hit and Damage switch places as to which is best based on your gear. Haven't done the math yet on other spells, but I imagine they're fairly similar. Also, check whether you have Totem of Wrath or Draenei racials checked in the buffs page. With a Draenei, you can subtract 12-13 hit from your cap, Totem of Wrath subtracts another 36-38 (depending on rounding). Lastly, The gap between 136 and 152 is only 16 rating, or about 1.3% hit. If you're only getting 500 DPS, that's a 5 DPS increase.

The spreadsheet doesn't yet account for the 2.4 spirit changes (or at least my copy doesn't, can't remember if I updated recently enough), but if mana is an issue for you in live as well, stack MP5. When you're mana-starved, MP5 is your best stat, bar none. If you can't cast, you can't DPS. Get a Spriest, chug pots, become a JC and get yourself an owl trinket, do whatever it takes. Once you can sustain a rotation for a whole fight, only then should you worry about making your spells do more damage.

Ambiorix, it's perfectly possible for Haste to be better than Hit, although I'm not sure exactly given your numbers. Both Haste and Hit have diminishing returns, so with low Haste you'll get a lot more benefit from a single point than you will with a point of Hit when you already have a lot. Look back a couple pages for a similar discussion of Crit versus Haste. Lamp, you might enjoy that as well.

A couple of questions to the general theorycrafters out there. I have a DMC:Blue Dragon as well sitting in my bank from back at 60, as I've been keeping it for an offhealing set if I ever need it. Bloodthorn sparked my interest in it, and with rough calculations using a rotation of IS, MF, SFx4 with an average total cast time of 14.5s, it works out to an average of a proc every two minutes, or 12.5% uptime. At my gear, that averages out to about 30 mp5, which is not insignificant. Anyone see any issues with this or know if the DMC:BD has been 'fixed' to not work for level 70 toons as well as level 60 toons? Otherwise, I'm thinking about using it on long fights without an SP. It also seems like it would scale fairly well with extra gear.

Second question, and this more in the vein of the upcoming math posts. I think I've finally decided on an overall Average DPS function for Starfire, but I wanna make sure I'm not missing anything before I move ahead with it. Does anyone see any problems with this function, assuming a two-roll system?

Assuming the following: Fully talented, is the base damage after Moonfury; is our damage coefficient after Moonfury and Wrath of Cenarius, 1.32; is our Damage Rating, or +spell damage, including the bonus from Int; is the Crit Rating coefficient, 1/2208 = .000453; is your equivalent Crit Rating after decomposing any percentage-based bonuses such as Focused Starlight or Totem of Wrath; is the Hit rating coefficient, 1/1262 = .000792; is your effective Hit Rating, again after decomposing any percentage-based bonuses; is the Haste Rating coefficient, 1/1576 = .000635; and lastly is your current Haste Rating.

[edit] Changed rating coefficients (except Damage) into percentage-based numbers, i.e. 12.6 -> .126
[edit2] Updated the function call to specify variables, though it was pretty evident anyway
[edit3] Increased accuracy of coefficients
[edit4] Bad Ador, no biscuit. Forgot to reciprocate rating conversions.

Just a couple questions on your equation,

Most moonkins use a CSD, and it makes crit a 209% modifer instead of a 200% modifier
remember hit chance caps at 99%, and a 100% crit chance will only lead to 99% over all crits on your cast. I have a throbbing headache and sick so yell at me if i'm making no sense.

your crits and hits aren't making sense to me, shouldn't the hit for example be
(.83+ .01(hit rating/12.6))

or is that the same thing as you have?

going to bed, i'll look again later :S

This won't take into effect CSD. It'd be nice, but for my initial math runs I won't be doing anything for specific gear sets. Once the initial math is done, it should be relatively easy to adapt it to gear-related equations using the spreadsheet. As for Hit vs. Crit, I think we're saying the same thing, that with 100% Hit and 99% crit, every cast that hits, crits. And yes, that's how the representation works.

Thank you for catching me on the rating conversions, though. Forgot to reciprocate, going back and editing.

I'm hoping to get some help in calculating the relative worth of [Timbal's Focusing Crystal] to compare it with the [Icon of the Silver Crescent] and the [Sorcerer's Alchemist Stone].

I'm basing my calculations similar to what was shown in the comment section on wowhead.

For reference, the conclusion was this:

Essentially, this matched what I came up with using slightly different calculations, raid buffed the proc gives approximately 22 DPS unbuffed it gives approximately 16 DPS. (No CoS, Misery, ISB, etc.)

Now the question I have is what value should I be using to determine how much DPS I get from 1 spell dmg. This is how I attempted it. Using the boomkin spreadsheet I fully raid buffed then went to add 1 nature and 1 arcane damage to the quick add section. Fully buffed with an ideal rotation it gave me 0.6 DPS for 1 Spell Damage. Unbuffed with a different ideal rotation it gave me 0.5 DPS for 1 Spell Damage. I realize this depends on personal stat/crit/haste values for example. I'm personally at 1193 Spell dmg 21.7% Crit and about 39 haste rating.

So using that information in a raid scenario Timbal's proc is worth about 22DPS / 0.6 DPS/Spell Dmg = 36 Spell Dmg or if I use 0.5 DPS 44 Spell Dmg.

In a non raid scenario Timbal's is worth 16 DPS / 0.6 DPS/Spell Dmg = 26.7 Spell dmg or if I use 0.5 DPS 32 Spell Dmg.

So 26-32 Spell Dmg without a raid or 36-44 Spell dmg in a raid.

So total that's a conservative 44 Spell Dmg + 36 = 80 Spell Dmg in a raid and 44 + 26 = 70 Spell Dmg without raid (de)buffs.

Icon is slated to be on average ~67 Spell Dmg if you average out the on use effect. Whereas the Alchemist Stone is a constant 63 spell dmg.

What's not considered here is the fact that you can stack cooldowns with the Icon, ie. Icon + Heroism + Destro Pot + More haste for better results than just using the trinket when it's available. Also, that the Alchemist stone gives you 40% more mana. (For the cases when you do not have an spriest in group and might be experiencing troubles)

Another thing to consider is that on trash (5-10 man) you most likely will not be always applying dots as mobs die to quickly. So getting full use of the Timbal proc is unlikely. Another point is that not every boomkin uses IS in their rotation because it means less personal DPS depending on their current gear level. I personally do for the raid utility.

So, my personal conclusions (correct me please if you feel I miscalculated above) with the assumption that you're keeping both IS and MF up, and do not need the Mana gains from the alchemist stone are:

[Timbal's Focusing Crystal] > [Icon of the Silver Crescent] > [Sorcerer's Alchemist Stone].

I'll personally be replacing my Quag's Eye with the Sorcerer's Stone (will hit exalted tonight) until I can get the Timbal to drop. 0/9 at the moment.

I'm almost always Resto in raids, so these numbers reflect mostly BoJ gear or gear that other casters didn't need, but using the trinkets listed in the spreadsheet, here's how my DPS numbers were with different trinkets with the IS, MF, SFx3 rotation:

That would make the Timbal's trinket the 5th best for me at the moment, with small increases past that up to The Skull.

Are we sure things like the Timbal proc are modelled in the spreadsheet? I assumed it wasn't other than the passive +spell dmg portion.

No, but then neither are click effects. So the Skull, the Hex-head, the Icon, all of those would also migrate upwards based on their on-use.

You have to remember that the spreadsheet doesn't take procs into effect, which is a big boost for Icon, Timbal, Hex and Skull. Less of a boost for the Eye, and no boost for Audacity (i was initially surprised to see it that high in your list)

Actually, the development version of Rawr.Moonkin takes into account proc effects and click effects, and the Skull and the Eye are still the top two that I can see. Now, I should caution you that this is if you're not hit-capped; I don't know any hit-capped moonkin personally, so I don't know how the Eye stacks up. I also don't recall seeing Timbal's in there. The Icon and the Hex Shrunken Head are right behind the Eye and the Skull, though.

[edit] I did find a (nearly) hit-capped moonkin to test on, and it still shows the top 5 trinkets as:
1) Skull of Gul'dan
2) Eye of Magtheridon
3) Hex Shrunken Head
4) Battlemaster's Audacity
5) Timbal's Focusing Crystal* (proc has not been added for Timbal's yet, sorry guys)
6) Icon of the Silver Crescent
Now I know a lot of you don't hold Rawr.Moonkin in high regard, but I worked hard on it and I like to think that my math is mostly accurate, so take these results as you will.

The main problem i have with rawr.moonkin at the moment is that it doesn't filter out mage gear (i suppose that will be in the upcoming version, so whatever). But anyway, I wasn't saying anything against rawr, it will certainly be a very good tool (i didn't get to experiment much on it for the moment but did use the feral one).

However, i can't agree with ranking Battlemaster Audacity higher than Timbal or Icon dps wise. A statical 3 or 4 +dmg increase is certainly not more than either Timbal's proc (+10-20 dps, see Zene's post on this same page) or Icon's on-use (~+20 dmg) added damage. Alerian said he used the spreadsheet to compare, and the spreadsheet doesn't take procs into account, which explains why he got that ranking.
I merely posted to make him aware of this element, not to criticize anything.

That being said, it does surprise me that Rawr also ranks Battlemaster Audacity higher than the Icon. I don't see any hypothesis that could explain that appart from giving a value to life/stamina (which i usually consider to be null myself for dps comparisons)

just raided on saturday... managed a 10th with 17 dpsers. not using the spread sheet prescribed rotation but a situational one (IS IFF always up, Wrath when i have a shorter window to stop moving MF when i have a lengthier window. however i think dps would be target specific. without a working omen, i did not pull aggro at all without stopping dps.
Mags eye > Icon of the Silver Crescent imo because of raw damage increase, the proc did go off sometimes irl. i am at 150 hit rating and i am suprised at the frequency i can get resists. of course for theory crafting purposes, the reverse is true.
I have not stacked haste before and i am wondering from the posts above if crit>haste.
Can anyone list, point for point, the stats to stack? i thought it was hit>dam>crit>int>spi>mp5. now with haste it would seem hit>dam>crit>haste>int>spi>mp5...
Any confirmations on this?

I would say that the priority list isn't really possible to do with different scalings and gear options. Tentatively I'd say Hit>Damage>Haste=Crit>Int>Spi>mp5

A lot of that depends on what pieces are your sources for each of those though. Quag's eye can eat up a lot of need for haste with the 50sec. proc (I believe it has the usual 45sec. ICD). It depends situationally...when BL/Heroism and drums are up I'd take haste any day. With Quag's eye and those two buffs I average about a 1.8sec. SF at 30% buffed crit. Quag's eye has always proced. on BL for me short of maybe a few times so it's fairly reliable. Of course that may be why I see such a big jump on DPS from SF spam and iFF only. When I throw in MF and/or IS my DPS drops by about 150-200. So in my personal case I'd say haste>crit, but that just may be with how I gear.

It doesn't give any weight to stamina. I'm not sure why either, because my napkin math says that the Icon's on-use effect should be worth ~26 spell damage if used on every cooldown, which puts the icon at ~69 effective spell damage, versus the Battlemaster's static 47. Perhaps that should be a hint to me to recheck my math.

Wanted to let you guys know that I'll be applying to be an author on the TTT Balance druid article so I can place these in a more easily-accessible place. That said, I promised a small workup this weekend, and so here goes. Note that my initial assessment of the relative difficulty of these was a bit skewed. As it turns out, Haste is actually the easiest stat for which to find a closed-form relationship to spell damage.

The following assumes a fully-talented Balance druid, though the end result is very-easily adaptable to any Balance druid. In fact, it's applicable to most nukes in the game.

Given that the DPS of Starfire is as follows:

, where is the base damage after Moonfury, 658; is the damage coefficient after Moonfury and Wrath of Cenarius, 1.32; is your current +damage, including the bonuses from Int; is the Crit rating conversion coefficient, 1/2208 = .000453; is your current effective Crit rating, after decomposing percentage-based effects; is the Hit rating conversion coefficient, 1/1262 = .000792; is your current effective Hit rating, again after decomposition; is the Haste rating conversion coefficient, 1/1576 = .000635; is your current Haste rating.

Let's define the benefit of adding one rating to any stat.


Previously I've used a division. It ends up, when comparing two stats, that the two methods will give the same answer. Each can be useful, and in this case subtraction is more so. We're concerning ourselves with finding the relative value of two stats, one of which is Spell Damage in general, and the second of which, specifically, is Haste rating. This relationship would be given by


Now that we know how to find what we want, namely , let's find it. First, let's determine the benefit of adding one Spell Damage:








And now for Haste Rating:




Now that we know both, we can calculate the relative benefits of the two:






And that's it. Very simple. Given my current set of gear, with 1020 spell damage unbuffed, and no haste gear, is equal to approximately .9635, or that 1 haste is worth .9635 spell damage for me.

Implications:
Now that we have a formula, it's fairly easy to find break points. Let's find out the first point, in spell damage, where Haste begins to be useful, i.e. has a spell damage value of better than 1:




Now, remember, this only works for Starfire spam. When used in a rotation, this number will be higher, as Moonfire and IS do not gain as much benefit from Haste as Starfire does. That said, it can be generalized that until you have more than 1080 Spell damage, haste gear is a DPS downgrade.

I love you, now to address probably the most pertinent issue for raiding moonkins. Most moonkin are around 1200 spell damage. Given that is there a way to figure which is more important for dps: crit or haste?