I felt like an appropriate follow-up to my last article would be to look at the talent responsible for creating such a low soft cap on haste. For a more intimate look at the talent responsible for crit's soft cap, visit Graylo's guest post over at WoW.com

Nature's Grace. The butter to Eclipse's bread. The speed increase associated with this talent adds excitement to the Balance Druid's rotation, allowing a player to rain down massive Starfire bombs upon the heads of his enemies or bring out a Gatling gun of Wrath casts. There really is nothing as exciting as watching a Wrath a second fly out from your druid and wreak havoc. And there's nothing as frustrating as watching that Wrath cast in .6 seconds only to have to wait almost as long as the cast took before you can start firing another.

Unlike Eclipse, many druids do not see a problem with Nature's Grace. They enjoy the faster Starfire casts and adhere to a 400 haste rule so that they never really feel the effect of the downtime because they gimp their stats to stay at 400 haste. In my last post, I looked at the haste soft cap and (hopefully) encouraged people to continue gearing haste after 400, despite its lowered relative worth. In this post, I will look at how Nature's Grace is causing issues with the global cooldown and haste.

Power

Nature's Grace is an extremely powerful talent. Previously, Nature's Grace affected a single cast. Now, the buff persists for 3 seconds, allowing us to cast 3 Wraths or 2 Starfires. While this seems like a buff, and I posted some other peoples thoughts on the matter, I am beginning to wonder if it is not part of the fundamental problem with Moonkin DPS and scaling. As mentioned above, Moonkin currently gear to 400 haste as a priority. At this point, haste takes a dive in its relative value. This is due to clipping the GCD with Wrath while Nature's Grace is active.

I have seen a plethora of ridiculous suggestions lately that include, but are not limited to: Dropping points from Nature's Grace to lower uptime, dropping points from Starlight's Wrath, dropping points from Celestial Focus, or dropping points from Improved Moonkin. All of these suggestions are ludicrous. First, if you did drop the points, where would they go? Our two floater points tend to be placed in one of three places. Gale Winds, Brambles, and Owlkin Frenzy all provide sub optimal DPS boosts. Other options are even worse. Genesis is a complete waste, even with impMoonfire and 2T9. Typhoon is situational at best. Also, dropping a talent that provides stats (as Nature's Grace does) only leads to gearing difficulties or multiple respecs in order to maintain the "right" amount of haste. Don't go down this path. It's a disaster. You would be sacrificing faster Starfire casts for Wrath, a poor trade-off. Another way to think of this situation is to look only at Wrath. Where previously, 400 haste brought Wrath to a GCD cast with Nature's Grace, you would now need more haste (and therefore more itemization points in haste and not crit or spellpower) to reach that same value. At the same time, your Starfire would suffer the same increase in cast time, essentially managing to boost the relative value of haste by cutting out haste boosts from your spec. That's like saying "I really value my left hand after cutting off my right!"

Scaling

Nature's Grace increases your casting speed by 20%. With a long nuke such as Starfire, this bonus is great. Channeled spells also reap great benefit from haste boosts such as this. The only pitfall with such a buff occurs when the spell being cast has a short cast time. There is a minimum of a one second global cooldown that cannot be reduced. Cast times can be reduced past 1 second. So what happens with the extra time between the end of the cast time and the expiration of the GCD? Exactly nothing. This presents a scaling problem similar to the one laid out by Graylo in his article. To quote:

The size of Eclipse not only amplifies other problems but it actually causes other problems as well. The prime example is how moonkin scale with crit rating. A lot of people assume that Eclipse increases the marginal value of crit rating for moonkin because Eclipse procs off of critical strikes. In reality Eclipse hurts the marginal value of crit rating relative to other stats more then it helps. This is due to diminishing returns felt as critical strike chance increases.

To put it in less mathy terms, increasing your critical strike chance increases the DPS of only part of your rotation. If you have a 50% crit chance and cast 100 spells, the 50 spells that would have crit anyway won't crit harder because your crit chance is increased to 51%. So, as you increase your critical strike chance, there are fewer and fewer spells that can be affected by increasing your crit rating. This decreases the marginal value of crit rating. When you have a buff like Lunar Eclipse that increases your crit chance by 40%, the marginal value of crit rating goes down a lot. In fact, the Lunar Eclipse buff is so large, that moonkin are able to cap their critical strike chance with normal ICC 25man gear, thus reducing the marginal value of crit rating to zero for Starfire while Lunar Eclipse is active. This is important, because it means other classes will improve with better gear more quickly then moonkin will.

Likewise, Nature's Grace actually hurts the marginal value of haste, although not in exactly the same way. Given the bottom line of the GCD, any haste that pushes a spell below the GCD no longer benefits that spell. My last article showed that this value with Nature's Grace is 400 haste and without Nature's Grace is around 1100 haste. Putting the Eclipse argument outlined above together with the Nature's Grace argument presented here implies that the only stat that Moonkin's scale fully with is spell power. Both haste and crit suffer from some type of soft cap.

I pulled this slightly out of context, but it came from Murmurs quite awhile ago. I think it does a good job explaining the interaction of haste, crit, and Nature's Grace. Murmur's is referring to Wrath casts here.

It's odd because haste and Nature's Grace have something of an inverse scaling with each other. In theory, the more crit you have, the less value the Haste from your gear is going to hold because you'll see less time benefiting from it, conversely the more Haste you have, the lower the impact Nature's Grace has on your overall casting time. If you already have 50% Haste, then Nature's Grace is worthless. If you have 100% crit, then Haste is worthless.

Putting the Puzzle Together

Nature's Grace is the cause of the very low haste soft cap. For progression raiders in ICC gear, running with less than 600 haste takes some tricky gearing. As such, we have begun to see a scaling problem with haste on our gear. Recently, we received a buff to Earth & Moon to help us keep pace with other DPS classes, but this is clearly a band-aid.

Likewise, as our crit rating increases, our Nature's Grace uptime increases, decreasing the marginal value of haste further. Additionally, as uptime increases, this talent becomes a larger contributer to our DPS. Changing the talent has the potential to drastically alter Boomkin DPS, so it is understandable that Blizzard is waiting to rework the talent (if they intend to do so).

Looking Forward (TL:DR)

It's probably foolhardy for me to lay out some ideas here, since they will most likely never reach the right eyes, but I wanted to consider some alternatives to what I consider a stat boosting spell. I also understand that mechanics that increase damage done by a flat modifier are pretty lame. It is my hope that if/when a Nature's Grace redesign occurs, it will not be converted to such a talent, as I feel this would be an injustice to our game play and remove the cross-spec utility of the spell. A good talent will retain its usefulness to both specs (resto/balance) and contribute to a dynamic play-style.

I tossed around several ideas that involved charges or complicated mechanics. I wanted to make sure that the talent still maintained a reliance on multiple stats (and yes, this will probably all be moot when Cataclysm hits). I finally settled on a spellpower boost that comes from your haste rating. Without a 20% haste buff, the soft cap rises to ~1100. Given this new (and much higher haste cap) along with the increase of haste priority in gearing, we would begin to see much larger haste values (~800). If all of these conjectures were to become true, then this is what I would propose.


I particularly like the fact that this talent would then scale with gear while not creating any specific soft caps. I did not try to balance the talent with anything more than the briefest of napkin maths, so the 25/50/75% business might be a little funky and need some fine tuning. It has the interesting side effect of increasing the power of the Lunar Eclipse (which as I understand it, is not scaling as well as the Solar Eclipse).

All suggestions aside, I hope we can agree that while Eclipse may be the most obvious focal point for Moonkin rage due to its seemingly RNG nature, Nature's Grace presents a problem that is as fundamentally challenging for the Moonkin rotation. Here's to hoping that Cataclysm holds some changes for us in this department!

It seems like it has been awhile since I've talked about theorycrafting on this blog, so let's begin a discussion on raiding soft caps. I chose this topic because I've seen a rampant number of posts over on TMR that just reek of misunderstanding. Posts along the lines of "Oh noes! My haste is 403 and I have this uber wep but it will put my haste to 480! Should I shard it!?"

Yes, Virginia, there are all sorts of folks misunderstanding what a soft cap means. So let's just takes a second to break down the difference between a soft cap and a hard cap.

Hard Cap

Hard caps represent a value of a stat after which increasing that stat yields zero gain. A perfect example is hit. For a Horde Druid (I always see people talking about Alliance druids...) the hit cap is 10% or 263 hit rating. At this point, you will not miss a spell cast against a boss level mob. The druid with 263 hit rating is going to garner exactly the same benefit as the druid with 363 hit rating. That extra 100 hit rating does absolutely nothing for the character. This effect is known as a hard cap. Other examples of hard caps come with expertise, armor penetration, and critical rating, among others.

Soft Cap

Soft caps represent a value of a stat after which the value of that stat relative to other beneficial stats changes significantly. Yes, this is a mouthful. Haste is an ideal example here. While casting Wrath, we often run into the problem of "clipping the GCD" while casting. This means that our Wrath casts in less time than the GCD takes (this is set at 1 second). When Nature's Grace is up, this happens at 400 haste. Any haste after 400 yields no benefit for a Wrath cast that is affected by Nature's Grace. However, any Wrath that is not affected by Nature's Grace, as well as all of your Starfires still scale with haste. So any value after 400 still has some benefit, albeit less than before 400 haste. This sudden change in value is called a soft cap. Other examples of soft caps come with expertise, armor penetration, and critical rating. Yes, I used the same examples and yes I did that on purpose.

Balance Druid Soft Caps

---

So now we will be discussing the Balance Druid specific soft caps and the relative loss of stat value that follows the soft cap. Hopefully this will demystify why any stat past its soft cap can still benefit you.

Haste

Since our cast time is bounded on the bottom by the one second hard GCD, we will look first at a haste cap that brings any Wrath affected by Nature's Grace down to one second.

So we need to gain 50% haste or 1640 haste rating (since we can't get half a rating) in order to bring our Wrath cast time down to one second. Now lets examine the help we get from our buffs and our raid (keep in mind that while haste rating is additive, haste percentage modifiers are multiplicative).


To predict Nature's Grace uptime we will use (1-(1- Critical Strike Chance)^Number of Casts) where 3 will be the number of casts that we will use, since that is the number of casts that can be started within one Nature's Grace proc. For simplicity, let's use a theoretical druid with 50% crit raid buffed. Then my uptime for Wrath will be:

Remember that we are only looking at the Wrath portion of our rotation, so I am able to say that 12.5% of my Wrath casts will be unaffected by Nature's Grace. Likewise, I am able to say that any haste after 400 is 12.5% as effective as any point of haste rating before 400 for Wrath.

So how much haste does it take to make Wrath a 1 second cast without Nature's Grace. Let's call this number the "less-soft-cap". Bear in mind as we calculate this cap that it is much less of a change than the previous soft cap, representing a very small change in the value of haste.

As we can see, this is a large amount of haste rating. This is generally considered an unmaintainable number under normal circumstances via gear. Yes, you probably could reach this value with procs, but that only represents a very small segment of time. Based on these observations, we currently ignore the less-soft-cap, however, we could say something along the lines of "Between 400 and 1137 haste, haste rating is roughly worth ((1 - Nature's Grace Uptime) * 100) percent of its sub 400 value for Wrath. Naturally, these values fluctuate from gear level to gear level, but they should maintain relatively constant. Also remember that these values do not include the large gains that Starfire still garners from haste.

Crit

It is impossible to crit more than 100% of the time. This creates a hard cap for critical strike rating. However, the unique mechanic of Eclipse, coupled with the high Eclipse uptime, yields a soft cap for crit rating. Lunar Eclipse is present often enough to consider it when considering a soft cap. It does not approach the Nature's Grace uptime, but it is still fairly high.

Crit percent is much more difficult to calculate because it relies on two variables: crit rating and intellect. In order to make our lives easier, let's assume that the druid in question has 1500 intellect. We will use the following formula to calculate our crit rating needed to hit the crit soft cap:

This translates to 43% on your character sheet with all raid buffs. Any crit past this percentage will be "ignored" during a Lunar Eclipse. However, we still fire Starfires outside of a Lunar Eclipse. We also fire every single Wrath outside of a Lunar Eclipse. Essentially, over 50% of our rotation is outside of a Lunar Eclipse, meaning that the additional crit is still contributing over 50% of the time.

TL:DR

While soft caps indicate a change in the value of a stat, they do not mean that the stat has lost value. Roughly, we could estimate the value of each stat at 50% of its pre-cap value (this is an underestimation). Even after this reduction, both haste and crit are better than spirit. So don't avoid haste, or crit, after the soft caps. Don't freak out if you go over 400 haste. In ICC it will happen. Just keep going for the sp/crit/haste pieces anywhere you're not putting a hit piece.

If you could change one talent in any of the druid talent trees, what would it be? You could talk about what you want it changed to, or just talk about what makes it a weak talent that should be either changed or replaced in some way.^[1]

This question grabbed my attention and I think it has some merit as a Shared Topic. I noticed Graylo made a post on the idea as well. And curse him for having the time to push this out before mine because I will be echoing his sentiments on many things. Oh well, imitation is the best form of flattery, or so they say.

Improved Moonfire

Improved Moonfire as a talent has been lost for awhile. So much so, that even as a second tier talent, it is often overlooked. It took one of the better set bonuses that we have seen to convince us to even consider the talent. This indicates that the talent needs to be reworked.

So what would I do to it? Scrap the increased critical percentage. As a modifier in PvP it does have some usefulness, but I think even using MF spam to finish off an opponent has lost most of its viability. Lets give this talent the 2T9 bonus. "Moonfire DoTs have a critical strike chance equal to 50/100% of your critical strike percentage. Also increases damage dealt by Moonfire by 5/10%."

I think critting DoTs have been on their way for awhile and this set bonus was a "test run" to see what it would do to our DPS. I also think it's been a good change and one worth keeping in the class for awhile. I considered extending crits to Insect Swarm, but decided against it based on two facts. First, IS is a talent. I really dislike talents that buff talents because I feel like it becomes mandatory to pick up the talent as gameplay was probably balanced around having both. I believe that critting IS would require a reduction in some other portion of the Moonkin arsenal and I'm not interested in walking that path. Second, IS is already glyphed for damage. I believe critting IS would require a re-evaluation of the glyph, reducing the damage increase (circa 20%).

Genesis

Raise your hand if you hate Genesis. From a Moonkin point of view, it simply does not deliver. Now yes, I have read the posts about how it's really for Resto and blah blah blah. Here's my problem: There are so few viable options in the second tier that we are forced into the least unattractive option. Nature's Majesty seems a pretty clear must bu what of the other three points? Clearly, Moonglow fills this out nicely, but what if you aren't experiencing mana problems? Where else can you put those points? In order to reach the third tier, you would need to put at least 1 point in Genesis if you didn't want Moonglow. I have Moonglow in my build simply because I wanted to make sure I didn't run out of mana. But in a 25 man, I'm sitting pretty with my mana pool and replenishment.

Fixing Genesis is a whole different beast. I believe it should maintain its viability for Resto Druids so that they can progress down the tree with some type of benefit. I also believe that the talent should offer some viable benefit to Boomkins. Since PvP oriented talents shouldn't really be prevalent until the upper tiers, I suggest a talent that maintains its healing benefits to DoT's and increases the damage and cast time on Wrath.

Wait . . . wha?!? A talent to make a spell . . . slower? Yep. I hate clipping the GCD. I hate having the value of haste reduced because it stops affecting a large percentage of my rotation. I would much rather have the spell seem more like Starfire (since they are fundamentally the same) and get the full benefit of my stats versus firing snotballs every second with no spell queuing and often clipping the GCD.

Resto Tree

While I agree that there are problems in the Resto tree as far as progression down the tree goes, I can't say that I am overly interested in posting fixes to it because I feel like a Resto redesign will mess up what I would say anyway. If you want to see some good ideas on the Resto tree, visit Gray Matter

Granted, once 3.2 is released, this becomes a bit of a non-factor but I wanted to address the "soft-cap" of 400 haste for rotations. I will open by making the statement that for a Lunar Eclipse, the 400 haste cap is asinine. That's right. It's complete fodder. By definition, the Lunar Eclipse has a 60% chance to proc on a Wrath crit, and while I realize sometimes we have a dry run with many Wrath crits and no Eclipse proc, that is often only our memory latching to the bad instead of the average.

Ok, so why is the haste cap ridiculous for a Lunar cycle? These events are independent (i.e. one Wrath crit does not impact when the next Wrath will crit). Instead of looking at Eclipse as a 60% chance to proc on crit, let's look at it as a 40% chance to not proc. Let us then look at the successive chances that Wrath will not proc Eclipse.
As we can see, having 5 independent crits with no proc is a 1% chance. Now, what does all of this have to do with the 400 haste soft cap? Well, the haste soft cap includes the Nature's Grace haste bonus. If you're not under the influence of Nature's Grace, you can afford more haste.

The current logic states that any rotation that contains Wrath should have a soft cap of 400 haste and I'm afraid I've just seen too many erroneous posts asserting the anything over 400 haste is a waste. It's not. Any rotation that contains many Wrath casts, including crits, with the intention of continually casting Wrath after the crits (read Solar cycle) should aim for a haste rating of around 400. Any rotation that plans to switch spells after a Wrath crit is more than okay with a haste value over 400.

I've been puzzling over how to represent this mathematically since it's very tough to model the probability of some chance encounter when I more or less know the previous events. The problem we encounter is that probabilities are not static once the sequence is in motion. Once I've gotten my first crit and the Eclipse did not proc, I reset the entire sequence. Due to the fluid nature of this problem, I have decided to make a table of likely sequences and attribute each sequence a percentage. That way I could see just how likely I was to be casting under the influence of Nature's Grace. I assumed a worst case scenario (which is that all the crits have 3 seconds of Nature's Grace associated with it). Below are the weighted values of the Nature's Grace time through n = 5.
As I sum this, I find that I get a value of 3.198 seconds of average uptime on Nature's Grace given my worst case scenario of non-overlapping crits. I would be working with for n=1 through infinity. I treated this like a Power series and came to the conclusion that it summed to 3.333 repeating, of course. Now, I did this over the course of several 5 minute spurts throughout a few weeks, so I could have fractured logic (in which case, tell me!!) but I'm forced to come to one conclusion here.

TL:DR

The results seem to indicate that I have an average of 3.3333 seconds of uptime on my worst case scenario in a Lunar precasting phase. That means that for any haste over 400, we clip the GCD for an average of 3.3 seconds. I really don't feel like this is such a bad thing because I showed earlier that the average Lunar cycle takes around 35 seconds. The rest of the rotation (32 seconds) is spent without clipping the GCD. That's 91.4% of the rotation that benefits from haste over 400.

The haste soft cap is exactly that: Soft. Especially for a basic Lunar Rotation. Naturally, math gets more complicated when Starfall is accounted for, although then we're talking 67% of 91.4% or 61.2% of the rotation. I say this because Lunar cycles are using Starfall at most every third Eclipse. If you plan on running a Lunar Eclipse (until 3.2), haste is still good for you. Don't be afraid to go over 400 haste!

Debates are raging! Procs are being watched! Combat logs are being combed through! Cowpies are being hurled with precarious disregard! And all the while we're wondering what the final judgment will be.

Here's what we know: It's a 15% chance to proc a GCD Starfire every time that Insect Swarm ticks. There is no internal cooldown.

I will be using a slightly uber druid for stats because with 4T8, I'm assuming you've garnered some good gear. Therefore, my druid has 3k spell power, 45% crit, and 389 haste rating. As always, there is a TL:DR for those of you looking for the "express" ride.

Math Stuff

Chances to proc: Insect swarm ticks 7 times over 14 seconds. So I have
Thats 4.5 procs per minute. To even out the numbers, I'm going to be looking at 120 seconds (two minutes) so that I would get 9 procs.

Nature's Grace Uptime: I will be using Starfire's NG uptime for this so
GCD:
Average Starfire Cast Time:
Average Starfire Damage (assuming SF idol and moonfire buff to crit):
I would get 9 instant Starfires @ an average GCD of 1.065 seconds for a total of 9.585 seconds of casting time. In this time I would normally do an average of 4.5 Starfires.

TL:DR

My theoretical druid picked up 328.78 DPS through his 4T8 piece bonus assuming that the buff never overwrote itself and wasn't consumed by the casting bug. That, to me, is a significant increase in DPS and warrants the inclusion of the 4T8 bonus into the BiS gear.

A small note on all of this: The benefit of the proc decreases as Haste increases. Logically, this makes sense because there is less gain in casting time per proc as you cast faster. In Bloodlust and Heroism situations my druid is popping off Starfires at a rapid 1.434 seconds per cast. Here, the DPS boost of the proc is minimal. I speculate that Solar rotations will make heavier use of the 4T8 than Lunar rotations using high haste. For example, a druid with 700 haste will find himself gaining around a mere .9 seconds of cast time through the proc. My druid was gaining well over 1 second per proc.

A big thanks goes to Murmurs for the Lunar cycle numbers as well as a plethora of formulas so that we could have a comparative Solar cycle. I have incorporated these values into a page of the spreadsheet based on your current gear and your BiS list.

Lunar Rotation Scaling

Percent of Damage: Insect Swarm: 12% Moonfire: 10% Starfire: 50% Wrath: 25% Starfall, Force of Nature, proc effects: 3% (ignored due to complications)

Spell Power - (0.14957948 * .12) + (.10524679 * .10) + (1.1023188 * .50) + (1.1865983 * .25) = 0.876283192

Haste - (0.0113592254 * .12) + (0.00841851174 * .10) + (1.05240836 * .50) + (0.411537562 * .25) = 0.631293529

Crit - (0 * .12) + (0 * .10) + (0.878345459 * .50) + (1.05686234 * .25) = 0.703388314

Spirit - 0.876283192 * (15/100) = 0.131442479

Intellect - (0.275459945 * 0.703388314) + (0.876283192 * (12/100) = 0.298909289

Okay, so, I know what people are thinking "What!? Crit is so not better than Haste for a Lunar Eclipse rotation! You've done something wrong!" Well, I may have done something wrong, but I don't really think I have, and, if anything, I favored Haste more than Crit in my calculations. The out come isn't all that odd though, and does make some sense. Haste is far better than Crit is when casting Starfire, there is no doubting that, however Haste is essentially worthless for Wrath at this point. However, and here's the fun part, the more Crit you get, the lower a percentage Wrath is going to take up in your rotation, thus increasing the portion of Starfire and netting more gain from Haste. I looked at a wide variety of DPS reports in order to construct these percentages, and I'm quite confident that they are accurate. As I recall, the highest Starfire percentage I saw was from a Druid on Hodir where he had 60% Starfire, and 27% Wrath - although he didn't use Starfall nor Force of Nature (which I thought was odd.)

Do not read this as me saying, or even the data suggesting, that Druids using a Lunar Rotation go out and stack crit and toss out all of their Haste. That isn't the case. Druid should always stack to 400 Haste before anything no matter what rotation they are going to use (and these numbers assume that you already have Wrath Haste capped) - at that point you probably want to focus on getting Crit. Trinkets that have an on-use Haste effect (Haste procs too, although be warry of those as a proc during Wrath spam is sadness and you'd get more out of Crit or Spell Power procs then) and Potions of Speed are still your friend. As is Power Infusion.

Solar Rotation Scaling

Percent of Damage:
Insect Swarm: 11%
Moonfire: 7%
Starfire: 32%
Wrath: 46%
Starfall, Force of Nature, proc effects: 3% (ignored due to complications)

Over 400 Haste

Spell Power: 0.877493361
Haste: 0.433181487
Crit: 0.506081945
Spirit: 0.263248008
Int: 0.244704508

Under 400 Haste

Spell Power: 0.87437463
Haste: 0.737575344
Crit: 0.552895115
Spirit: 0.262312392
Int: 0.257225415

Ok, so my synopsis right now is that this wont hold much weight. First, I used Mumurs' theoretical druid to calculate this so that the stats would be relative to his stats. 30% crit, 15% haste, 2700 spellpower. Reading this, we see that under the haste cap, we still benefit most from haste. Once we hit the soft haste cap, crit takes over. Spell power continues to reign supreme (and did so through some very large numbers of haste/crit/SP).

When I ran the calculations for my BiS geared druid, I got some different weights but the rankings stayed mostly constant. I put together one of the highest SP sets I could as well and found that in these instances, crit and haste began to overtake spell power. That said, the high spell power set performed poorly within my DPS model and I would highly recommend against seeking such a thing. Spell power is built in to all of our items with the possible exception of trinkets. You shouldn't actively seek the highest spell power piece, but instead search for the best "secondary" stats on the item. In many cases, the highest spell power piece was NOT the best performing piece given either my weights, Graylo's weights, or Murmurs weights. I'd like to give a nod to Kiliani too, for putting together some very nice dps gear sets. Unfortunately, I could not run his weights because I did not have them.

TL:DR

1. Hit till capped.
2. Spellpower is greater than all =]
3. Haste till ~400 then Crit
4. Int~Spi
5. Going too high on haste with a solar rotation causes clipping during the wrath portion of your rotation. This is akin to lost dps.

Nature's Grace plays a very important role in the solar rotation (and in the lunar rotation). Among other things, it creates a soft cap of haste rating for druids beyond which haste becomes less effective on a per point basis. Keep in mind this is a soft cap, not at all like the cap on hit, after which a player receives zero benefit from additional points.

My motivation here is that I am trying to build a spreadsheet to calculate my theoretical DPS. One of the most important components in this calculation is cast time. The haste part was easy, but after hours of scouring the internet, I was unable to come up with a reasonable formula for finding the uptime of Nature's Grace. (apparently I didn't look very hard) Woe is the balance druid's underpopulation. And so it was that I began re-inventing the wheel to make my own spreadsheet. There is was a lot of math here. If you're not interested in that, hit ctrl+f and search for TL:DR. Your brain (and eyes) might thank you.

Allow me to steal a page from Graylo and talk about my theoretical raid buffed druid, casting with 48.8% crit, 3061 spell power, and 22.2% haste. I am ignoring on use trinkets and eclipse bonuses. I expect the mob to have moonfire on it while casting starfire.

Allow me to quote Murmurs:

Nature's Grace: Buff or Nerf or what?

This one is pretty much a no-brainer. For Balance PvE DPS, this is a huge buff, probably one of the largest buffs we've seen since they re-adjusted the base-line damage of Starfire and Wrath to non-suck levels.

Here's how the new Nature's Grace formula works for calculating cast time:

((Base Cast Time / (Haste Effects)) / ((1 + (Haste / 100))) / (1.2)

Average cast time formula should look like this: ((Base Cast Time / (Haste Effects)) / ((1 + (Haste / 100))) / ( 1 + (.2 ((1-(1- Critical Strike Chance)^Number of Casts)))

Using the numbers above, Starfire's average cast time should look something like this:

((3 / 1.114) / ((1 + (18.15 / 100))) / (1 + (.2 * (1-(1-.3749)^2)) = 2.03

Under the current model Starfire's average base cast time should be this (note, I'm auto deducting the effects of Celestial Focus, Improved Moonkin Aura, and Wrath of Air Totem from the base cast for simplicity)

(2.7 - (.5 * .4049)) / (1.1815) = 2.11

While the proc rate of Nature's Grace hasn't changed, it still has the same effect on Starfire's cast time and now efffects 2 spells instead of only one which is why the average cast time is going to be lower than the current version. The only time there would be no net change over the Live version and new version, for Starfire, is during 100% critical strike situations - so basically Loatheb.

And now allow me to sum that up. See that little part that I put in bold? That's the part that tells us how to calculate the uptime. Think of it like this: We assume that Natures Grace has just procced. Then the chance of that buff staying up is equal to the chance of one of the three wraths critting. This is the uptime. So in go my numbers and out comes my uptime! Here is the theoretical chance that one out of n casts is a critical.

n Casts	Wrath	Starfire
1	48.82	51.82
2	73.80	76.78
3	86.59	88.81
4	93.14	94.61
5	96.49	97.40

My theoretical druid has an 86.59% uptime on Wrath and a 76.78% uptime on Starfire. Notice that this coincides with the percentage chance of at least one crit per the number of casts in one Nature's Grace proc. Back to the spreadsheet to continue plugging in numbers.

TL:DR
Calculate your Nature's Grace uptime by using (1-(1- Critical Strike Chance)^Number of Casts) for each spell. Remember, Starfire gains crit from both Moonfire and a Lunar Eclipse. Take that into consideration as you are calculating.