I'd like to hear more examples of what casting rotations people use and why.
One simple trick I use is if I can anticipate I am about to get a pushback from melee or magic damage, I work in a moonfire or insect swarm so it will occur during the 1sec GCD so there is no delay in casting.

What rotation strategies do people use against certain mobs or to improve mana or dps efficiency?

First off great job on this spreadsheet. Works very well however I have found a few isues:

1.) The time till oom seems low, but I can't figure out what exactly causes it
2.) Nether of my trinkets are avalible, (Neltharion's Tear and Shard of the Scale) I realize these are old trinkets being from nef and ony, but I have yet to find anything better even since I started raiding at 70. (I have added them to my copy but I think they are worth adding to yours)
3.) If you enter a fight duration lower than the Mana Regeneration Calculation delay then all the calculations are n/a even if the use is set to no for that one, (should ignore it if the fight is shorter than delay no matter what the yes/no is)
4.) With a latency of 0.4 the sheet says I should do MF/wrath and with 0.2 it says MF/IS/wrath mabey I'm missing somthing but wouldn't more lag benifit from dots rather than something you have to keep re-casting?

edit:
5.) Would it be posible to add a new field, time till oom, (ignoring the Expected fight Duration just using when your out of mana the first time, might be longer than the expected fight duration, might be much shorter. Currently the time till oom counts regen after you initialy go out of mana before the expected fight is over)

Time until oom will always seem low, because when theorycrafting you often ignore the huge amount of time in combat you spend not attacking. This has a huge significance on time until oom in reality, but in theory it doesn't matter because that's time you're not DPSing anyway. It only really matters in terms of mana regen. Overall it's not something you can't include, it's just very hard to figure out the % time while not casting without extreme analysis of said bossfight. The more regen you have (from raid buffs etc) the more significant this factor is.

I found the times pretty believable when I input my gear & buffs. Especially for, say, mostly unbuffed, chain-casting IS,MF,Wrath. Would regularly run me OOM in ~2 minutes.
I've been neglecting pre-BC gear because there are going to be relatively few Moonkin with those items. They were, however, part of my motivation for adding the custom-gear section, and providing instructions on how to add items on your own copies.

Fixed, for future versions. Just had to use some "if" statements to keep Excel from trying to divide by 0. A convenient work around is to put all the delays to 0.

NG is applied against the laggy cast time. So, instead of 1.7 non-NG, 1.5 NG, you're getting 1.9 non-NG & 1.5 NG. The NG procs do nothing for the DOT cast times. Also, what degree of variation are you seeing between the two cycles? In RL, the impact of crits & finite sample size are likely to have an observable variance that's often within the same range as the top 2-3 suggested cycles.

This would be a self-referential loop. It would check to see how long you could cast to know how much mana you'd regenned, then estimate how much longer you could cast with that mana, then check to see how much more you'd regen, then check to see how much longer... It can be done, but it's annoying. I don't really know what the utility of that information is.

EDIT: Err, and I know there is probably a closed form solution to the iterative loop. But the sheet & my brain aren't designed towards finding/using it.

I had the same problem when I first tried to build my own spreadsheet, (VERY limited compared to yours) I got around it by using a second sheet with 1000 iterations of the loop, not perfect, but it worked ok. The reason I like this number, (or more accurately the total damage I cause in a fight this duration) is I find it's a good guide to tell if a new piece of gear is better than my current in all situations whether the fight is short or long this figure sort of averages it out so I can say x mp5 is worth y spell damage.

oh one other minor issue, fight time is in minutes but time till oom is in seconds, I can convert it fairly easily but I think it would make more sence to be in the same units.

Thanks. I'm glad people are getting use out of this! With my new versions I'm seeing 300 downloads within 2 weeks, so it seems worth the few hours per week I'm spending on it!

The two main factors on W vs. S is CoS & NG. For those with "normal" lag, Starfire seems to reign supreme. There could also be some "cycle rounding" effects of W vs. S and how closely the predicted average cast times, including lag & NG procs, coincide with a 12s cycle.

Since I didn't really develop this spreadsheet until after I had stopped raiding, I'm glad to hear someone has confirmed it's real-world utility. As far as disparities you see in raids:
1) Debuffs - in addition to it being absolutely imperative that CoS doesn't drop, how does the uptime of JoW and/or SS relate to what you've input in the spreadsheet? Further, are you responsible for casting FF? And, perhaps most importantly, is your shadow priest--if you have one--applying VT with the "highest priority" in their spell queue?
2) Buffs - it's entirely possible you might sometimes get the non-talented DS even though you have a DS priest in the raid. Same with say BoW, impMotW, and/or forgetting to eat your +dmg food.
3) Mana regen - the spreadsheet makes the simplifying assumption that NO MANA REGEN EVER GOES TO WASTE. Of course, in practice this can be quite difficult to accomplish.

Suggestion: if you have a PC that can manage it, try recording your fights. You might be surprised at the areas you see where you can improve when you get to watch through with no pressure to perform

WWS seems to be an excellent tool; if you don't have the capacity to record the fight, I'd say it's possibly the next best thing.

As far as hampering raids, I understand your concern. I felt I had to quit when I was regularly falling 7th - 10th on damage meters during boss kills, because even though I was beating some "pure DPS" classes, I had little room to improve and lacked the time necessary to invest in continued consumable usage at the rate I had been. And I would be willing to bet your guild would rather have you switch to feral than resto, unless 2.1 has done that much for Lifebloom.

I guess my concern is really much more one of advancement. Right now, looking at my numbers compared to other Gruul/Mag level guilds my 800 or so dps on a tank and spank seems solid and comparable to mages/locks/shadow priests at that level. However looking at other guilds WWS reports I pretty much never see a moonkin in a similiar position, especially at a higher level. Is the conventional wisdom about moonkins right, and I'm just an anomaly trying way too hard, or is there real potential there.

I would love to see a moonkin from a higher end guild post a WWS log or similiar about what sort of real raid damage they are doing in late SSC or TK.

I might be able to provide some data for it soon. The main problem is that while offtanks are not that easy to come by, ranged damage dealers are just piling up everywhere. Thats a big point in "letting" someone specc moonkin for raids in a working raiding guild setup.

The next thing is equipment. Most druids come from a resto point in raids, then went to the feral part because of its new viability. Providing full gear and enchants plus pots is a big money sink with only boe epics and even as an enchanter like me for a third specc.
The next thing is what raiding gear you get. Taking away epic rings/necks/cloth items from pure raid dmgcaster is frowned upon to say the least, leaving you with an awesome 1 epic bop leather spelldmg item far into BT and the set tokens you wont be able to take for reasonable prices for an offspecc until you are at least 1 full tier ahead raidprogression wise.

Meaning you will try to compete in a t5 environment in t4 gear or t6 in t5. The most reasonable choice for me at least will be to take the surplus t4+5 which are just there due to random amount of one token dropping very often, using the boe epic set ring and neck and picking some cheap items like the magtheridon spelldmg ring plus some kara gear and to replace the tokens i simply cant get with gladiator gear from surplus points.

All alot of effort for maybe a simple "no, cant compete" in the end. Ill do it anyway, im sure of it

I think my situation is a little abnormal in that my guild has always had a billion druids (we once had literally a third of the raid made up of druids, I love my class, it worked surprisingly well). I've gotten to be "the moonkin" for a long time now and my guild has been really cool about upgrades, etc. I'm pretty solidly established as a primary dps caster, and have equal priority with mages/locks/priests etc. So I'm not worried about falling behind in gear.

My worry is more one of taking that gear, but possibly fallling behind or that it would be better used by others. I love my spec, I love that I can be dps in one fight and change gear and be a healer in the next. I guess I always just feel a little guilty whenever we are short a healer for a fight (even if there is already 2-3 resto druids).

I just feel like I'm playing a completely different class from the one that I see on forums, and I'm worried I'm missing some big liability. That's why I think this spreadsheet is an amazing product in that it's been confirming the numbers I've been seeing in game. It's extremely nice to get some hard theorycrafting to counter when the dumb mage goes "lol oomkin".

Who knows, maybe in a few months conventional wisdom will be bitching about overpowered moonkin. It took a long time for the community opinion on shadow priests to come around, I just hope to see something similiar.

The problem is not the raid gear. The raid gear is terrible for caster DPS - and even moreso the moonkin/shaman/shadow priest DPS set. It's all about spellfire/spellstrike, which druids avoid for some reason. If you use a lot of nature I'd understand not taking spellfire but not taking spellstrike is pure DPS gimpage. And you have to settle for cloth if you want your DPS up, there simply isn't enough leather caster DPS gear out there, and what is out there has too low DPS on it.

You are correct Efejel, acquiring the the spellstrike legs is not required for adequate dps. There are many viable options outside of the craftable legs.