We've already had the discussion for this trinket, I questioned nakedduck's calculations myself, and I was shown to be wrong. Reread his posts, his calculations are accurate, match up with kenpachi's results, and can be used to estimate the benefits for your own equipment.

You do not subtract the time spent casting from the beginning of the proc, you only subtract the wasted time at the end. The example with the 2.5 second trinket was just an attempt to show you not to subtract the time spent casting from the beginning, but I guess it wasn't clear enough.

If we take an example of 4 starfires, assuming the starfires take 2.5 seconds to cast:
(Each character is 0.5 seconds, and the X shows where the proc takes effect)

4 starfires is a total of 10 seconds.
Out of these 4 starfires, 3 of them benefit. So the proc has an effective uptime for 7.5 seconds. So the effective proc was up 75% of the time.
If we subtract the wasted uptime(1 X) from 8 seconds, we also get 7.5 seconds. Again, the effective proc is up 75% of the time.
The method you suggest would be 8 seconds, - 0.5 from the end, and then another -2.5 from the beginning. This gives us an effective uptime of 5 seconds. This is an effective uptime of 50%.

But you can clearly see that 3/4 starfires gain benefit. This is 75% uptime.

Time spent casting a starfire that gains benefit is not wasted, and thus you do not need to subtract the first cast time off the beginning of the starfire. Time spent casting a starfire that does not gain benefit is wasted though, so you definitely should subtract any time wasted from the end.

What are the boomkin minimum stats, what should be my spirit and mana regeneration? Things of that nature.

Refer to announcement up top "teach me how to play."

Spi, are you trying for a specific haste threshold, or just stacking it? I need to change the quick lionseye in my helm and gloves to reckless pyrestone, to get 1197 +dmg, 101 hit, 16% crit and 319 haste while still maintaining my t6 4 piece (wtb vanq dropping in sunwell, gd prot). I figure 319 puts me at starfires under 2.5 seconds, and with drums Im at 399 haste =P.

Alrighty, I'm back from vacation with a gift: I've finished up the calcs for both Hit and Crit, just need to format them for the boards and the article. I'll post them here first so I can get confirmation that I didn't completely hose the math on this one, and then I'll see about getting the article published once I'm sure things are right. After that, Wrath should be pretty easy to model, and IS, MF might take a little time. Then I'll move on to rotations, mana, and non-static buffs (Trinket procs, 4T5, etc).

I'm attaching the latest version of my spreadsheet, which has been drastically updated, including the new Crit calculations. It has sections which aren't implemented yet in preparation for the future, like spells which are not Starfire, and rotations. There's a new page for Talents and Buffs, and a math page for interesting info. In general, if something is greyed-out, it's not implemented yet. Make sure to read the comments on entering Hit and Crit ratings.

I was using ef's spreadsheet and I was stacking it because i dont clip my dots, so i dont really care to hit threshold. However, my itemization is obviously incorrect now that I saw ado's math.

Adoriele - I looked at your spreadsheet but I guess it just isn't dumbed down enough for me. I don't understand what the numbers mean under the pure rare gems column to the Hybrid Epic Gems Column.

I want to double check that I'm understanding everything correctly. After 1077 spell damage, you want to stack haste in a 1:1 ratio with damage. Since I have around 1265 spell damage when in my pve gear(armory has pvp gear on it now), doesn't that mean that having around 185 some haste rating is pretty close to the sweet spot?

I've seen someone say don't use +10 haste gems till 1394 spell damage but at that level shouldn't you already have 317 haste? Is there really a benifit of using orange gems over yellow gems when you are already way above 1077?

I guess I just get confused between the two because I really don't know which rule I should be following when you guys say both are correct. The add haste in a 1:1 ratio after 1077 rule is great. I love it because it's so easy for me to do. I'm obviously not a math guy in the slightest so the simpler the better and this fits.

Like I mentioned before I'm at roughly 1265 spell damage and 182 haste. So I'm really close to that 1:1 mark however to do this I used some +10 Haste gems. So that seems to be right, but I'm no where near 1392 spell damage which is where you said I should be to start using +10 haste gems.

You guys kinda make it sound like if you could add 1 gem slot to your gear Spinels are a better choice because of the 1.2:1 ratio of gems. But if your upgrading a piece of gear aren't you more than likely adding at least a few spell damage also. If a item added more spell damage to your set wouldn't it be better to add a haste gem so as to keep your set closer to the 1:1 Spell Damage to haste ratio?

Sorry I'm kinda dense when it comes to these things.

Can anyone help me calculate the spell coefficient on moonfire?? I am writting a tutorial about moonkins, and I would like to write down how that is calculated. :x

The thing is, I can understand how the calculations on the wowwiki is made, but the one at the think tank on ej I can´t. And the thing is, the results posted on EJ and wowwiki don´t match.

If anyone can run the math for me, I would appreciate.

Don't worry about it. It's damn confusing sometimes to me, too. Really, the 1077 number is a theory thing. If we had the ability to design our gear as we wished, we'd want to always have our spell haste equal our spell damage minus 1077. But we don't. We get close with gems, but then Blizzard throws a curveball and makes Spinels and Rubies have more stat points than Dawnstones and Lionseyes. So we have to deal with the trade. And as it turns out, the trade is only better once you get to 1392 Spell damage. At that point, you'd want to keep an equal number of Spinels and Lionseyes. Before that point, the extra stat points allocated to damage on the red gems outweigh the deficit caused by straying from 1:1.

What's 1077 useful for, then? Well, it's a good rule of thumb for when to start gearing toward Haste. Not socketing - we've shown that 1392 is better for that. But it's a good point to start seeking out actual gear with haste on it at that point. I'll try and pick up some examples later tonight.

Here's the best I've seen it: WoW Forums -> YABi2.4R (Yet Another Balance in 2.4 Rant)