xaijin_sama, if you read my write-up you would see that I have listed cower as being the 6th option, right underneath committing suicide (which I hope noone actually takes seriously, meant it as an option by way that it would reduce your threat, but not in a desirable way). In my original post I should probably have stated it was to reduce threat relative to the tank, rather than reduce threat absolutely.


The point is, switching to cat and cowering will reduce your threat. It will burn mana, but for a lot of raiding moonkin I am sure mana is not an issue. It may get you killed, and the bigger effect is that it effectively encompasses the 3rd option of doing nothing while you are running to the boss, cowering, and getting back into position.

The Shrouding potion isn't fantastic in WOTLK, but I included it after the suggestion here on the criteria that :

1 : it is something you can personally use without relying on others
2 : even if the amount is small, it does reduce your threat
3 : what other option do we have aside from doing nothing :P

The only option that significantly reduces your absolute threat is using a paladins HoS

On an unrelated note, I am happy to see the tier 8 changes. As has been noted it had to happen, and it puts an end to druids recommending to upcomming druids to go for older tiers. T8 was nice while it lasted but it's time to move on!

From MMO champion:

Really dont know what to make of this. my napkin math is too inferior for it. but this could mean ol' school haste stacking and farming SF eclipse only.

It's messy because that's lost DPS time too. If you have to wait for 2 seconds, it's probably not worth it. If you have to wait for .2 seconds though, it is worth it. Another reason why serious DPS optimization in this new model is an incredible pain.

To flesh it out more, let's say that WE added ~300 DPS to a floor of 8000. Around 4% (just grabbing rough numbers from sheet). 4% of a 30s cycle is around 1 second. So that's a first-order estimate of a cutoff for how long it's worth waiting to time the Starfire. Note, however, if you're planning to wait more than 1s, you should be using that time to cast an instant, further complicating this process. Just to make it even worse, recall that the correct amount of time to wait depends on whether your previous spell happened to proc Nature's Grace.

So, for optimal DPS, you have to see when your second-last Starfire is going to land. You have to then decide whether to fit in an instant or not before you start the final Starfire. You have to do this computation for two cases however, depending on whether or not the second-last one procs NG.

This is getting beyond the point where I expect anyone to do it consistently without an addon (there's no reason not to make such an addon though, that tells you what to cast and gives a "golf"-style timer as someone mentioned above). But for now I'm just pointing out how absurd the mechanics become when you try to optimize the currently presented scheme.