EDIT: I'm a little scared to post this cause I'm afraid people will say I'm an idiot for suggesting against the common rotation. I just did the math on it and it seemed solid. I'd love to see math countering it or anybody pointing out a mistake I made.
EDIT2: I already found a mistake in my own math, lemme fix that. (currently editing post)
EDIT3: Fixed my mistake
Basically, I saw someone suggest this, and I did some looking at it. The thing I hear all the time is that you should prioritize Thundercloud Procs over Firestarter Procs because Thundercloud Procs have a chance of immediately overwriting themselves. Someone claimed that the standard practice of prioritizing Thunder III over Firestarter procs is not necessarily a good idea because Thunder III does less and can eat it's own proc every 3.0 seconds, there's only a 5% chance of it eating itself. But Firestarter has a 40% chance of eating itself and you can use Thundercloud procs to buffer after a Fire to improve your chance to not miss a Firestarter proc.
This is the evil bane of the standard rotation:
Fire (Firestarter Procs) -> Fire (Firestarter Procs) -> Use Fire III
Under the standard wait until the next Fire to cast Fire III, you have a 16% chance for Firestarter (40% * 40%) to overwrite itself and lose it's proc. The worst part, is hypothetically it can happen twice in a single Astral Phase. It's not common, but it has a 2.56% chance of occuring, that you get Firestarter on 4 of the Fire Casts who's Firestarter can be used during Astral Phase.
What about Thundercloud? Well, it's almost never going to overwrite itself in the standard rotation. And of course it has a very high visibility of those 0.25% chances to proc Thundercloud twice in a row (Which happens on average at least once per 10 minuter fight). If you waited on the Thundercloud one GCD, it has a 0.25% chance to overwrite itself. If you wait two GCDs, it has a 0.4875% chance of overwriting itself. But it should reduce the number of eaten Firestarter procs by 5%, which means you would only have 14.4% of Firestarters Eaten, or 1.56% more Firestarters landed.
-IN A NUTSHELL-
Standard Thundercloud > Firestarter Priority
16% Firestarters Eaten
0% Thunderclouds Eaten
Hypothetical New Rotation: Firestarter > Thundercloud only used after a Fire I when you don't have a Firestarter
14.4% Firestarters Eaten
0.4875% Thunderclouds Eaten
1.56% more Fire starters landed
To make this even more solid, look at the damage from Firestarter and Thundercloud:
Firestarter Astral III = 396
Thundercloud Proc = 340
Ultimately it's not a huge difference, it's only about 1.08% more successful procs landed, though every little bit helps, and it has the added bonus of being /slightly/ less RNG reliant. I may have made a math mistake somewhere, but lemme know. This was only some rough napkin math I did, and I would need to do more extensive math to actually present this with certain accuracy.

Reply With Quote






