I've spent 12+ hours playing around with simulationcraft since this article went live. The only thing I can say for sure is that haste is still our 2nd stat (behind intellect). Mastery does appear to be pulling ahead of crit like the article says - it was doing this in 372 gear on live, so this is not too suprising.

Attempting to build a "BIS" set using these weights will lead you to a set that does 500 or more less damage than a haste-focused set. I can't tell you why Simulationcraft gives bad stat weights for haste for shadow priests sometimes. I've formed a few theories, but ultimately I can't find a way to prove any of them with the program. For now all I can say is that "sometimes Simulationcraft fubars the haste stat". It just does... there's no real profound explanation beyond it.

If you redo your trinket calculations with haste set to somewhere between .55 and .6, you'll get a more accurate view of what they are woth in relation to the trinkets that are already out there.

Another thing that is important to note is that the stat weights in the article are based upon 372 gear. As you move into heroic tier 12 (iLevel 391), you will start to see all of the stats scale higher in relation to Intellect. Each stat will bump up nearly .8 points in value (except for spellpower which remanis constant). For example, mastery will be closer to .49, crit will be closer to .48, haste will be closer to .6 or more. This is no different than the gradual increase in value that we saw when moving from 346 gear to 372 gear (so again nothing new).

As a result of the above, the "worth" of the trinkets with secondary stats will gradually get better as your overall gear level increases. Because the stats generally stay relative to each other, you won't see any significant changes. But you may occasionally see trinkets that are close in value flip back and forth (I think this just means if you have a choice between two trinkets that are very close in PP value, you can opt to choose the one that you enjoy using (or perhaps percieve to be stronger) over another that is a few PP difference.

Anyway, I hope all of this research helps. Thanks for writing the articles. I look forward to the next one.

-Kilee

Reply