Catching up on forum stuff after the holidays
I’m looking at this case in particular, because it’s an interesting one. It seems like zero haste is a pretty extreme solution, even if haste is less valuable than some other stats.
I see a couple of things going on:
1.) With the gear available, it is possible to heavily favor two stats at the total exclusion of the other two. So, any calculation that finds crit/mastery to be stronger stats will be able to push you into gear that gets almost exclusively those stats when any item in the game is considered. In the past, this wasn’t possible so you’d always see some of every stat.
2.) My calculations do find haste to be less valuable than crit and mastery - but not by a lot. Forcing haste onto the gear is only reducing the score by something in the range of 2%. I’m sure that simc is finding haste to be slightly better, and thus all the other guides are saying haste is the best so stack haste. But my calcs show crit/mastery to be slightly better so the optimizer is saying to stack crit/mastery.
There isn’t much of a downside to stacking haste, crit, and/or mastery on your gear. I’m trying to figure out what it is that is making the other guides out there like haste so much… I haven’t quite found it yet. I’m not sure exactly what I could do to the calculations to make them favor haste that much.
A big chunk of your soul shard generation comes from demonbolts that are a result of dreadstalkers giving demonic core charges - more than half. The rate you get those is unaffected by haste (dreadstalker cooldown is not affected by haste). So, to me, that is a significant hit against the value of haste. It makes mathematical sense for haste to be a weaker scaling stat.
More haste could give more value to grand warlock’s design, but it actually has a surprisingly small effect. Wild Imps don’t do more damage with more haste, either (but you can summon more imps total) - more haste in AoE might actually make it harder to get big implosions and implode every imp. Imps only cast 6 felfire bolts before they despawn, and their cast time is affected by haste. Ideally you want to be casting implosion less often than you cast hand of gul’dan - every other hand would be great if possible to reduce gcds spent casting implosion.
I’ll keep an eye on it, but going through my calculations, nothing is jumping out at me right now that I could change to skew the results toward haste. I think what I’d have to do is somehow make mastery less valuable.
Also, the warlock set bonus proc rates are not available in the spell data. Simc is using a guess and I’m using the same value they are for consistency right now, until we can figure out what it is.