For a bm hunter, is going past 60% crit just a waste of time ? I’ve read in the past 40% is good, but with sever etc, I can get into the 70 % range pretty easily.
It’s not a waste, but you will get more dps by mixing a few Expedients and Masterfuls instead.
If I’m not mistaken the main reason why crit loses value is that you end up capping the uptime of Dance of Death. I’m at 64% crit with crit food, and my DoD is usually up above 90% of the time. I have a few logs with 99% uptime. My sims say I shouldn’t take any more severe at this point.
But your damage will still go up if you go ham on crit. You only lose about 2% or 3% dps at most, compared to a more optimal corruption setup.
1 Like
This, and also crit has diminishing returns by its nature. If you have 59% crit, you have a 59% chance that adding another 1% will do nothing. The higher your crit, the more likely your chance that adding crit will have no effect.
Meanwhile, haste, vers, and mastery have much less natural dr.
Iiiiiiish. I suppose that might make some intuitive sense, if you squint at it a bit.
It’s more that additive stats all have a bit of innate “DR”, as it were. The main reason is that increasing your damage from, say, 150% to 151% gives you less of a percentage increase than increasing your damage from 100% to 101%, despite both requiring the same amount of stats. 100 → 101 is a 1% increase, while 150 → 151 is a 1.51 / 1.5 = 0.667% increase.
This is true of all additive stats. However, different stats combine differently, and have different baselines. Haste rating is additive with other haste rating. For example, if you have 680 haste rating, which is 10%, and you gain another 680 haste rating, you now have 1360 rating, or 20%. (I know, simply arithmetic, but it’ll make sense in a sec). Haste rating is, however, multiplicative with other percentage haste effects (and those are multiplicative with each other). For example, if you have the above 20% haste from rating, and you get Bloodlust (30% haste), rather than having 20% + 30% = 50% haste, you in fact get 1.2 * 1.3 -1 = 56% haste. Because of this, haste rating effectively starts at 0, your first point of haste rating is at full strength, because it’s multiplicative with any other sources of haste.
Versatility is additive with other percentage sources, but there basically aren’t any other sources of percentage versatility, so again, your baseline is 0, first point is at full strength.
Mastery is also additive, but it also lacks other non-rating sources, but you start with what effectively amounts to 8 “points” of mastery, which in BfA has the same impact as 576 mastery rating (the actual percentage benefit of these 8 “points” is variable based on spec, but it’s proportional to the amount of benefit you get per point of mastery rating). This makes mastery rating start on DR, since you’ve already got some built-in mastery that the rating is adding to.
Crit is basically in the same boat as mastery. We get 10% crit innately, and crit is all additive, so your first point is already adding to that 10% existing crit, rather than 0% like haste. Since it starts already on this “DR” effect, it also progresses down it faster.
So it’s not DR insofar as you have an increasing chance of more crit not doing anything. At 200% critical damage (which most classes have, including hunter), 1% crit is an additive 1% increase to your total average DPS. The probabilities average out.
One other thing that factors in here is that increasing one stat innately increases the marginal value of other stats, while decreasing the marginal value of more of the first one. For example, if you increase your haste, then you’re getting in more damaging abilities per minute, which means increasing your crit chance will increase your damage per minute (and thus DPS) by more than it would have without that haste. The haste increased the absolute DPS gain per point for crit (and for vers and mastery for the same reason). However, because of the additive nature of the stats (above), that haste also decreased the value of further haste. This is why stat weights can change so wildly as the relative balance of your secondary stats shifts with gear upgrades.
That said, Khallid nailed it, at least for BM’s Dance of Death build:
Basically, Dance of Death scales exponentially with crit chance. The higher your crit chance, the more likely Dance is to proc, and the more often Barbed Shot resets and thus can be used for another try at Dance of Death. Once you reach a certain point, your uptime is already so high (though you’ll never truly hit 100% uptime unless you hit 100% crit chance) that adding more crit can only increase it so much more. The marginal value of more crit goes down rather drastically once you hit that point, and keeps going down. At that point, Expedient for faster GCDs and Masterful for more pet damage during those super-often Bestial Wrath windows strongly takes over.
This is actually rather in contrast to how percentage corruptions work for most other classes. For most of them, once you get a percentage corruption, you stack that stat even more, which makes more of that same corruption stronger, which makes more of that stat even stronger, and it feeds itself. The corruptions themselves are additive with each other (at least, I’m about 95% sure of that), but the stat effect feeds it.
For example, as MM (ya, I know, I just really don’t like BM), the sims are telling me to stack at least 2 more Expedient beyond the 4 I already have, and on my rogue, I already have 5 Expedient, and the sims want a 6th, and I kinda at a toss-up on whether I should get a 7th, or swap to stacking Severe at that point.
But neither of these specs really have softcaps on those stats. In the Rhythm will just happily keep scaling into eternity, and it takes a staggering 333% haste for Aimed Shot to hit the current GCD floor (and I’m only at around 190% with lust and most of my procs). As for my rogue, those bleeds just keep scaling as well, with no cap, and Exsanguinate, Shrouded Suffocation, Subterfuge and just plays into that stupid well.
Sim your gear. Without seeing your hunter, I’m not sure how anyone can tell you specifically what to do.
I think the other suggestions are all have it right; but, sim whatever trinkets and rings (potential procs). I run 2 crit, 2 haste, 2 mastery atm. Keeping within our raid leaders’ request of “39” corruption, I won’t be able to put a low crit (10) till next week.
“In my opinion”, (I’m currently 50% crit, 25% Haste, 46% Mastery and 5% Vers) if you are @ 60% crit, mastery or haste must be suffering. I have a 475 Harlans, and a 475 Galecaller (Prefer VIta curse you Ra-den) and 465 Logic Loop of Division and a 465 Overclocking bit band. When I simmed on AMR/Raidbots more Haste would benefit me more.
…but, keep in mind that you have to SIM, sim, sim. I also do 20ish keys and usually keep a healthy balance for raid and mythic+. So maybe my opinion is skewed.
Use Raidbots, please. AMR’s “sim” is a freakin joke. It’s a pile of guesswork, rules of thumb, estimations, and incorrect mechanics calling itself a “sim”. You’d literally get more accurate results by simply guessing.
So, yes dance of death incentivizes trying to keep non-crits off the table, so crit is a hige stat for bm. Arguably, you could just stack crit to make your life simple and do great.
But, i really dont understand anything else you said in how it relates to my point.
Versatility and mastery will always increase your damage. Crit is rng, so it is possible it wont. Thats not to say it’s bad. Its just to point out that the more you have, the less value it normally gets, unlike other stats.
Other stats lose value “in comparison” to other stats. Crit loses value “in comparison” and by the nature of what crit does.
Obviously, you could have over 100% mastery. You could never have over 100% crit as an extreme example.
The closer you get to 100, the worse it is for crit. And, it really does start to fall off a cliff. Getting close to 100% mastery is not an important breakpoint, as an example.
This is why crit is only good for specs that snowball it like feral and bm.
But yea, sim is always the answer.
They literally sim the same and use the same information. You’re doing something wrong I’d guess on the values.
RNG doesn’t matter, the average is what’s relevant. If you have 50% crit, half the time it does nothing, half the time it doubles your damage. The average is +50%, and the average is what’s relevant here.
No, my point is, that’s true of all stats. Additive stats give less of a percentage gain, relative to your current DPS level, the more you have of them. Crit is no different from other stats in that respect. Your intuitive assumptions about the RNG benefits are leading you astray here. Going from 10% crit to 20% crit provides the exact same increase to your average DPS as going from 10% to 20% versatility.
Yes, caps exist for some stats, but since they are largely out of reach except in the most extreme (and not recommended) corruption configurations, they aren’t relevant here. Caps are entirely different from the diminishing returns you are arguing exist here, because the crit cap is a point where the returns literally just suddenly vanish, not just the point where they happen to hit zero. Going from 99% crit to 100% crit is still a 0.5025% increase to your DPS (ignoring non-linear scaling effects like Dance), and the same is true of moving from 99% to 100% versatility.
That’s only true because of how Dance works. Outside of dance, it’s literally just the additive nature of the stat, that going from, say, 99% crit to 100% crit provides barely more than half the benefit as moving from 0% to 1% crit (though you can’t have <10% crit). But the exact same mechanism applies to the other stats, going from 99% mastery to 100% mastery provides barely more than half the benefit as moving from 0% mastery to 1% mastery (though, again, you can’t actually have 0 mastery).
They literally do not. Raidbots uses SimC behind the scenes. AMR uses its own proprietary engine, and if you scan through their database of spell effect, you find soooooo many errors, incorrect mechanics, or “close enough” things in there. AMR is considered worthless by the theorycrafting community for a reason.
They use the same information and you “literally” are doing something wrong. AMR if anything will sim slightly lower.
They straight up do not. Examples:
Killing Machine is modeled in their system as a flat 50% proc chance on crit. They even note that it doesn’t precisely match in-game, but that they don’t really care:
https://www.askmrrobot.com/wow/theory/mechanic/spell/killingmachine?spec=DeathKnightFrost&version=live
Comments
The actual proc mechanic for this has not been verified by blizzard. Data indicates that it is more than 4.5 rppm. 50% proc rate is what it comes out to on average. The actual mechanic may be different, but for simulation purposes 50% gets us results that match what we see in game.
In actuality, Killing Machine is a progressive 30%/60%/90%/100% proc chance on successive critical auto-attacks, reset on proc.
They still have a maximum extension configured for Metamorphosis via Demonic for Havoc (https://www.askmrrobot.com/wow/theory/mechanic/spell/metamorphosis?spec=DemonHunterHavoc&version=live, despite that having been removed in 8.0.
They only use Death Sweep at 3+ targets, despite it being better at 2+ (3+ is for Blade Dance): https://www.askmrrobot.com/wow/theory/rotation/amrdemonhunterhavocmain?spec=DemonHunterHavoc&version=live
Double Tap is only used if Trueshot is ready to use as well, which can significantly delay it if you aren’t running Calling the Shots. They also prioritize Rapid Fire over Aimed Shot, even during the Careful Aim phase.
https://www.askmrrobot.com/wow/theory/rotation/amrhuntermarksmanshipmain?spec=HunterMarksmanship&version=live
They still have an outdated 50% damage bonus for Aimed Shot on a target that has not yet been shot, which was removed in 8.0: https://www.askmrrobot.com/wow/theory/mechanic/spell/aimedshot?spec=HunterMarksmanship&version=live
Aspect of the Wild has no consideration given for the Barbed Shot charge granted by Primal Instincts, nor is it lined up with Bestial Wrath, it’s simply cast on cooldown and the extra charge is wasted.
https://www.askmrrobot.com/wow/theory/rotation/amrhunterbeastmasterymain?spec=HunterBeastMastery&version=live
It also fishes for Dance procs regardless of critical strike chance (should only be done if you have 3x Dance of Death and a sufficient amount of critical strike chance).
Oh, and their “gearing strategy” sim still says it needs to be updated for 8.2, much less 8.3.
Oh, and their “sim” says my hunter should be doing about 8.5k less than SimC does, and under-anticipates my Dance uptime by nearly 10%. For no readily apparent reason, it thinks I should only have a 90% uptime on the Focusing Iris minor, on a patchwerk fight (SimC properly reports at ~99.8% uptime).
Ya, man, AMR is the exact same. /s
Place is a dumpster fire. If you want to continue using it, go right ahead, but please stop trying to encourage others to. SimC, and by extension Raidbots, is a much more reliable (and regularly updated) source.
1 Like