I noticed that exchanging items with +25 ilvls did not make any difference to my DPS so decided to run simulations to see what’s happening. My Rogue is very new at 440 ilvl and not great gear. This is the stat weights as I have determined:
In other words 100 points in Crit are worth the same as 114 points in vers, or 140 points in haste, or 225 points in mastery (what!!!) and 230 points in agility (?!), which is my main stat…
How is this class designed so badly with such great disparity in stat importance?
That has nothing to do with bad design, stat weights are influenced by many things and those weights may change wildly, even when you only swap a couple items around.
I can’t speak to rogue’s specifically very much, but most classes/specs have moved away from stat weights as a reliable gearing strategy, even moreso once you throw some corrupted gear into the mix.
Because rogue’s abilities directly interact with how much crit you have?
Crit/vers are your best stats, you get haste passively from azerite/adrenline rush/alacrity and outlaw’s mastery just sucks.
I don’t see how having to pay attention to your stat weights is bad class design.
It is expected to have some variation in stats weights, but 2.3 times X? What is the point of mastery, which every class is meant to have as their “special” stat, with special entries in the spellbook specifically for that class? Someone sat down to design mastery for the outlaw rogue, and design the extra abilities it would introduce, and it is plainly pants. Were they just lazy or not bothered at all ? Probably a bit of both. And this does not apply just to the Outlaw, I bet there are more classes with “2.3 x” or even worse.
your worst stat being 2.3x less worthy than your best stat is a very mild range. I think you’re mistaken in how big of a deal that is.
It doesn’t make crit-gloves 2.3x as strong as mastery-gloves for example, as the secondary stats are only a part of the deal.
First off every item will have 2 stats, so that already cuts the impact in half, next the primary stat (and thus by extension, the ILvL) comes in which often plays a bigger role (the agi you listed above doenst belong in the list and is misleading), and those things in turn are overshadowed by the corruptions.
Look at it like this, at the current level of gear most of us use, about 5% of the power (stat value) of a Leg-item with a socket, comes from it’s gem.
A gem is the cleanest place to compare secondaries as you can freely pick between +50 of any of the 4 stats (we’ll just assume you’ve already exhausted the 2 unique primary stat gems on other gear to keep things simple).
So when using these Legs, an informed “you” would pop in a Crit gem, and a less informed person might pop in a Mastery gem.
That means the value of your item is 95% + (5% / 1.0) = 100% of it’s potential, and for the other guy the item only lives up to 95% + (5% / 2.3) = 97.1% of it’s potential.
In other words, the BEST choice makes the item 3% better than the WORST choice.
The depreciation of mastery is mainly due to essences and corruptions. If you use a passive major essence, like Vision of Perfection, and corruptions like Strikethrough or Void Ritual, instead of Infinite stars or Twilight Devastation then you’ll see mastery being closer to other stats.
I’m guessing this has been an unforeseen consequence but this is equivalent of picking a bunch of “do this on crit” talents and then be surprised crit sims miles better than your other stats.
No, it’s all simmed. If for example I take a +100 crit food buff, I would need to take a +230 agility/mastery food buff to get the same DPS increase. If I fit a +100 crit gem, I would need a +230 agility/mastery gem to get the same DPS increase. My only corruption is a single 50 Infinite Stars. Maybe I can sim without it to see how the weights change.
If you’re looking for how you should build your char, throw sims out the window, find someone who is well geared and well known for playing said class in the content you’re interested in taking part in, check their stats in the armory and aim for those. Had I stuck to sims, I’d be spamming vers/haste on my fire mage instead of mastery.
I get that, there’s no other way to get accurate weights.
Not really sure what point you’re trying to make.
It still remains the case that, the best stat being 2.3x as valuable as the worst stat isn’t some massive distribution. Even (much more extreme) 8.0x scenarios haven’t been that uncommon in earlier expansions.
I’m well aware of how stat weights work. Your example is correct in theory, but completely meaningless to the actual game.
When it comes to stat foods, the question is “which of them gives me the biggest increase”, as you can only use one at a time. So your complaint:
has absolutely no value here. It wouldn’t matter if crit was 2x as good, 1.03x as good, or even 50x as good as mastery, in all these cases you’d pick crit food over mastery food.
And sure, you could come to that correct decision by simming your stat weights, but it’s completely irrelevant what the exact weight values are, it only matters which one is best.
Where the stat weights do come into play, is when comparing multiple items. Say you looted 2 Belts, they have the same ilvl, socketcount and the same amount of Agility, and neither has a corruption.
Belt A has 230 crit + 70 haste
Belt B has 190 crit + 110 vers
Someone who only knows what his best stat is (crit) might pick A (because it has the most crit), but has no idea if it was the right choice.
Someone who knows his stat priorities (crit>vers>haste>mas) might pick B, because it has the 2 best stats on it, but also can’t really know if it was the right choice.
Someone with stat weights does actually have the information needed, to get the right answer. using your weights:
The problem with using stat weights though, and the reason why i said it wasn’t the recommended basis for selecting gear on anymore, is that you have to recalculate the weights every time you change anything.
Switching your current belt for Belt A above will change the stat weights for all 4 stats. Switching your belt for Belt B will change them all as well in a different way.
Changing a talent, getting a corruption on an item, swapping a trinket(effect), …
All these things can change the stat weights around.
And since you have to use a sim to get those weights anyway (if you want accurate ones), it’s a lot more sensible to simply sim possible upgrades you’ve looted and compare the sim results directly.
I understand and already know what you are saying. You think that a small or large difference in stat weights does not matter since you will choose the one item (even marginally) better regardless. That is correct.
But in terms of stats potency distribution it is a very bad design that results in such big discrepancies. You have mentioned 8x I am not sure I have seen it but then again I have not looked hard enough.
My thinking is you cannot have a system where some of its parts are made irrelevant. What is the point of mastery if I will knowingly remove it? What is the point of that Blizzard developer spending time trying to build/code the mastery effects on the Outlaw rogue (Mastery: Main Gauche), when no one would even care, and in fact, everyone actively staying away from it?
It is like a developer spending time and effort to create 10 class spells, or talents, and because of this unbalance, players only ever using 5 spells or talents, and completely ignoring the others.
In my view all stats should have their role to play and they should all be competing for first place, eg haste comes first but if you add too much then crit comes first and then perhaps mastery etc. I’d prefer a system like that where all its parts are almost equally relevant. And I would also prefer all the parts to have synergies resulting in different playstyles that you could choose, eg Rising Mist / Spirit of the Crane build.
This is interesting, why would sims give wrong results?
And people keep mantras like “sim it”…
I had similar situation with my enha shammy - experienced people saying stack haste-mastery, while sims said crit-versa even with haste as low as 17%.
Who to believe?
It all depends of gear level, there is early gear (let’s say 440) and those are different weights for most classes, then with high end gear (let’s say everything 470+) you drop those stats for completely different ones because how scalling works. Both advisors and sims are saying truth, the problem is that you are not talking about exactly the same situation.
Edit: Note that even picking different azerite traits can change your weights completely.