Mitigation improperly considers mob buffs

[ficocelliguy]

Mitigation currently uses the base unbuffed physical hit to calculate the damage prevented, but it should include monster buffs to avoid showing negative mitigation in incorrect scenarios.

[PsychicMuffin]

this is a tricky issue, because imagine the only adjustment you have is abyssus, you will end up with negative mitigation right? I think because we are considering map/curse adjustments to be part of the damage adjustment, if they make your mitigation negative, so be it. I think it would be much weirder to see you mitigate 120 damage of a 100 damage hit, since we don't show the "buffed" damage anywhere really.

[ficocelliguy]

I think I'd like to add a row to the tool tip for 'buffed hit' or something that only appears if monster buffs are applied. Negative mitigation from abyssus makes sense, but I like the idea of showing the actual amount of damage prevented including buffs.

[PsychicMuffin]

So would we do vulnerability curse the same way as well? Still seems a bit odd to me, but I think to me it boils down to why are we showing 'damage mitigated' stat in the first place? It seems like the goal is to see how much of the original hit you avoided, so the number loses context. Yeah we can show a 'buffed hit' but it seems like that adds steps to give the other number the context that the user expects it to have without looking for another number. This is why user requirements are good isn't it.

[ficocelliguy]

Honestly the entire motion to show damage blocked is the same as in world of tanks: its satisfying to see how much your defenses are doing, and showing the largest version of that number is the most satisfying. It's not actually a useful stat, its just for context to understand slightly different how eg. armor works. The idea is to be able to look at how much you actually took and how much you mitigated in that context. If you were to describe what percent of damage you blocked it would be in relation to the bottom line hit, including mob buffs etc.

[ficocelliguy]

Maybe it should be a percentage instead of a flat number?

[PsychicMuffin]

Hmm, maybe, but then we definitely have to stick to the original value. We might have to just rely on user testing to give us useful feedback about this value.

[ficocelliguy]

I'd like to do testing on it, but I think the root of the problem is that the original hit value is kind of incorrect. If there are monster buffs, you are not actually taking the damage of the listed hit size, its a larger hit than that when your defenses start working. I honestly don't think that many people will even use the monster buffs section, because it is a changing status every few minutes, unlike the rest of their consistent character. (We also don't support the mods that give mobs more damage as x element, for example)

Also, I disagree with your point about definitely sticking to the original value. Everything in this site is about the calculation output, the bottom line, and that should be the context that mitigation percent should be in relation to. If we use the base unbuffed hit as the basis for the percent, then players could feasibly 'mitigate 100%' of the damage and still be taking damage on top of that, which doesn't really make sense.

[PsychicMuffin]

I would calculate the mitigation like so: 1 - (damageTaken / originalHit), so the mitigation would vary from -X% to 100%. I guess we could do the calculation differently for the buffed hit percentage? But I DO plan to support more damage as X element eventually. The calculation is halfway to supporting it after I separated the elements. After I fix #24 it will be fully supported in the logic and we'd just need a way to add it to the UI.

[ficocelliguy]

If someone mitigates 20% of all incoming damage it would be misleading to show that as -10% or whatever just because the monster has a buff. There's no point in showing mitigation in that case. The whole idea of showing mitigation is to get a feel for the change in percent of incoming damage prevented over the different sized hits. Making the user mentally estimate some corrections for monster buffs in the mitigation field is not user friendly.

[ficocelliguy]

I'm a user who is interested in finding out how much of a difference armor actually makes to my character. I would like to see, for various amounts of incoming damage, what percent of that incoming damage is prevented by my defenses. I would also like to know what my effective one-shot health is in terms of incoming damage.

If you would also like to have information provided for how much map mods affect my build's defenses, or what 'base hit' value would one-shot me after monster buffs are applied, we can work on doing that, but I don't honestly know how we would explain those things to the user. I think it might be useful in cases of mobs doing extra damage as an element or whatever. But right now, even as someone who is working on the site and understands what is going on, I don't want to us the monster modifications section because it might mess up the calculations of what my actual build is doing. I think of my effective HP and percent phys mitigation as not changed if the mob attacking me is a rare mod with the "savage" 25% inc. damage mod.

I understand enfeeble can be thought of as a monster modification, but we're not calculating all of the accuracy and crit changes it applies, so when I have been using the site I just put it as -30% damage in the Damage Taken "More" table.

Maybe I'm just misunderstanding what you imagined the monster modification section to be for, so that may be worth clarifying at least between us if not in the description text. I guess as a user I wouldn't include monster untyped % modifications (other than maybe curses) unless I was specifically seeing if I could survive eg the Oversoul's slam attack, or something else where I know the unbuffed hit amount like that. But that's mostly just for messing around, not the core purpose I came to the site.

[PsychicMuffin]

Hmmm, well there are a lot of uses of the site. Enfeeble is kind of a core modifier that goes there, and if you have enfeeble on blasphemy you definitely want it to be considered for your normal damage mitigation. But one thing I considered is say, having your character set up configured via bookmark, rolling up a map, and then plugging the map mods in to see how they affect your damage taken and highest survivable hit. That's a pretty cool thing to do, and is totally separate from seeing if abyssus is more dangerous on vulnerability maps for instance while theorycrafting.

I definitely think this change would have to be at the same time as adding 'effective hit size' to the tooltip (preferably above mitigated amount), but I'm curious what users would like more. I can honestly see the pros and cons of both approaches, but personally I don't think the mitigated value tells me much so I don't really look at it when I'm using the site, which is why I've tried to think what is the most intuitive for that number regardless of use case as I analyze this feature.

But I still think the fact that some items increase your damage taken can make your mitigation negative regardless, so the user will have to be used to it? It's hard to say. Is it more intuitive that the monster buffs section is the one section that isn't part of your 'mitigation'? Could be.

[ficocelliguy]

I definitely think that items that increase damage taken should make your mitigation negative, unless your armor or whatever more than balances it out. I like the idea of bookmarking a build then adding map mods, and I guess in that case the user would expect changes in the end calculations that might look like what you are describing. We should try and formalize the question before testing, I think.