+/- % protection in item comparison

[muzzz]

Does anyone know how the +/- % protection value, that's displayed for in-game item comparisons, relates to the actual damage mitigation formula (as in (1 - armor) * (1 - dodge) * (1 - res))?

[MagisterLudi]

I remember seeing a Blue Post somewhere about it now reflecting Resist as well as Armor. Not sure about Dodge though.

Seems to be a very confusing stat to try and weigh Armor mitigation vs. Resists.

[Durikkan]

I just did a little bit of research and the following seems to be accurate. I'm going to leave out the actual calculating of Damage reductions for the sake of simplicity.


Just by intuition, I think it's the final total damage mitigation %, including dodge.

The Melee bonus is taken into account.

I.e. my monk has 6487 armor, 876 res all, and 1528 dex.

That's 91.94 % Damage reduction (DR) before dodge, and 94.35% DR after melee bonus, and 96.35% DR after dodge

If I equip a diff amulet I change to 6580 armor, 888 res all and 1571 dex.

That's 92.09% DR before dodge, and 94.46% DR after melee bonus and 96.44% DR after dodge

Switching between these amulets gives me .09% final DR (96.44 - 96.35), which shows up as 0.1% protection in the comparison tab. This is slightly misleading because going from 96.35% to 96.44% is actually a 2.7% increase to my Total effective HP after dodge.

Edit: second test

I change from 6580 armor, 888 res all and 1571 dex.

That's 92.09% DR before dodge, and 94.46% DR after melee bonus and 96.44% DR after dodge

To 5191 armor, 824 res all, and 1482 dex

That's 93.16 % With Armor, res and melee, and then 95.54% final DR after dodge

96.44 - 95.54 = 0.9% final DR lost, but the character sheet shows 1.0% prot lost, so I'm not sure if that's a rounding issue or if I made a small mistake somewhere. If I use maximum precision that's .901% DR lost, so I suppose it is remotely possible that's being rounded up to 1.0 but again it could be my mistake.




The higher your defenses, the smaller the numbers will be on the protection comparison. At this point, equipping an item with a +2.5% on the protection comparison would double my effective HP, while someone with very little armor might see a +6% for equipping an armor that raises their armor by 100 points that would give them ~6% more effective HP.

Green would always be better, but the number itself will be misleading.

It does take the Melee Damage reduction into account, so Melee characters will usually see less protection bonus on that comparison card because they're starting from 30% DR instead of 0, even though they're getting the same relative benefit.

[torand]

The guy above me is correct; albeit verbose. The "Protection" stat is how much more or less damage you will take on average per hit out of the total damage the attack would do without any protection at all. I wish your character screen would show you your current "Protection", as it would make comparing items a bit easier.

[Katniss]

I actually was wondering something along these lines.

Durikkan, thanks for supplying your detailed findings. You mention the Monk's natural 30% DR and that got me thinking about the use of String of Ears. If I have a SoE with 16% melee reduction equipped and then hover over a possible belt choice that has 75 AR (for example) and the protection number displays green, does the calculation actually take into effect not having that 16% melee reduction? I guess I don't know why I would question it, but I'm curious if my survivability would actually go up wholesale using a rare belt with more AR than my middling self-found SoE.

Thanks!

[Durikkan]

...does the calculation actually take into effect not having that 16% melee reduction?...

Just tested it, and that is completely ignored by the comparison formula. I don't know if that's an accident or just due to the complexities of determining how useful the modifiers are. Depending on a person's play style, they might get hit by mostly melee attacks, or mostly ranged attacks, or somewhere in the middle (though they could just average it like they did for resistances).

So there wouldn't really be an easy way to know if that helps other than doing a quick calculation if able, or just using a spreadsheet.

[Katniss]

Thanks for the research and quick response! I guess I'll do some field work testing and see what makes me feel less squishy, heh.

[torand]

I actually was wondering something along these lines.

Durikkan, thanks for supplying your detailed findings. You mention the Monk's natural 30% DR and that got me thinking about the use of String of Ears. If I have a SoE with 16% melee reduction equipped and then hover over a possible belt choice that has 75 AR (for example) and the protection number displays green, does the calculation actually take into effect not having that 16% melee reduction? I guess I don't know why I would question it, but I'm curious if my survivability would actually go up wholesale using a rare belt with more AR than my middling self-found SoE.

Thanks!

The "Melee Bonus" that Katniss was discussing in his testing post was the 30% damage reduction from all damage that melee get. "Melee Reduction", "Ranged Reduction", and "Elite Reduction" are all specific to some damage, and the "Protection" stat only takes into account things that reduce ALL damage by a predictable amount (Same reason it doesn't take into account block).

[The Rockman]

The melee damage reduction Durikkan talking about is the 30% bonus global damage reduction all monks and all barbs get. TBH it would be better to call it melee classes damage reduction really.