Highest Damage Reduction

[Knot Write]

I was just curious as to which tanking class has the highest amount of damage reduction from talents/glyphs/what-have-you. I figured druids would have the most, considering they have little means of reducing incoming damage other than obscenely high armor and their savage defense ability, but I'm really more concerned about which class would take the least amount of damage from a spell.

[krc]

The armor cap is 75% damage reduction which against a boss is around 49,000 armor. I know for a fact that druids have the highest amount of armor, I am not sure of the excact number but I know that they can get very close if not hit the armor cap buffed.

With the addition of all the bonus armor gear armor gap has been shrunk since fully buffed a warrior can get around 40,000 armor and if you gem completely agility and enchant as much armor and agility as you can a warrior can get 45,000 armor buffed. The reasilistic number is 40k but it is possilbe to get huge amounts of armor now.

[Proletaria]

Wearing armor gear I can hit 40k raid buffed. That's with all the badge gear (save the belt), legionarre's gorget, ony ring, twins ring, 1 armor trinket, 1 stam trinket, and everything gemmed for stam. If you went DW frost, picked up a few ardent guardians, and broke the stam gemming a bit for agi, i'm sure you could get really close to raid-buffed = armor capped assuming the rest of your gear was icc badge and a smattering of icc heroic loot. Not very practical, but quite funny.

In a more practical/balanced set I've got around 36-7k armor and a lot more hp.

I've found the full-armor setup to be a little less than impressive. As blood I've found the stam gearset to give me better survival in practice (mostly tested against festergut). As frost you're going to surpass the armor cap even without glyphing UBA using that much armor gear. Haven't tried it as unholy, but i'm not really interested in using that spec to tank at the moment.

[Edgewalker]

I'm at about 37,500 armor unbuffed and still fall short of the cap raid buffed by a decent amount. It would impossible I believe to hit passively without Ironshield or procs.

[Airowird]

If we're talking spells:

Druid: 12% (Protector of the Pack)
Warrior: 10 % (Def Stance) + 6% (Imp. Def Stance) = 16.6%
Paladin: 6% (Imp. RF) + 6% (Guarded By the Light) + 3% (Shield of the Templar) + 3% (Glyph of Divine Plea) = 19.202724%
Death Knight: 5% (Blade Barrier) + 8% (Imp Frost Pres.) + 6% (Magic Suppression) = 20.204%

This is the best guarantee you can have of a spell hitting you the first time (otherwise Acclimation would be awesome, but dependant on other resistance)

[Satorri]

For the OP, if we're talking general class comparison with damage reduction abilities?

First, passive damage reduction from stances/talents/personal buffs:
Warrior = 10% (Def Stance)
Paladin = 6% (Imp Righteous Fury), 3% (SoT talent), 3% (Glyph of Divine Plea) = 11.56%
Death Knight = 8% (Frost Pres, 10% if Frost), 5% Blade Barrier = 12.6% (14.5% as Frost)
Druid = 12% (Protector of the Pack talent)

Everyone can get the 3% raid buff from BoSanc or Renewed Grace.

For armor, all tanks can use the same cloaks, trinkets, rings, and necks, and all plate wearers can wear the same plate gear with bonus armor. All the plate classes also have standardized armor benefits from talents and possible meta gem. The one distinction for plate wearers is DKs get bonus armor from Frost Presence, and Warriors/Paladins have shields. No one currently gets any armor from their ranged/relic slot, so we'll disregard that difference. Bears have no bonus armor on leather currently, but they get a multiplier in Bear form to translate leather armor to tank levels (totals a muliplier of 7.014 with meta and talents).

So, at a quick glance, here are armor sets (thank you WoWhead) compiling the high armor gear for plate, and I'll just use the t10 feral set for druids (all 264 pieces):
Plate Gear = 15181 base, 4312 bonus, 0 agility
Leather Gear (WoWhead is currently allergic to leather) = 3824 base armor, 910 agility
Common Items (broken too) = 177 base, 6386 bonus, 0 agility

And we can use Distant Land - Item - World of Warcraft as the Druid staff (152 Agi), and Icecrown Glacial Wall - Item - World of Warcraft for shield bearers (8873 base armor). We'll assume the DK is using a Str 2-hander (no Agi) for simplicity. We'll neglect enchants, but apply Kings to Druid Agi.

So, tallies:
Warrior/Paladin = 37,885 (69.49%)
Death Knight = 38,269 (69.70%)
Druid = 36,785 (68.86%)

Note: these numbers could be fudged a bit by enchants (which *could* be consistent from class to class) and gems (marginal differences in gem colors, 6 red slots in the plate, 7 in the leather). They do not include any raid buffs, aside from Kings on the Bear agility, though all static armor increasers and damage reductions should be the same.

The block mechanics take too many assumptions to sum up cleanly here, and all the damage reduction cooldowns work out to roughly the same average damage reduction if used on cooldown. That said, DKs are the only tank class without a block type mechanic.

So, make what you will of the margins. =)

[Dracunis]

If we're talking spells:

Druid: 12% (Protector of the Pack)
Warrior: 10 % (Def Stance) + 6% (Imp. Def Stance) = 16.6%
Paladin: 6% (Imp. RF) + 6% (Guarded By the Light) + 3% (Shield of the Templar) + 3% (Glyph of Divine Plea) = 19.202724%
Death Knight: 5% (Blade Barrier) + 8% (Imp Frost Pres.) + 6% (Magic Suppression) = 20.204%

This is the best guarantee you can have of a spell hitting you the first time (otherwise Acclimation would be awesome, but dependant on other resistance)

Excuse me if this is a dumb question but... if druids get 12% passively from PotP, wouldn’t it be better to just stack to 63% dmg reduction since the cap is 75%? Do they stack?

[Insahnity]

For the OP, if we're talking general class comparison with damage reduction abilities?

First, passive damage reduction from stances/talents/personal buffs:
Warrior = 10% (Def Stance)
Paladin = 6% (Imp Righteous Fury), 3% (SoT talent), 3% (Glyph of Divine Plea) = 11.56%
Death Knight = 8% (Frost Pres, 10% if Frost), 5% Blade Barrier = 12.6% (14.5% as Frost)
Druid = 12% (Protector of the Pack talent)

Looks good so far.

Everyone can get the 3% raid buff from BoSanc or Renewed Grace.

Add Vigilence to the list (not stacking with BoSanc), and not sure if RG still stacks with BoSanc/Vigilence. Also, you might also want to add Sacred Shield in a difficult to quantify way (more on that below).

For armor, all tanks can use the same cloaks, trinkets, rings, and necks, and all plate wearers can wear the same plate gear with bonus armor. All the plate classes also have standardized armor benefits from talents and possible meta gem. The one distinction for plate wearers is DKs get bonus armor from Frost Presence, and Warriors/Paladins have shields. No one currently gets any armor from their ranged/relic slot, so we'll disregard that difference. Bears have no bonus armor on leather currently, but they get a multiplier in Bear form to translate leather armor to tank levels (totals a muliplier of 7.014 with meta and talents).

Sorry Satorri, but Thick Hide adds 10% for leather at max rank (as well as cloaks too I believe).

It should also be noted that armour denoted as being Green (for example, Frost Emblem Cloak's armour value) DO NOT get multiplied. Instead, the portion of the armour that is normally found on that iLVL of item and armour class (plate/leather/cloth/etc.) get multiplied, and the the static bonus (the part that makes it green) gets added on afterwards.

So, at a quick glance, here are armor sets (thank you WoWhead) compiling the high armor gear for plate, and I'll just use the t10 feral set for druids (all 264 pieces):
Plate Gear = 15181 base, 4312 bonus, 0 agility
Leather Gear (WoWhead is currently allergic to leather) = 3824 base armor, 910 agility
Common Items (broken too) = 177 base, 6386 bonus, 0 agility

And we can use Distant Land - Item - World of Warcraft as the Druid staff (152 Agi), and Icecrown Glacial Wall - Item - World of Warcraft for shield bearers (8873 base armor). We'll assume the DK is using a Str 2-hander (no Agi) for simplicity. We'll neglect enchants, but apply Kings to Druid Agi.

So, tallies:
Warrior/Paladin = 37,885 (69.49%)
Death Knight = 38,269 (69.70%)
Druid = 36,785 (68.86%)

Note: these numbers could be fudged a bit by enchants (which *could* be consistent from class to class) and gems (marginal differences in gem colors, 6 red slots in the plate, 7 in the leather). They do not include any raid buffs, aside from Kings on the Bear agility, though all static armor increasers and damage reductions should be the same.

Mostly OK, just wish to elaborate that Kings on Bears affects armour pretty significantly, in that 1 AGI -> 2 Armour. A bear running in excess of 1000 Agi gains 200 armor with kings, and realistically in ICC gear you are rocking way more than 1k agility. Also need to note that Strength of the Earth/Horn of Winter is useful to tank survival, it's not just a DPS epeen boost. Agi for armour/dodge, Strength for slight BV to block tanks and parry for DKs, albeit in arguably negligible quantities unless in a min/max discussion.

The block mechanics take too many assumptions to sum up cleanly here, and all the damage reduction cooldowns work out to roughly the same average damage reduction if used on cooldown. That said, DKs are the only tank class without a block type mechanic.

To clarify, bears have a block mechanic by way of savage defense, which is not quantifiable by percentages reduction, but can be measured in uptime relative to crit rate, modified by its strength based on the bear's AP. As mentioned above, any tank can have sacred shield, but its hard to quantify for the same reasons.

So, make what you will of the margins. =)

To the OP, despite my nitpickings, Satorri has quite correctly illustrated in awesome detail that all tanks are not far apart.

However, if you truly want spell reduction, Unholy DKs win with Antimagic Shell as well as Antimagic Zone are pretty much kings, beside being about as rare as a dodo bird in endgame content. No other tank has magic resistance quite like them.

The next closest tank is warriors, followed somewhat by Blood DKs, who can spell reflect direct damage back (warriors can do so actively on command with spell reflect, DKs can do so occasionally and passively with Spell Deflection talent). Doesn't work with non-targeted spells (i.e. stop standing in clouds made of fire/poison/shadow/etc, and a dragon's breath weapon is not good for your health)

Finally, EVERY tank has access to magic resistances.
-You can enchant your cloak with magic resistance
-Certain classes have magic resistance goodies (Priests/Pally-Shadow, Pally/Shaman-Fire/Frost, Shaman/Hunter-Nature, ye be screwed with arcane)
-You can quaff a magic resistance flask
-Certain items offer magic resistance (Ony rings are an excellent example)
Magic resistance for your purposes is exactly like armour, but for magic damage. As with armour, it is capped at 75%, and generally hitting the resistance cap will gimp you quite nastily, and most mobs since Sarth do significant combinations of magic and physical damage so that you can't stack magic resistance and ignore physical armour.

Excuse me if this is a dumb question but... if druids get 12% passively from PotP, wouldn’t it be better to just stack to 63% dmg reduction since the cap is 75%? Do they stack?

12% is for all damage
63% is for physical damage. The 12% does not preclude you from getting another 12% physical damage reduction from armour. So in theory, you could have both, but it is not "Additive".

I could be wrong on this (somebody correct me here if necessary, especially order of reduction), but if you are facing a 40k hit, it goes something like this.

40k hit meets Savage Defense (4000 AP, the shield is 1000 points strong) -> 39k hit.

39k - 12% = 34.32k

34.32 - 75% from armour capped -> 8.58k hit.

Compare that to a 63% reduction from armour:
34.32 - 63% -> 12.7k hit

Please note this is purely physical. The OP might be after reducing magical damage.

[Kazeyonoma]

armor is calculated first, then damage reduction, then savage defense/sbv no?

so that same 40k would be something like:

(40,000*.25*.88)-1000= 7800

a 63% reduction from armor would be:
(40k*.37*.88)-1000 = 12024

[Airowird]

armor is calculated first, then damage reduction, then savage defense/sbv no?

so that same 40k would be something like:

(40,000*.25*.88)-1000= 7800

a 63% reduction from armor would be:
(40k*.37*.88)-1000 = 12024

Actually, it's Armor -> Block -> %damage modifier -> Absorb (such as SD)
This is why Paladin/Warrior tanks are so good at anub adds, the damage modifier also boosts the relative value of block.

Example:
10% dmg reduction, 2k shield blocks:
10k post-armor swing > 9k dmg taken
10k swing > 8k after block > 7.2k dmg taken
The block reduced the incoming hit with 1.8k dmg (2k - 10% dmg reduction)
This is why Shield Block is not as effective in combination with other dmg reduction abilities
e.g. 50% reduction:
10k swing > 9k > 4.5k dmg taken
10k swing > 6k (4k from SB) > 5.4k > 2.7k dmg taken which is & 1.8k reduction, not 4k (as the combat log will show)

Note that 50% damage reduction abilities will effectively cut damage taken in half. But you have to consider block as a relative portion of the damage incoming (as intended in Cata) and absorbs as pre-emptive healing buffs.
The end formula for dmg taken = (Post-Armor - block) * dmg modifier - absorbs

[Satorri]

Sorry Satorri, but Thick Hide adds 10% for leather at max rank (as well as cloaks too I believe).

And DKs, Paladins, and Warriors get Toughness, the same 10% armor buff on non-bonus armor, all 4 of which were included in the calculations. And yes, it affects cloaks as well.

It should also be noted that armour denoted as being Green (for example, Frost Emblem Cloak's armour value) DO NOT get multiplied. Instead, the portion of the armour that is normally found on that iLVL of item and armour class (plate/leather/cloth/etc.) get multiplied, and the the static bonus (the part that makes it green) gets added on afterwards.

Also included in the calculations, I'm not sure why you think that needs to be clarified. Maybe I should have been more expressive in all the elements, I was trying to reiterate work I've done a few times before, concisely. =)

Mostly OK, just wish to elaborate that Kings on Bears affects armour pretty significantly, in that 1 AGI -> 2 Armour. A bear running in excess of 1000 Agi gains 200 armor with kings, and realistically in ICC gear you are rocking way more than 1k agility. Also need to note that Strength of the Earth/Horn of Winter is useful to tank survival, it's not just a DPS epeen boost. Agi for armour/dodge, Strength for slight BV to block tanks and parry for DKs, albeit in arguably negligible quantities unless in a min/max discussion.

Also factored on an approximation from summing the agility on the gear, only left off the discrepancies of raid buffed agility which should be small enough to not register a significant % on armor for this comparison.

However, if you truly want spell reduction, Unholy DKs win with Antimagic Shell as well as Antimagic Zone are pretty much kings, beside being about as rare as a dodo bird in endgame content. No other tank has magic resistance quite like them.

The next closest tank is warriors, followed somewhat by Blood DKs, who can spell reflect direct damage back (warriors can do so actively on command with spell reflect, DKs can do so occasionally and passively with Spell Deflection talent). Doesn't work with non-targeted spells (i.e. stop standing in clouds made of fire/poison/shadow/etc, and a dragon's breath weapon is not good for your health)

For consideration, a DK with 20% parry (typical 226+ geared DKs will be in the low 20s depending on weapon runes) would get an average of about 9% magic damage reduction from Spell Deflection. The distinction is that you don't get guaranteed reduction on every hit, but when it does proc it is a huge reduction. (And the "direct" damage application is relative to auras/dots/ground zones, which in unavoidable proportions is usually less than 10% of the total magic damage a tank will take).

AMS vs Spell Reflect is tricky too. AMS always works and is 75% of the damage (100% with Magic Suppression) up to 50% of the tanks health. If we take a very well geared tank with 50k health, that's 25k absorbed if the hit goes that high. AMS is a 1 min CD. Spell Reflect is a 100% reduction in damage whether or not the spell is actually reflected (which I LOVE when it does, the ICC 5s this is silly big damage), but it doesn't always actually reflect things. Spell Reflect has a 10 sec CD. If it works at least to avoid the magic damage, Spell Reflect surpasses it. AMZ is great, and if nothing else it gives you a second slightly smaller AMS, though generally I'd argue for the value being better spent on the group as a whole when you use it well.

Warriors and Paladins also get passive 6% magic reductions at all times through talents to match the passive contribution of Magic Suppression.

One value that is hard to measure but can be terribly significant where it matters most is Acclimation for Frost DKs. In situations where there is consistent enough damage to keep it stacked you can be looking at upwards of 40% magic damage reduction against that school, but if the damage is infrequent the average can skew way down.


And Dracunis, damage reduction %'s do not stack additively, it's a sort of inverse multiplicative.

Like so:
65% reduction from armor
12% reduction from talents
3% reduction from buffs

Becomes:
Total Damage Reduction = (1-0.65)*(1-0.12)*(1-0.03) = 29.876%

Or, they are applied step-wise (same thing just illustrating with real numbers). If you are hit for 60k before mitigation:
60,000 reduced by 65% = 21,000
21,000 reduced by 12% = 18,480
18,480 reduced by 3% = 17,925.6
(17,925.6 / 60,000 = 29.876%)

So the logical extension if you cannot see it is that stacking % damage reductions can never quite fully close the gap to 0% damage taken.

Also, the different effects do not contribute towards the same cap, i.e. Armor is capped independently at 75%, but total reduction is not. In reality, if you factor the contribution of avoidance in total average damage reduction as well, well-geared tanks will be looking at 88-92% total physical damage reduction before even using cooldowns.

Also, by interesting extension, your 50% Divine Protection for Paladins will only actually reduce your physical damage taken by 2-4% from the original unmitigated values, depending on gear, etc. Though it will be a 50% reduction on the numbers you're used to seeing pop up after all your passive mitigation.

[Satorri]

Actually, it's Armor -> Block -> %damage modifier -> Absorb (such as SD)
This is why Paladin/Warrior tanks are so good at anub adds, the damage modifier also boosts the relative value of block.

Are you sure Block comes before non-armor damage reduction? Do you have a link to the testing, I've been working with this stuff for a while but I hadn't see that, and that would be somewhat meaningful in the order of operations.

[MudNova]

Are you sure Block comes before non-armor damage reduction? Do you have a link to the testing, I've been working with this stuff for a while but I hadn't see that, and that would be somewhat meaningful in the order of operations.

It does, the place where I've been paying attention to it is on anub's adds... when I use Divine Protection I take about 2k per hit, instead of 4k

[jere]

Actually, it's Armor -> Block -> %damage modifier -> Absorb (such as SD)
This is why Paladin/Warrior tanks are so good at anub adds, the damage modifier also boosts the relative value of block.

The end formula for dmg taken = (Post-Armor - block) * dmg modifier - absorbs

I'm not sure that is correct. I am pretty sure that, aside from AD which was specifically changed, damage modifiers come before block. I'm have the same reservations as Satorri.

It's hard to nail that down because 50% of 3000 BV is 1500, and boss damage varies much more wildly than that. I don't think it is possible, even using anub adds, to go from 4k to 2k and assume the operations. You either need a fixed damage value to test with or a lot of statistical data.

EDIT: The easiest way to test is find a class that has an ability with a fixed damage value (but is physical in nature and blockable). Test it before and after shieldwall/divine protection.

EDIT2: I just went back and pulled a festergut attempt around the 3 stack area:

[21:39:08.982] Festergut hits Jere 36581 (A: 1760)
[21:39:09.978] Festergut hits Jere 32420 (A: 4190)
[21:39:10.804] Festergut hits Jere 23101 (A: 4483, B: 1966)
[21:39:11.003] Jere gains Divine Protection from Jere
[21:39:11.008] Jere casts Divine Protection
[21:39:11.949] Festergut hits Jere 14393
[21:39:12.686] Festergut hits Jere 16830 (B: 1966)
[21:39:13.725] Festergut hits Jere 14702 (B: 1966)
[21:39:14.711] Festergut hits Jere 17268 (B: 1966)
[21:39:17.591] Festergut hits Jere 14085 (B: 1966)
[21:39:17.730] Festergut Gastric Bloat Jere 4475 (R: 1233)
[21:39:18.508] Festergut hits Jere 13388 (B: 1966)
[21:39:20.370] Festergut hits Jere 12542
[21:39:21.470] Festergut hits Jere 12581 (B: 1966)
[21:39:23.052] Jere's Divine Protection fades
[21:39:23.699] Festergut hits Jere 30804
[21:39:24.506] Festergut hits Jere 27304 (B: 1966)
[21:39:25.581] Festergut hits Jere 29750 (A: 1498, B: 1966)

you can see just from the excerpt that it is hard to nail down which happened:
[21:39:12.686] Festergut hits Jere 16830 (B: 1966)
[21:39:13.725] Festergut hits Jere 14702 (B: 1966)
[21:39:14.711] Festergut hits Jere 17268 (B: 1966)
[21:39:17.591] Festergut hits Jere 14085 (B: 1966)

[Satorri]

And the distinction would be very small. If the unmitigated hit is for 60k, and we have 65% armor, 15% from stance/buffs, 2k block value, and we'll play with 50% Divine Protection on or off.

So, we're all assuming armor is first, that puts the hit at 21k.

If Block comes next, that's 19k, followed by 15% stance reduction it becomes 16.15k (8075 with DP up).

If stance/% reductions come next it's 15.85k(7925 with DP), 300 (150) difference. Like Jere said it is a small enough deviation that it could hide in the variation of swing damage from bosses.

My assumption, and it is an assumption, was that block was applied the same as any other absorb mechanic. It was applied after all the mitigation effects had been factored. It is what let's bubbles be most effective, and I assumed it was designed the same for shields. Why have shield block behave differently when it is essentially the same mechanic?

[Insahnity]

And DKs, Paladins, and Warriors get Toughness, the same 10% armor buff on non-bonus armor, all 4 of which were included in the calculations. And yes, it affects cloaks as well.

Yes, I agree with you there. I just got confused by this statement.

Bears have no bonus armor on leather currently, but they get a multiplier in Bear form to translate leather armor to tank levels (totals a muliplier of 7.014 with meta and talents).

Upon re-reading, I see your rolled it into the statement with the meta gem, but the first part is what stuck on the first read for me.

Next,

Also included in the calculations, I'm not sure why you think that needs to be clarified. Maybe I should have been more expressive in all the elements, I was trying to reiterate work I've done a few times before, concisely. =)

Satorri, you do great work, those in the know understand this, yet we still repeat ourselves/your work, sadly because people don't sift through the material. I just mention this because I had to sit an explain this to a new tank just last week as he was drooling over the ICC cloak. I was trying to persuade him to save for his feral T10 before cloak.

Also factored on an approximation from summing the agility on the gear, only left off the discrepancies of raid buffed agility which should be small enough to not register a significant % on armor for this comparison.

Again, you know this, and so do I, but it helps people at the OP's level. I don't intend to dispute anything you said, except possibly for the thick armour clarification (mostly from that one statement).

For consideration, a DK with 20% parry (typical 226+ geared DKs will be in the low 20s depending on weapon runes) would get an average of about 9% magic damage reduction from Spell Deflection. The distinction is that you don't get guaranteed reduction on every hit, but when it does proc it is a huge reduction. (And the "direct" damage application is relative to auras/dots/ground zones, which in unavoidable proportions is usually less than 10% of the total magic damage a tank will take).

AMS vs Spell Reflect is tricky too. AMS always works and is 75% of the damage (100% with Magic Suppression) up to 50% of the tanks health. If we take a very well geared tank with 50k health, that's 25k absorbed if the hit goes that high. AMS is a 1 min CD. Spell Reflect is a 100% reduction in damage whether or not the spell is actually reflected (which I LOVE when it does, the ICC 5s this is silly big damage), but it doesn't always actually reflect things. Spell Reflect has a 10 sec CD. If it works at least to avoid the magic damage, Spell Reflect surpasses it. AMZ is great, and if nothing else it gives you a second slightly smaller AMS, though generally I'd argue for the value being better spent on the group as a whole when you use it well.

Two things
1) Value for group - Yes, I wholeheartedly agree, which is why I have started to pop Divine Sacrifice more often as a tank, to de-stress the healers. Because WoW is supposed to be fun.
2) I'm don't have any experience with spell deflection. Are there situations where the direct damage is deflected and damages the caster, and others where the talent is procced but no damage to the caster? Can you give some examples, this is interesting! (I'm guessing mobs vs trash)

Warriors and Paladins also get passive 6% magic reductions at all times through talents to match the passive contribution of Magic Suppression.

One value that is hard to measure but can be terribly significant where it matters most is Acclimation for Frost DKs. In situations where there is consistent enough damage to keep it stacked you can be looking at upwards of 40% magic damage reduction against that school, but if the damage is infrequent the average can skew way down.


And Dracunis, damage reduction %'s do not stack additively, it's a sort of inverse multiplicative.

Like so:
65% reduction from armor
12% reduction from talents
3% reduction from buffs

Becomes:
Total Damage Reduction = (1-0.65)*(1-0.12)*(1-0.03) = 29.876%

Or, they are applied step-wise (same thing just illustrating with real numbers). If you are hit for 60k before mitigation:
60,000 reduced by 65% = 21,000
21,000 reduced by 12% = 18,480
18,480 reduced by 3% = 17,925.6
(17,925.6 / 60,000 = 29.876%)

So the logical extension if you cannot see it is that stacking % damage reductions can never quite fully close the gap to 0% damage taken.

Also, the different effects do not contribute towards the same cap, i.e. Armor is capped independently at 75%, but total reduction is not. In reality, if you factor the contribution of avoidance in total average damage reduction as well, well-geared tanks will be looking at 88-92% total physical damage reduction before even using cooldowns.

Also, by interesting extension, your 50% Divine Protection for Paladins will only actually reduce your physical damage taken by 2-4% from the original unmitigated values, depending on gear, etc. Though it will be a 50% reduction on the numbers you're used to seeing pop up after all your passive mitigation.

And finally..

And the distinction would be very small. If the unmitigated hit is for 60k, and we have 65% armor, 15% from stance/buffs, 2k block value, and we'll play with 50% Divine Protection on or off.

So, we're all assuming armor is first, that puts the hit at 21k.

If Block comes next, that's 19k, followed by 15% stance reduction it becomes 16.15k (8075 with DP up).

If stance/% reductions come next it's 15.85k(7925 with DP), 300 (150) difference. Like Jere said it is a small enough deviation that it could hide in the variation of swing damage from bosses.

My assumption, and it is an assumption, was that block was applied the same as any other absorb mechanic. It was applied after all the mitigation effects had been factored. It is what let's bubbles be most effective, and I assumed it was designed the same for shields. Why have shield block behave differently when it is essentially the same mechanic?

Mathematically speaking, unless the fixed amount absorbs/mitigation preceed all (Block,Savage Defense/ SacredShield), the significance of the relative order won't be that huge. In one of the numbers listed above, the difference was 12.7k vs 12.2k, I wouldn't quibble over the 500 damage with a proper healing profile, especially if we are not talking ToC Damage profiles. It does however illustrate (roughly) the idea.

[Satorri]

I don't have any experience with spell deflection. Are there situations where the direct damage is deflected and damages the caster, and others where the talent is procced but no damage to the caster? Can you give some examples, this is interesting! (I'm guessing mobs vs trash)

You can think of it, as the name suggests, as the ability to parry spells, but in a more traditional deflection sense. Rather than negating the damage it cuts it in half (45%). So, imagine you are tanking a dragon, you have a chance equal to your parry chance to take only 55% of a breath hit.

I'm currently sitting at about 26% parry raid buffed. So if I had Spell Deflection against a dragon with, say a 43k frost breath (coughSindragosacough), 26% of the breaths would be reduced by 45% to only 23,650. The average reduction in that case would be 11.7% reduced damage taken from the breath from that talent alone. It's not bad, and obviously 3 in 4 will still get through, but the one that doesn't hit in phase 3 will be a game changer.

[Airowird]

It does, the place where I've been paying attention to it is on anub's adds... when I use Divine Protection I take about 2k per hit, instead of 4k

It's tricky in that it works on the debuffs on you that increase damage taken (and last time I checked this is consistent, which means also on buffs that reduce damage taken) but it does not work on e.g. Saurfang, who has a buff to increase damage done.

[Insahnity]

You can think of it, as the name suggests, as the ability to parry spells, but in a more traditional deflection sense. Rather than negating the damage it cuts it in half (45%). So, imagine you are tanking a dragon, you have a chance equal to your parry chance to take only 55% of a breath hit.

I'm currently sitting at about 26% parry raid buffed. So if I had Spell Deflection against a dragon with, say a 43k frost breath (coughSindragosacough), 26% of the breaths would be reduced by 45% to only 23,650. The average reduction in that case would be 11.7% reduced damage taken from the breath from that talent alone. It's not bad, and obviously 3 in 4 will still get through, but the one that doesn't hit in phase 3 will be a game changer.

Ok now you have me confused, partly because I am still in the warrior spell reflect mindset. Warrior Spell reflect only works on targetted single target spells, so for the aforementioned breath weapon, a warrior is SOL. If you are suggesting that a DK can parry away area of effect spells (classified as either conical or plain old Puddle of Acid/Fire/Shadow), this would be fascinating.

I am still trying to grapple general resistances. I've been tinkering with frost resist gear for VoA25 on Toravon on my arcane mage, and I am finding the following

with 426 resist, Recount reports

1/3 25% partial resist
2/3 50% partial resist
One instance of a 75% partial resist.

With 466 resist, recount reports

Almost exclusively 50% resists
3 instances of 75% resists in one fight
5 instances of 25% resists in the same fight.

I'm guessing with 450ish resistance, you can expect a normal distribution around 50% partial resists, with +3s being 75% resist and -3s being 25% resists. I'm anxious to try its effects with polar gear on my bear where it is most effective.

Finally, Spell Deflection sounds like the old avoidance argument; in the long run statistically speaking, it's as good as mitigation, but unless RNG loves you, that killer spell ain't gonna be deflected when your health is redlining and your healers are dead, and there ain't nothing you can do about it.

[Satorri]

It will not work on damage zones as they count as "indirect damage" along with auras, actual debuff dots, and that sort of thing.

Dragon Breath/Cone attacks do constitute "direct damage" spells though and are subject to the effect.

And yes, it follows the rules of chance. When it happens it is huge, but it won't happen every time. That's not really a reason to disparage it of course.