Shield Block vs. Shield Barrier: The Numbers

[Breccia]

This is an exceptionally long post filled with math. There is no TL:DR version.

Introduction: Everyone who has looked at warrior tank mechanics in the Beta knows that they have Shield Block, and Shield Barrier. It has been said many times that Shield Block will be better on harder-hitting bosses, and Shield Barrier on weaker. So what is the dividing line? When is, in fact, Shield Block better? How hard is "hard-hitting" in this context?

I started with some warrior stats similar to mine in the Beta. Yes, my warrior is still 85, and for this comparison, that's good, because I can compare my results to currently existing content that has been in the game for months. This means I'm using a dodge/parry/miss combined chance of 40%, a block chance of 30%, a critical block chance of 70%, and no unusual spells, effects, or items. I am assuming there is no clipping of durations of either Shield Block or Shield Barrier. I'm also using a boss swing timer of 1.5 seconds and a maximum Shield Barrier value of 40,000 absorption.

Now: Shield BLOCK can be used twice per 15 seconds, and it lists having a duration of six seconds. That means, at the maximum, you can have Shield BLOCK active for 12 out of every 15 seconds. I will now call that "Max" Shield Block uptime. I will also designate a "Standard" Shield Block uptime of 6 seconds out of every 15. To keep things fair, I will also assume the same amount of uptime/usage for Shield BARRIER, i.e. either two uses of Shield Barrier at "Max" or one use at "Standard" settings, per 15 seconds.

Block is now on a "second roll" system which makes it happen a lot less often than Live, especially when combined with the much lower Block chances on the Beta as well. If this warrior with these stats were attacked 10,000 times, then of those 10k times, 4000 of them would miss, be dodged, or be parried. Of the remaining 6000 attacks, 30% of them, or 1800, would be Blocked, and of those 1800, 70% of them, or 1260, would be Critically Blocked. The remaining 4200 attacks would be normal hits. If each attack did D damage on a full-strength hit, this becomes
(4000x0)+(1260x0.40D)+(540x0.70D)+(4200xD) / 10,000 = 0.5082 D
Without using Shield Block, each swing does, on average, 50.82% damage.

Now if you then use Shield Block, during its duration, every single attack that does not miss entirely becomes Blocked. 4000 attacks still miss, but of the remaining 6000 attacks, 70%, or 4200, are Critical Blocks, and the other 1800 are Blocks. This means
(4000x0)+(4200x0.40D)+(1800x0.70D) / 10,000 = 0.2940 D
While Shield Block is up, the damage per swing averages to 29.40% of its normal "hit" value.

If we went with the "Standard" Shield Block uptime, then this would be the following averaged result over 15 seconds:
(9 x 0.5082D) + (6 x 0.2940D) / 15 = 0.42252D overall average over 15 seconds

Now it's already been accepted that stronger hits (high values of D) make Shield Block better, while weaker hits (low values of D) make Shield Barrier better. Where is the break-even point? Well, with 10 boss attacks in that time, and with one use of Shield Barrier absorbing 40,000 damage in the same span, we have
10x0.42252D = 10x0.5082D - 40,000
D = 46,685.3 damage, per hit

Let's put this value in perspective. On a Morchok 10H run, with the current 25% nerf, Morchok is hitting most DKs for about 40 to 45 thousand damage, INCLUDING the fact that he's using a Stomp debuff on the tanks that causes them to take +50% physical damage, about one-third of the time. Because warriors use shields, their armor is a little higher, and take a little less on a "hit". D, on the other hand, is HIGHER than this value. So, this means we are talking about a boss that is currently hitting harder than even a 10H boss while you have a significant damage taken debuff on you often.

With me so far? Good. Now let's crank this fight up to "Max" Shield Block uptime.
(3 x 0.5082D) + (12 x 0.2940D) / 15 = 0.33684D overall average over 15 seconds spamming Shield BLOCK
10x0.33684D = 10x0.5082D - 80,000
D = 46,685.3 identical. In other words, Rage income seems to have no bearing on the Shield Block vs. Shield Barrier debate SO FAR. We'll come back to that topic later.

Now so far, this looks bad for Shield Block. But it gets worse. Here are my real stats on the Beta: 41.64% dodge/parry/miss against a raid boss, block chance 24.62% against a raid boss, and critical block chance 61.70%. These changes are relatively minor, true. However, I was able to generate enough Vengeance to get a 66,000 Shield Barrier by fighting level 80 trash monsters, and in a raid situation, my Attack Power couldn't be any lower. Following the same template as last time, we see:
-- without using Shield Block, the average damage is now 0.51388D
-- while Shield Block is active, the average damage is now 0.30049D
-- using "Standard" Shield Block uptime, the damage over 15 seconds averages to 0.42852D. Compared to a single 66,000 Shield Barrier, this gives
10x0.42852D = 10x0.51388D - 66,000

D = 77,319.6 damage. I will remind you that this is based on my level 85 stats. This much damage will three-shot just about any tank in the game, and those that survive are barely standing. Heroic Morchok 25 is doing about 70,000 average damage per hit these days, including the aforementioned +50% damage Stomp debuff uptime. What I am saying, is at this time, pretty much the only bosses capable of doing this kind of pure damage per swing, are the 25-man Heroic DS bosses with the 25% nerf removed. This is not just "hard-hitting". This is the top of the top, the things that not even 1% of the entire playerbase does. I'm not a PvP expert, but I'm willing to bet that 77k melee hits don't happen very often in PvP, and if they do, then I'm glad I don't participate.

-- using "Max" Shield Block uptime, the damage over 15 seconds averages to 0.34317D
10x0.34317D = 10x0.51388D - 132,000
D = 77,324.1 damage, difference explained by rounding. Once again, Rage income seems to have basically zero effect on the Shield Block vs. Shield Barrier dispute SO FAR.

It is also worth noting that, in both cases, the value of D was just a little over the value of the Shield Barrier. Think about a time when a tank is in trouble, is taking a lot of damage and needs help FAST. In such cases, popping Shield BARRIER will prove a smarter move, as even if the next attack is a completely unblocked hit, a very large percent (over 85%) is absorbed, and the attack can still be blocked or critically blocked. Popping Shield BLOCK, on the other hand, would at best reduce the damage by 60% assuming the attack is critically blocked, which is not 100% of the time, and there is no backup afterwards. If the tank is low on health, there is no question.

Here are a list of things that make using Shield BARRIER and even smarter idea:
-- any bonuses to Attack Power you can get, especially Vengeance, raises the Shield Barrier value and, therefore, raises the value of D. You could expect to see a large number of warrior tanks using Battle Shout because of this. Worse, you can expect to see warrior tanks USING PVP/DPS GEAR ON PURPOSE and claiming it helps them survive, and sadly, they'll be right.
-- any bonuses to armor. Armor lowers all incoming physical damage, and reducing the damage of the blow makes Shield Barrier a better move. There aren't a lot of bonus armor effects or items, and most tanks don't seem to think highly of them, but technically, they make Shield Barrier better.
-- any unblockable physical attacks, such as Heroic Morchok's Stomp attack. Most tanks in most Morchok fights take more damage from Stomp than from melee attacks, based on the logs I was able to browse. Other common examples include Blackhorn's roar and the Madness Impales.
-- any form of magical attacks, which cannot be blocked at all, make Shield Barrier a MUCH smarter choice.
-- any "chance on hit" effect that applies on a blocked attack, but not on an absorbed attack. Nothing important leaps to mind in this category, however.
-- healing debuffs make Shield Barrier more attractive, because while Shield Block causes you to take less damage, Shield Barrier causes you to take none at all, buying your healer more time. The only exception is a healing debuff coupled with really high-damage attacks, also known as "instant death".
-- high-rage fights. Shield Barrier has no cooldown, nor does it have a lockout on its maximum uptime, which Shield Block has. Shield Barrier can also be used with under 60 rage, allowing you to top up nearly at will, which Shield Block can't do, either. Rage bursts caused by lucky procs and shouting can always be dumped into Shield Barrier, but can't always be dumped into Shield Block.
-- higher avoidance levels reduce the number of times an attack could be blocked at all, which not only reduces the value of Shield Block but of mastery as well. For instance, if you use the first set of stats mentioned, but replace the 40% total avoidance with 50%, then D jumps to 56,022, roughly a 20% increase in the amount of damage per hit needed for Shield Block to remain a good idea. This is roughly the amount of damage Heroic Morchok 10 was doing without the nerf in place.

Here are some things that make Shield BLOCK a smarter move:
-- bosses whose melee attacks have a "chance on hit" effect that fails on a block. I honestly can't think of a lot of those off the top of my head, maybe Blackhorn's Devastate or the bloods on Cho'gall.
-- fights with INFINITE Rage mean you can use both Shield Block AND Shield Barrier. However, at this time, Rage is not generated from being attacked, so barring odd mechanics like Vael, you won't see this, either.
-- high values of mastery make blocking better, because that's simply what mastery does. However, they'd have to be REALLY high. The honest stats I mentioned above involve me stacking mastery fairly heavily already. Eating a +90 mastery food in the Beta caused my Block chance to increase by 0.25% and my Critical Block chance to increase by 1.11%. Assuming both of these figures are linear (which, because Block has diminishing returns, they can't be), if I somehow got my hands on +500 extra mastery at level 85, then the value of D would drop from 77,319.6 to 75,862.1 which is a drop of about two percent. And this is the equivalent of ten whole epic gems ALL stacking mastery! By contrast, +500 strength would increase my maximum Shield Barrier by 2,500 (based on the scaling I was able to find), which in turn would change D from 77,319.6 to 80,253.1, a change just shy of four percent, i.e. almost double the effect in the other direction.

Conclusion: By the current mechanics, using currently existing raid models as a comparison guide, the only players who would ever want to use Shield Block are those who are fighting the very, very hardest of encounters. The vast majority of all warrior tanks, by current content, will never fight a boss capable of doing the kind of damage that is needed for Shield Block to outshine Shield Barrier. Therefore, almost all warrior tanks will spam Shield Barrier at all times and never use Shield Block, regardless of the context. Worse, the mechanics also not just allow, but encourage the same tanks to gear, gem, and enchant for Attack Power rather than defensive stats or mastery.

In other words, the current situation of Shield Barrier is teaching warrior tanks to be bad.

Breccia

[Deleted]

Well... the formula I was thinking about to compare Shield Block and Shield Barrier was:

Melee hit = Absorb / [((1-%block)*(0.7*(1-%crit_block)+0.4*(%crit_block))*Number of hits on 6 seconds*(1-%parry-%dodge)]

I would disregard the fact that you can only get SB 12 seconds out of 15. You only have to think, what do I want for 60 rage in the next 6 seconds, 100% assured block or an absorb shield? Let's ignore the case where you use a 20 rage Barrier instead of a 60 one.

With the numbers of your tank a 1.5 swinging boss will need to hit for 42511 pure damage or more to make Shield Block better than Barrier for pure melee in those 6 secs. Even tier 11H bosses were hitting that hard (I wouldn't use a 25% nerfed Morchok... he never was a heavy melee hitter to begin with, due to that 100% physical debuff he leaves). The moment magic damage/unblockable damage/boss stops to cast something and don't melee, there is a thinking process behind of what to choose, it's not obvious which one to pick (Ie, let's say the boss melees for 60k. How much magic damage will you have to receive to make shield barrier better on those 6 secs? Let's take Ragnaros, even if he don't melee that hard either (since he leaves a dot too on the tank). How many stacks of Burning Wound do I need to have on me to switch Block to Barrier? It is worth it to use Block if he is going to stop to cast Sulfuras Smash during those 6 secs and melee 3 times instead of 4?)

What if you are about to die, as you say? What about the increased rage of shield block? Will you need to watch your vengeance before hitting barrier?

We would see how it fares on level 90, but for me, it's more or less on the sweet spot right now, where you will have to decide on each moment.

[Satori]

You're completely ignoring the effect that Shield Block has on rage generation through Enrage uptime.

[Deleted]

And I personally forgot that each hit parried/dodged males Barrier more valuable than S Block. Too late for me. Will edit.

[Breccia]

You're completely ignoring the effect that Shield Block has on rage generation through Enrage uptime.

If this was aimed at me, then it's been discussed. Rage income has effectively no effect on whether Shield Block or Shield Barrier is better until you get more than 120 Rage per 15 seconds. At that point, Shield Barrier wins even more because you can burn the excess. Below that, the numbers are unaffected.

[Deleted]

I kind of fail to see the relevance of using lvl 85 numbers on beta when most, if not all, numbers are "work in progress".
If at least it were lvl 90, it might be worth discussion, but it barely registers as it is.

[Breccia]

I kind of fail to see the relevance of using lvl 85 numbers on beta when most, if not all, numbers are "work in progress".
If at least it were lvl 90, it might be worth discussion, but it barely registers as it is.

Because any issues with level 90 raid bosses might not be issues with the abilities, but the bosses. Cataclysm is done being coded, it makes a reliable control group. In addition, it puts the "hard-hitting bosses" in better perspective if people actually know the bosses you're talking about.

[Satori]

If this was aimed at me, then it's been discussed. Rage income has effectively no effect on whether Shield Block or Shield Barrier is better until you get more than 120 Rage per 15 seconds. At that point, Shield Barrier wins even more because you can burn the excess. Below that, the numbers are unaffected.

If I understand your text correctly, it wasn't discussed:

With me so far? Good. Now let's crank this fight up to "Max" Shield Block uptime.
(3 x 0.5082D) + (12 x 0.2940D) / 15 = 0.33684D overall average over 15 seconds spamming Shield BLOCK
10x0.33684D = 10x0.5082D - 80,000
D = 46,685.3 identical. In other words, Rage income seems to have no bearing on the Shield Block vs. Shield Barrier debate SO FAR. We'll come back to that topic later.

You're comparing the breakpoint of max Shield Block uptime to max Shield Barrier uptime. What I'm saying is that, by using Shield Block you're closer to getting to max Shield Block uptime whereas using Shield Barrier will never allow you to get to the maximum uptime. This is because Shield Block greatly increases your chance of getting a Critical Block and thus an Enrage (from a 16% chance to a 43% chance for my character) and a prot Warrior rage generation is fixed, unnaffected by haste or any sort of outside effect.

Now, I'm lazy and not smart enough to have the actual numbers, but for the sake of argument let's take your equation but instead of making the rage expenditure equal between Shield Barrier and Shield Block (which ignores that Shield Block increases rage generation, whereas Shield Barrier doesn't), we'll assume a 50% increase in rage income (as if 100% enrage uptime was achievable):

3.3684D = 5.082D - 60,000
D = 39.768

On top of that, while you can appreciate the time you put on this, your 85 testing has very little meaning. Don't take this wrong, but because you're not 90 your numbers are way off. For instance, Galleon, a world boss (so that means he's pretty weak) hits for 60k per swing. Your avoidance and mastery ratings will take a huge hit at 90, further increasing the value of shield block (for reference, in the pvp gearset with no reforges or enchants, my character has 5% dodge, 14.49% parry, 18.89% block and 25% critical block).

[Shamgar]

The funny thing is that if they make all our rage dependent on blocking, we will be completely useless when we can't block. And you can bet your ass they will make more bosses that bypasses block for whole or parts of the encounter.

[Daetur]

Did you post this in the main Beta Class Balance Feedback thread on the Blizzard forums? If not, you should. If you are unable to, someone should for you.

[Satori]

The funny thing is that if they make all our rage dependent on blocking, we will be completely useless when we can't block. And you can bet your ass they will make more bosses that bypasses block for whole or parts of the encounter.

That's an extremely good point. It's already really bad in rbgs and situations where you can't make contact with the opponent.

[Breccia]

OK Satori, you made some relevant points. Again, I left out level 90 testing, mostly because the damage of level 90 bosses is nowhere near set in stone, and nobody has T16 gear to show the full extent of the code. DS, however, is something everyone and their alt is comfortable with by now and makes much clearer examples. I could use example stats, such as yours, and re-run the numbers at 90, but I'm not sure what to compare them to. At the very minimum I'll need info about a maximum Vengeance 60-rage Shield Barrier the example warrior can possibly generate. I can almost guarantee, however, that a level 90 warrior in a raid group will easily have the AP to Shield Barrier that 60k hit you were talking about, considering my level 85 can do that now solo.

As for the increase Rage caused by blocking, I admit, it could be possible that Shield Blocking gives more than Shield Barriering, but I'm not sure by how much. Using your stats, your extremely small 25% crit block is just not going to generate full Enrage uptime. I'll also have to include Berserker Rage, because you know for a fact everyone's going to macro that into everything. This will very likely lower the value of D, but I expect to see a bigger change with my much higher critical block than yours, which doesn't help Shield Block that much. I'll see what I can find.

I would also like to add that, unless I'm misreading the tooltip, Revenge and its Rage generation only procs on a Dodge or Parry. This won't help a lot in PvP, although dodging ranged attacks might help some.

[kopcap]

Sorry, but 46k swings were last relevant in t11. Running numbers on 25% nerfed content tuned for 391 gear is more or less a waste of time.

ps

10x0.42252D = 10x0.5082D - 40,000

This theory crafting gem is sticky-worthy.

[Breccia]

Running numbers on 25% nerfed content tuned for 391 gear is more or less a waste of time.

First of all, I was running numbers on warrior stats. The results were compared to current content to put them in perspective.

Second of all, I was pretty specific about saying "What I am saying, is at this time, pretty much the only bosses capable of doing this kind of pure damage per swing, are the 25-man Heroic DS bosses with the 25% nerf removed." content which is NOT tuned for 391 gear.

Thirdly, these changes will hit live before MoP's release, forcing every end-game warrior tank to use the new mechanics, making them quite relevant, if only for a few weeks. Existing warrior tanks still interested in raiding and/or re-learning their class will be testing their abilities, and researching them online as well.

Finally, every single warrior tank that wants to get to MoP raiding will, at some point, run content from level 81 (when they first get Shield Barrier) to 85 (where these numbers are based) on their way to 90. In theory, they should be learning how to play their class in that time. These levels SHOULD remain relevant to the game, unless you really want everyone in MoP's 86-90 range to run around like headless chickens, with no ability or desire to play correctly. I realize it's all fashionable and meme and stuff to say "the game begins at 90" but do you really want to be the bad tank everyone complains about from 10-89? I know I don't. Most warrior tanks who read these forums probably don't, either.

[kopcap]

Perhaps you should rename this thread to "Shield Block vs. Shield Barrier: 5man and Cata Numbers", rethink your math and reword the bottom third of your post to avoid confusing the shit out of any inexperienced tank who is trying to catch up with the new mechanics.

[Breccia]

I was finally able to get a few more calculations going involving both the Enrage effect, and also higher-level stats. Here are my updated results:

Assuming a warrior did the following :
-- started every fight by using Battle Shout, then Charging
-- uses Berserker Rage on cooldown, unless already enraged
-- uses either Shield Block or Shield Barrier, depending on which one is being tested, at 60 rage
-- Uses the attack priority Shield Slam, then Revenge, then Thunderclap if expired, then Battle Shout when ready AND rage less than sixty, and finally Devastate if nothing else is available
-- No other unusual buffs, debuffs, abilities or items
-- Each fight lasts 10 minutes.
-- The warrior engages in 1,000 such fights.

We'll start with my stats at 85 to keep things consistant. The version that spammed Shield Block was able to maintain a 77.605% uptime averaged over the one thousand fights, and a Rage generation of 7.8195 rps. Meanwhile, the version that spammed Shield Barrier had a Rage generation of 7.0865 rps. This is about a 10% difference in Rage by spamming Shield Block instead of Shield Barrier, which is definitely noticeable, but not overwhelming. It also leads to 1.7716 full-strength Shield Barriers per 15 second period.

Following the new Shield Block uptime, and the new number of Shield Barriers that follows, the resulting equations become
(0.77605*0.30049D)+(0.22395*0.51388D)=0.34828D average damage taken while Shield Block is up
10x0.34828D = 10x0.51388D - 1.7716*66,000
D = 70,607

This new value of D, taking into account Enrage from critical blocks, is about 10% lower than the previous value, lowering the threshold between what damage is high enough that Shield Block becomes the smarter move. As noted before, heavy melee hitters in Dragon Soul 25H are doing about this damage now, so any change this would make to your strategy on whether to use Shield Block of Shield Barrier, in this context, would be minimal at best, especially your only option with the 25% nerf (oh sorry, 30% nerf) are "on" and "off".

BUT.

Everyone wants to talk about level 90 content, despite of the overall lack of information or the raid boss damage/abilities being far from finalized. Fine. We can do that. I justed need some information that Zej, on the Beta forums on Blizzard's site, was happy to provide. Whoever and wherever you are, Zej, thank you! You made my job a LOT easier. Zej posted the following stats:
-336,187 Health
-42,759 Armor (48.03% pDM)
-8.11% Dodge Chance
-12.08% Parry Chance
-19.17% Block Chance
-27.05% Critical Block Chance
-13,606 Attack Power (No Vengeance)
Incidentally, this means his Shield Barrier, with a full stack of Vengeance, will absorb a staggering 118,000 damage with 60 Rage.

There are a ton of changes here, mostly involving that ALL of these avoidance/mastery abilities are a LOT lower than mine. This is fair. I'm at the end of one expansion's final tier, he's barely started the next one. His values will be low, but they should be appropriate for entry-level raiding based on the ilvl he posted.

I re-ran all of my calculations using his numbers, including the Enrage uptime variant, for both Shield Block and Shield Barrier spamming. I noticed a few things while doing so:
-- The difference in Rage generation, between using Shield Block and Shield Barrier, was again almost exactly 10% higher rage per second when using Shield Block, despite the massive changes in stats.
-- This "hypothetical Zej" was able to maintain 56.3143% uptime on Shield Block or 1.4625 Shield Barriers per 15 seconds. Both of these values are lower than mine, due again to his much lower stats and, therefore, much smaller chance to proc Revenge and get a critical block for Enrage.
-- And finally, the value of D turned out to be 130,383 damage, after armor and other reductions of damage, on a full-strength hit.

A few things to note about this new number:
--First, like the number I got for myself using current content, this is just barely over one-third of his health, meaning that this, too, refers to a monster capable of three-shotting the tank. Just as with current stats and current content, this refers to a Lich King-era boss that destroys tanks in under 5 seconds without healing, a model that Blizzard has said multiple times they're not really interested in.
--Secondly, I had some difficulty finding how much melee damage a boss does. Satori, at mmo-champion, did mention a world boss Galleon that does about 60k damage per hit. Thanks Satori! But this is over TWICE as high as the attack you listed.

To put this number into better perspective: the boss fight Will of the Emperor, according to wowpedia.org, has two major abilities that do physical damage. Both are labeled with a red skull mark, calling these abilities out as "Deadly". They each do 300,000 physical damage. I am not sure that the boss will do the same damage on a melee swing, and I would guess it could not. After all, this "deadly" damage comes from two avoidable zone attacks, it is supposed to be a punishment for standing in the wrong spot. However, purely for the sake of argument, let's pretend that this particular boss is so VERY strong that it does the same damage using melee attacks as a "stand in the fire instakill" megasmash, and use that 300,000 damage. After the 10% reduction from thunderclap, and the effect of Zej's armor, this damage will be reduced to about 141,000 damage, just barely over the amount D found for Zej. And, to make it clear, I do NOT think that bosses in T14 are actually going to hit that hard! Reducing this absolutely ludicrous damage by even a mere 10% means that, once again, Shield Barrier wins.

This means, once again, that EVEN including Enrage uptime, EVEN taking into account level 90 stats, that the ONLY time it would be worth using Shield Block over Shield Barrier is when you're dealing with EXTREMELY high amounts of damage. Not just "heroic dungeon" high, but "you took this kind of damage because you did something stupid and now you must pay" high.

[kopcap]

Do you realize that each Chogall add with one stack on the tank was hitting for nearly 90k 18 months ago? That Zonoss was smacking for up to 180k in December? That even a timid boss like Ultraxion still hits for 100k even with current nerfs even on ten man?

[Deleted]

If you want to put it in perspective, right now bosses like Protectors or Tsulong (10 Normal) melee for up to 110k. Take that value for what it is (considering low numbers+high gear+normal difficulty). I would save the model, Breccia, and judge the value of Barrier vs Block once the actual numbers are on game. After all, you have done the hard part already.

[Breccia]

Do you realize that each Chogall add with one stack on the tank was hitting for nearly 90k 18 months ago? That Zonoss was smacking for up to 180k in December? That even a timid boss like Ultraxion still hits for 100k even with current nerfs even on ten man?

Hold on. I'm not sure you can back that last one up. With current nerfs, he's hitting about 40-45k in normal and 65k in heroics in 10man. Grab a worldoflogs at random and see for yourself. Remove the current 25% nerf and that doesn't even let him break 100k. Oh sure, his magical attacks might do that much, but you can't block those so who cares?

By the same token, the random logs I'm browsing suggest that Zon'ozz 10H does about 40k on an unmodified melee attack with the current nerfs. Again, grab a log at random and see for yourself, look for the first melee hit in the fight. Remove the 25% nerf (because we are talking about December), pretend that 25H is 50% more damaging, and for him to do 180k on a swing, he'd have to be at six stacks or higher. In other words, an un-nerfed boss, carrying around a big damage buff (which he will not have for very long), on the hardest possible difficulty. Since my main point was that Shield Barrier is better for all but the hardest of the hard bosses, I don't think this is the best example to use here.

I'll be honest, I don't have any numbers for Cho'gall or his adds offhand. I don't remember the adds hitting that hard, and I was the add tank for my raid groups for both normal and heroic, but maybe that's because I am a warrior and blocked just about everything. wowpedia claims that their big nasty interruptable spell does under 80k on 25H mode, so I guess maybe with a debuff stack from Cho'gall the melee might do 90k on a hit at that difficulty before the 20% nerf (18 months ago), but that makes him again, one of the hardest bosses, at the hardest difficulties, without the nerf.

If you want to put it in perspective, right now bosses like Protectors or Tsulong (10 Normal) melee for up to 110k. Take that value for what it is (considering low numbers+high gear+normal difficulty). I would save the model, Breccia, and judge the value of Barrier vs Block once the actual numbers are on game. After all, you have done the hard part already.

Now THIS is information I can actually use. The 110k is lower than the D discussed above, but a lot of people (including you) seem to think this damage is low. Should they bring it up a bit (or if people go after Heroic mode), and if people go at it with lower gear than Zej, then yes, Shield Block vs. Shield Barrier becomes an actual choice again.

The model's not going anywhere, and it will be easy to adjust if, say, Blizzard lowers the AP scaling or lets players block for 35% or whatever.

[Deleted]

After reading all of what you wrote Breccia, I still don't think our shielding abilities needs any tweaking. In fact, you just convinced me they don't, as I still saw SBar to be absorbing too little damage for a whole 60 Rage package.

What you should know is bosses are gonna hit for way more than 100k damage in MoP Raids once it goes live, even in normal mode, and mostly twice that much, 200K on heroic.

Why ? simple. Because Tanks when full buffed are gonna be at around 400/450K health more or less.
With that amount of health, if bosses don't hit at least that hard on heroic, they are simple gonna be too trivial to tank. This isn't an assumption but, rather an observation based on previous content.

Let's take DS HC (before nerf). On average bosses like, Morchok, Zon'ozz, Ultraxion, Blackhorn, Madness, meleed for something between 70/100K (85k on average thus), after armor/buffs/debuffs DR, but before blocking (which is equal to the numbers you saw on WOG after blocking).

On average this is almost half of the tank's health since he had around 200/220K back then, making him being 2 shotable if he wasn't able to block everything and always reduce that 70/100k damage by at least an extra 30%, and mostly 60%.

This will be the same thing on MoP, but since, we will be at around 400/450K health bosses will be hitting for 140/200K on HC, after armor/buffs/debuffs DR, in order to keep tanking and tank healing interesting.

Now since, we aren't CTC caped anymore, those hits are gonna fully hit us most of the time when Shield Block wouldn't be active. It would then take a simple bad luck to get 2 shot in 2 seconds.
And when that scenario is even possible, RNG always makes it that it happens more frequently than you'd expect it to (don't ask me why, it's a simple observation, why else would you want to be CTC caped and not sit at around 95% of the capping?).

So lets say you decide to go all in on Shield Barrier. and can even increase the absorb to some 140/200K with a full 60 rage shield (which is very generous). You'd be able to absorb 1 full swing, so according to your calculation, 1 swing every 9 seconds more or less at 7 RPS.

What then ? How are you possibly gonna survive during the remaining 8 seconds (5 melee swings) when you'll be waiting for your Sbar to be usable again ? You can't possibly cross your fingers all the time and hope to dodge most hits so you don't get 2 shot, can you ?

And there is where Shield Block gets interesting :

Shield Block allows you to reduce damage spikes to an extent that you wouldn't be 2 shotable anymore, even though Shield Barrier on the paper reduces the same amount of damage (or even slightly more).
It does this by allowing you to CONTINUOUSLY reduce the incoming damage, and not totally absorb the first hit and let you unprotected for the remaining 5 attacks.
It will reduce each attack by at least 30% (and mostly 60%), for the whole 6 seconds duration, and healing you would then become more predictable and less of a gamble. Sure, you can't keep it up 100% of the time, but you can keep it up so often (at least some 60/70% of the time unless you are doing it all wrong) that when it wouldn't be up, the boss simply wouldn't have the time to 2 shot you, or in the worse of the worse case scenario you'd simply need to pop a lighter Shield Barrier, or even a CD, to completely eliminate any unmanageable combo.
You would never be at 100% HP a second then in the next 2, drop to 10% HP, but that's exactly what will happen if you where to go all in Shield Barrier on everything.

As a tank and a healer, I can tell you for a fact that never I have to blindly spam heal a tank (unless he's a pretty bad one), because if I did I'd be oom in a matter of seconds, not minutes. A good tank (whether he's even aware of it or not) will always be trying to reduce damage continuously and smoothly, rather than in a manner that would inevitably create damage spikes. I think it's clear that tanks hate seeing their HP drop to nothing in a matter of seconds.