I specced into every other talent that increases my survivability by a little bit and the prot tree is actually totally bloated anyways. I'd have to skip either thunderstruck, piercing howl, war academy or gag order and i really like all of them.

But my real concern about it: i don't hit devastate that often. Between revenge proccs, shockwave, heroic throw and concussion blow there just not much time. And devastate hits really, really soft. I use Cookies Tenderizer (2.8 speed) and it averages on 6600 damage for me, only 1.5 times higher than my average melee hit. And with actual gear dps matters, on fights with high vengeance uptime (eg omnitron council) you easily ends up bypassing most of your dps while on kick duty and moving those bosses around. Having a 8.000 hitpoints heal (at 160.000 hp raidbuffed) every so often for swapping to a pure devastate spam in execute range and investing 2 talent points looks pretty underwhelming to me. Without the 5% restriction i'd love to take it.

...

Pinnick you have seen a few heals around 55k is this from the Victory Rush heal criting?

Based off of Myul's 160k hp total from raid buffs (this total could vary from tank to tank), Im only seeing the total heal from a spec'd and glyph'd Impending Victory build which includes field dressing being around 19k (non-crit of course) if it does crit then we are looking at a 38k heal, which as you stated is "pretty significant"

Which would make and Impending Victory build a viable spec at least early on to improve survivability.

I'm seeing 55k heals as well with Victory Rush, but only the regular one after you kil a mob that gives exp (Victory Rush is glyphed). I think Pinnick was talking about regular VR, there's no way you can heal 1/3 of your life with Impending Victory.

My brief experience in raids suggests that Impending Victory is a no brainer for now, especially since they fixed paladin mana regen. When you glyph Victory Rush, it heals for around 10%. I have roughly 160k health, I'm seeing between 15k and 18k heals. Haven't seen a crit yet though.

On a side note, I'm feeling Vigilance is pretty useless. Vengeance goes up really fast without it, it doesn't offer 3% dmg reduction to the other tank, and I haven't see fights where you need to taunt all the time. I'm probably keeping it in the back of my head for adds heavy fights, but for now it seems like a wasted point, even in a raiding situation.

...

Pretty new here, but I figured I'd add a little.

I just tried out spec'ing for Impending Victory prior to our raid tonight, and I seemed to like it. I decided to give up 1 point in Thunderstruck and Vigilance for it... I haven't yet found Vigilance to be terribly useful. During our 9 minute long Omnitron kill, our priest healer died towards the end, leaving the 2 paladins to finish the fight. In this case, the heals I got from Victory Rush were quite useful, and probably helped secure the kill.

Here's the log from our kill, if people wish to dig further: World of Logs - Real Time Raid Analysis

Note: I forgot to pick up the 50% bonus glyph, so the heals are significantly less than they should have been. But still strong.

Anyway, after tonight, I would believe that the point loss in Thunderstruck and Vigilance is worth it. Even in heroics... my healer got killed by Corborus's burrow but i managed to finish the fight alive with Impending Victory. If you wanted to min/max for single target, you could probably get rid of piercing howl and put the point back in Thunderstruck, but I get so much use out of the ability that I would hate to give it up.

Which is exactly what Blizzard wants. The whole reason for these new talent trees was so that there are more optional talents and less "you must go this build."

I personally refuse to take Impending Victory mainly because of the 50% chance for it to proc. If it was 100% after a Devastate then it would be interesting, but it currently is too unreliable for my taste.

I am loving Safeguard though. With the 30% damage reduction it is basically a mini shield wall on a 20 second cooldown for any player in your raid. The fact that the distance requirement was removed from Intervene makes it all the more appealing.

Putting vigilance on the other tank gives you a vengeance boost and near zero cooldown taunts.. I would have a huge problem dropping that for such a situational talent.

I´m also new here so forgive me if you guys consider the "which avoidance to take"-issue resolved, but as far as i can tell, by now it is no longer about how much damage we can reduce, but more about how much mana we use up per hitpoint lost, isn´t it?

seeing as how my healers have either casts that give me a regular decent amount of life for ca.2k mana or a big fat amount for 6k mana (ca threefold increase in costs but only 1.5-2 times increase in healing put out) the whole issue becomes a question of taking out spikes rather than going for pure avoidance and forcing healers to use a big fat chunk of mana at times, while at other times they still cannot will-regen because they have to heal the raid.

As far as i can remember during BC there where two theories or rather tank-concepts on how to build a tank:
EH (Effective Health) and pure avoidance (dodge/parry) - seeing as how pure avoidance gains better/more stats (as was discussed earlier in this thread), this would seems the choice to make, but I have to disagree: with the new deminishing returns on the dodge and parry stats it seems that once again we are pushed to EH in order to be most effective in a raidsetting, since we eg. lack the massive armor and shield of a druid, which allows him to go for dodge only (the OT in my guild has about 10k more armor than me and i guess I´m not the worst equipped tank out there with average ilvl of 351)

anyhow those are just my 2 cents on the matter...

For the most part to talent and glyph for Victory Rush will be up to the individuals play style and what they like which is as intended by Blizz as stated by a previous poster. I personally have found it usefull when I get a chance proc it does help to save the healers a little bit of mana. Are you sacrificing some dps for it, sure, I am sure there are some number crunchers out their that could tell us exactly how much if X were this and Y were that but for me it is about a 1-1.5k loss depending on the fight. Seems like a huge chunk, but sub 20% how long does a fight take 90 secs? tops? this means at most I would be giving up 135k dmg. Now since healers are doing triage more now than ever before who is number one on the list, us. Well if we can do something to help lessen the amount of mana required to keep us up then that might allow the healer the opportunity to move a heal to a dps target doing 12k dps and keep them up. Its a proc so obviously unreliable but still in my opinion a very useful talent to take and one that has kept me alive for the time needed to kill the boss when the healer went down.

As to Avoidance/Mitigation my healers have told me I am much easier to heal now that I focused on mitigation then I was when gemming and enchanting for stam. Currently I have 47% block, 14% parry, & 8% dodge. When a healer has to blow a big heal on you because you took an unmitigated hit it is a huge mana drain, avoiding the dmg all together would be nice, but the healers have told me they are much more comfortable with me taking steady damage then they are when I had gemmed and whatnot for stam making the dmg more spiky. My focus has been on pure Mastery since I am not really a big fan of Hold the Line since it seems like a chance for a chance if a chance happens. All of the above are just my opinions based on the couple weeks we have had in live to play around with this.

I do have one question though - seems kinda no brainer but never hurts to make sure (and could be updated on top post):
is critical block really just the double amount and does it stack with the "normal block"?
What I mean by this is the following: if I have 50% block chance and 25% crit block do i block 75% of all ingoing hits or just 50% and 1/4 of those become critical blocks?Havent seen anything so far on the matter, hope someone out there ran some tests on this...

Has anybody done any research on at what gearlevel we will be reaching near 100% dodge+parry+block?
Im already @ 12,5%+12,5%+46% with half heroicblues half 359 epics wich brings up the concern of what will the other tanking classes look like compared to a warrior at the gearlevel required for >100% dodgeparryblock?

edit: just as a clarification to what im after here, when you compare a paladin and a warrior
Paladins get more %block per mastery but once they reach a point where they block dodge or parry everyhing mastery becomes a wasted stat->forces a paladin to get more parry+dodge or exp/hit
Warriors get less block% per mastery but we would still benefit from mastery past the point where we block everything due to critblock

so given gearlevels would allow it this would already make our mastery far superior survivabilitywise.
Im interested in knowing what would the same thing look compared to a dk or a druid, even if gearlevels would never in reality go that large

Blizzard stated that bosses of later tiers will gain expertise, hit, avoidance and crit mitigation, in order to keep the effective crit chances / avoidance / etc. in control. We should no longer have mages that start the expansion 12% crit on critters and reach 70% crit on Lich King / Deathwing. Or tanks that were so slow they took about each single hit of a gnoll and are able to dodge or parry (!!!) more than half of the big dragon hits.

Hence, this question is somehow meaningless (and from my personnal point of view, it's good to remove those caps that are "too effective", and quickly become required for any serious raiding).

Tarilya of For the Horde wrote an excellent warrior tank guide. To bad it's in German. Manaflask - by Ensidia and For the Horde - Listandira's blog - Krieger Tank - Catacylsm
But the numbers still stand.

You need 102.4% to push white hits from bosses of the table. (Raid bosses only)
This can be achieved by 5% default enemy miss, 11% dodge, 11% parry, 25% block from shield block and 50.4% block via mastery/block rating.
If you want to achieve that number without shield block being active you need to stack mastery like Xav already stated.
If we assume some dreamy numbers of 15% dodge and parry, 5% enemy miss then you will need 67.4% block from mastery to push white hits of the table.

102.4-15-15-5=67.4%

But if I understand that post from tankspot correctly ([Warrior] Critical Block Mechanics) before you reach that point, avoidance will outplay mastery.

Regarding meta gems. The one that increases your shield block value will increase your mini shieldwall (blocked damage) from 30% to 31% and critical blocks from 60% to 62%.

Isn't miss chance versus a boss, this tier at least, 4.4%? I though that's where the extra 2.4% comes from, it's your parry, dodge, block, and miss chances reduced by 0.6%.

Also, Zork, that post was made back in August, and the testing was done at level 82 vs much lower level mobs. Wrong level and way too early in beta for the information to be valid now. I assume you're referring to the first post that is.

Mastery is still the best stat if anything because it helps reduce and smooth out incoming damage, making you easier to heal.

There's no question that the Mastery cap is much softer for Warriors than Pallies and it obviously depends on your specific stats, but do people think its worth stacking Mastery beyond the point of being unhittable if you could avoid the damage instead? I still have to crunch the actual numbers, so I'm not sure if it something we need to worry about (I'm fairly sure its already an issue for Pally tanks, but I know warrior Block rates are lower) taking into account the on use Block is also an issue. I know its not up nearly all the time, but it would still makes some component of Warrior Mastery useless a portion of the time, doesn't it? I doubt its something to worry about yet, but what about later tiers?

Taken from Dellingr's post over on the warrior boards:

Isn't there more to mastery than this though?

For one, as was more important back in wotlk, isn't it usually better to have a more consistent and less intense version of damage avoidance? Even if we are given that tank deaths due to spiky damage is removed when the tank performs appropriately (and judging from some previous posts, I'm not sure we are), could it really ever be removed so much that it makes up essentially 0% of all tank deaths? I guess what I'm saying is, shouldn't mastery be given a little bit more credit than just what damage it avoids, because it does help mitigate large spikes in physical damage better? I believe the line between damage avoidance and EH should be a bit less binary, with at least some consideration being placed on both. I'm not saying that we should rely on mastery solely based on this virtue, I'm just saying given that the line is so thin anyway, shouldn't this consideration tip the scales a bit?

Yes, that is my thought as well. BTW, these are the graphs that are being referred to:

[Warrior] Critical Block Mechanics

As I understand them, it means if you have a combined parry/dodge/mastery rating as defined in the title of the specific graph, increasing mastery rating from that point(x axis) will increase or decrease damage comparatively to an equal amount of dodge/parry. For example, the 4600 rating graph means mastery isn't as good as parry/dodge at reducing average damage. And after you hit the softcap(that's what those inflection points are), it still isn't better.

But, as you can see, the percentages for damage taken(y axis) are in tenths. This is really nit-picking the point. Which is fine with me, on a theoretical level. However, I honestly doubt your healer is going to notice the difference. If anything, you probably just caused a bit more overheal. After that realization, I'll go back to favoring mastery. Consistent damage taken will have much more of an impact and it's easier to fill out the attack table with mastery.

I don't think everyone is really interpreting those graphs and what they mean to Warrior tanking - they are just looking at them and saying "the graph says Dodge/Parry is better than Mastery at <insert irrelevant total "Avoidance" rating>."

Increasing Avoidance (Miss/Dodge/Parry) has always been the best way to take less damage over the course of the fight. When patch 5.0 hits and some other mechanic is added to the game, Dodge and Parry will most likely still be the best way to reduce the damage you take over the course of a fight. The thing is - tanks don't die over the course of a fight. They die with in seconds, or even fractions of a second. In very small time frames, Avoidance doesn't mean jack. What saves you is guaranteed damage reduction.

Regarding the quote from Dellingr's post: I don't understand why you would ever advocate moving away from becoming unhittable. Especially when tip-toeing the line between being unhittable with Shield Block up, or not.

The same regarding Impending Victory and saving healer mana. Healer mana will only be saved if the healer doesn't cast the spell. Healers will only stop casting a spell if they know a heal is incoming, and for how much. The amount healed with IV is predictable, the timing is not - therefore a healer will not stop casting. I'm not saying it's not going to increase your survivability, I'm saying it's not going to save healer mana.

I can not completly agree here. I only raided 10man so far, so maybe this is different in 25man, but tank deaths in current raid content are very much predictable. There are no spikes that insta-hit a tank that happen completly unforseen. Either its some special mechanic (Magmaw Crunch or how its called, Maloriak add stacking) or its some kind of void zone effect/effect on other players that cause large damage spikes. I am not a healer, but as far as I see it, there is not special tank healer anymore, so I guess if the tank does avoid one large blow, or even 2 in a row, the healers have more time to heal up the raid with their small spells.
On the other hand, if the tank takes damage the whole time, one healer has to stay on him exclusively to keep him up. I do not argue that mastery stacking will be more predictable for tank healing, but as soon as the healers are getting used to an avoidance geared tank, I think they could conserve more mana that way.

First of all, it somewhat depends on how your raid is organized. I know in the past, in large raid groups, healers have healing assignments to mitigate the possibility that after a large spike in damage to three different raid members (or in your example, the tank and two others), all healers focus on one target and the other two die. In the case of healer assignments, your example is more or less invalid, because the assigned healer will not lock up his GCDs if the tank isn't taking damage (and so would not heal other raid members). In some situations, healer assignments may not be needed - but if the need for healing is light enough that healers don't need some method to organize who is healing who on the fly (i.e. that overhealing - or wasted healing, is not planned for), is a couple percentage points more avoidance going to matter?

Secondly, if your tanks drop only due to a forseeable mechanic which occurs periodically, realistically, your healers must prepare for you to take full damage for a set period of time (as avoidance does not work all of the time) unless you are able to attain an avoidance cap that your "EH geared" counterpart could not. And keep in mind that when we say "avoidance geared" or "EH geared", we're mostly talking about trinkets or the selection of which avoidance stat one chooses. But at a very basic level, when only the most severe scenario (you dying - which is binary) matters, and since the number of periods of tank spike damage very quickly decreases the affect of avoidance on the most severe case (unless the number of attacks in each "spike" is very high, in which case, avoidance's mitigation of spikes in damage approaches its' avoidance value), EH is going to reduce the number of tank deaths faster than avoidance assuming reasonable values for both. Wotlk should have taught us that.

It really isn't as simple as "if one tank takes damage the whole time, and another doesn't" because of diminishing returns, because the mastery geared tank only takes more damage over small intervals a certain % of the time - and therefore, depending on overhealing, may only take less healing an even smaller % of the time, and because mastery and avoidance stats compare so closely in terms of avoided damage. Your last statement implies that in raiding gear, converting x mastery rating to x avoidance rating will create a substantial enough difference to affect healers' decision making. How large is the x in this statement and do you have any concrete support for this idea?

My experience with tanking has been that when I just barely avoid dying through avoidance I say "Man, glad I had that avoidance" even though I don't know for a fact I would have avoided the blow naked, and when I just barely die I say "Man, why couldn't I have avoided that? Better get some more avoidance" when I don't know for a fact that any level of attainable avoidance would have been enough to avoid the blow. I guess what I'm saying is that basing your decisions solely on perception, especially pertaining to randomly occuring events, is a poor choice. Several studies on the affect the human brain has on the perception of random events support this claim (For example, see the study published in MONEY here: http://www.jasonzweig.com/uploads/11...oeconomics.pdf).