[MoP] Protection Paladin Guide

[Drummerboy]

Just for clarity, keep in mind that what I'm offering is not a "proof" by any formal definition. It is a set of data, and the analysis accompanying that data. Any good scientist would balk at calling that a proof, because it isn't a universal truth derived from fundamental equations and axioms.

Ah the breaking down of word usage. Nerd arguments are so much fun.

I don't consider you a paladin saint but I do respect your well thought reasoning and your time. I respect Firefly's time (and everyone else's time) as well. After all we aren't making money doing this (usually). I see that your spending a lot of effort in this. And I'm only defending your efforts as I would defend my own in any project I work on.

Of course I could come up with suggestions as well but I don't want to make more work for you. For example instead of simming a "healthless" tank you could sim a "mid-tier geared" tank and simulate a constant incoming heal on the tank based on the average "mid-tier geared" healer and call a sim failed when the tank dies. I could suggest running only 1 minute tank tests since most swap fights consist of roughly that amount of time tanking. I could make these suggestions but I won't. :P

You could sim yourself to death but there's a point where I start to see it as "too much" with little gained other than a more close "human" simulation and I don't think that is needed. Help out SETI or protein folding instead. jk

[Nairobi]

Curious how other people have been approaching their trinket slots.

I avoided using accuracy trinkets for a long time, esp last tier when we had good stamina trinket options, for fear of being outside of caps if/when swapping.

However, I have since abandonded that.

I use HC Feather (Ref Hit->Haste) and HC Spark probably 99% of the time. I won't bother to re-spew my diatribe on the wonders of these trinkets as stat banks, their procs and scaling with our absurd haste, or their undeniable positive interactions with DPS and mitigation, but you can probably find those thoughts on the front page somewhere.

I have a Soul Barrier that I use VERY infrequently on progression if/when magic damage is an issue. I swap that in over Spark, but I hardly use it anymore to be honest. The on-use is terrible, even in 10man. If there was a similar trinket to last tier's stam ones (with on-use mastery!) to macro to HA, I may opt for that, but as there isn't, I don't. I also have vendored my JiKun's Winds POS. I have not attempted to roll for a HC Fort of Zandalari, but that would/could be a good option...I just don't see myself needing it currently (only 6/13 HC tho).

So yes, my TL;DR is embrace the accuracy trinkets, as they are beast as all hell. Even the SPA rep trinket, which I used prior to Feather, was a great pick-up. Once you get a feather and see it's uptime (PS, learn to watch it's proc and refresh SS at 10 stacks for bonus shields!) you'll wonder what you did before it.



Just an addition here, but I wanted to jump on the HA train as well. I know, stochastically, that at moderate haste levels DP will pull ahead for overall uptime. I use it on fights like Tortos (well, really, ONLY on Tortos) where I just stand there and take a constant amount of damage. Sometimes, I can get procs on procs, and stack up 8-10 seconds of coverage. Sometimes, I feel like I forgot to actually learn the talent. I have always, and will continue to swear by HA > all for it's reliability, DPS contribution, strength as a CD, duration, ability to be cheesed with mastery-trinkets, and even the fact that it allows you to "coast" for 30 seconds while basically not worrying about timing ShotR. HA on HC Sha of Fear is up for every other N+A phase (assuming tank swaps), which makes 50% of the tanking mindless and error-proof. Its DPS is also nothing to scoff at, and the fact that it comes WITH a powerful defensive CD is huge. On Phase 4 Iron Qon, popping HA (with wings) lets me burn through dogs upwards of 600k DPS, while running a cooldown that is stronger than GoAK. I just can't see losing it for anything else except a Patchwerk fight like Tortos.

[Theck]

Of course I could come up with suggestions as well but I don't want to make more work for you. For example instead of simming a "healthless" tank you could sim a "mid-tier geared" tank and simulate a constant incoming heal on the tank based on the average "mid-tier geared" healer and call a sim failed when the tank dies. I could suggest running only 1 minute tank tests since most swap fights consist of roughly that amount of time tanking.

I talk about this some in today's blog post, actually. You could certainly sim that, but I'm not sure it tells you anything interesting. Constant healing isn't a very realistic scenario, for starters. Real healers ramp up their healing when you take a spike. So a simulation with constant healing isn't that different from a simple time-to-live analysis, where you just calculate how long the tank lives on average in the absence of healing.

The one difference is that if you choose the constant healing appropriately, such that it exceeds average throughput, your tank lives until he takes a big enough spike (or combination of smaller spikes in rapid succession). And the information that you get out of that is highly correlated to the magnitude and frequency of spikes the tank takes - the tank that lives longest in that sim is the one who either takes the smallest spikes or takes spikes least frequently. But spike magnitude and frequency is exactly the data my simulation provides already.

[Drummerboy]

Perfection...

It's like a the baseball player hitting the 90+ mph pitch. Can an average person hit that pitch? No. Can a seasoned player who has seen that pitch many times before hit it? Yes. It takes perfect timing and bat placement for the batter to make proper contact with that pitch to hit it out of the park. Dare I say perfection (or at least damned close).

Perfect raiding... actually it is something that I think progression teams try to strive for. Progression raid teams are in the majors. Everyone is expected to act precisely when on the "gear-cusp" of that encounter. And with most spikes occuring and predictable times these can and usually are planned and handled.

One could extrapulate out the killing of a progression boss and compare that with the batter homerunning the fast pitch. Many things have to happen in a sequence of events for the desired outcome to occur. The outcome is not insurmountable nor are those conditions. And you could say gear equates to the size of the bat. Using a bigger bat probably makes the objective of hitting the ball easier. The bigger bat reduces the margin of error.

TLDR? Sim for perfection because it is applicable. Also I ramble.

[lakhesis]

One could extrapulate out the killing of a progression boss and compare that with the batter homerunning the fast pitch. Many things have to happen in a sequence of events for the desired outcome to occur. The outcome is not insurmountable nor are those conditions.

That's actually one of the elements I've really enjoyed about tanking Lei Shen. One of the keys to that fight, at least for my group, was precisely timing platform swaps off upcoming ability timings so that we deliberately deactivated a pillar just before he was about to use its ability.

Simply looking at the default approach of "his health is X, the pillar energy is Y, therefore we should platform swap" was invariably a mess, because we'd end up randomly overlapping annoying/dangerous abilities. Having to carefully thread abilities is much more interesting than watching health/power bars.

[Deleted]

I think what you're hitting on is the major difference between 25 and 10: tank throughput. 10-man tanks deal with much smaller melee attacks because of the healer constraints in that format, and as a result, they naturally have a little smoother intake than 25-man tanks. ...

To somewhat offset this, I wonder if 25 man tanks have more reliable healing intake than 10 man tanks? With 6 healers for 2 tanks, there will be more dedicated healing than with 2 (or 3) healers for 2 tanks. It might even show up as more overhealing of the tanks and the tanks being kept topped up more consistently. I sometimes feel I die in 10 mans when the healers are focusing elsewhere and I am left to my own devices. It's one reason I don't value avoidance at zero as Firefly does, as I don't think it's benefits are always lost in overhealing. If I am left to my own devices, those dodges or parries might save me.

So it doesn't really surprise me that you find yourself nearly invulnerable in 10-man raiding, because the survivability threshold is lower in that format.

Another factor here is player skill. A skilled tank with skilled healers will need less effective health and less survivability to kill a boss than less skilled players do. I can believe that highly skilled players (i.e. those focusing on heroic modes, topping WoL etc) such as you and Firefly would feel invulnerable in 10 mans. But as a less skilled player (one who'll likely never see a heroic mode fight save Morchok), I don't recognise the feeling of invulnerability in 10 mans. I've dropped dead after an Elegon breath, after a dance + gas on Will, a second overwhelming assault on the blade lord etc. Sometimes it's me not hitting a CD; sometimes it's healer inattention; sometimes both. It doesn't happen often - they are mistakes - and are particularly rare once a boss is on farm. But experiences like this are why I wear two stamina trinkets in 10 man normal. I'm gearing to feel as invulnerable as I can be, but I'm not there yet.

[Deleted]

I think what you touch about 10vs25 tanks having more reliable healing econ.
Is that under 'normal' circumstances aka only boss melee hits, a 10 tank can easily heal himself for 75% of the damage he takes. So when the damage spikes on him, say boss enters a phase where he deals 150% more damage, suddenly the healers have to heal 600% more than before, even though it is not high numbers it can catch unseasoned healers off guard. In 25 tanks require constant attention so healers are more prepared for the spikes.

I remember at the start of t14, we were using a pug healer, who was... Not bad, but mediocre at best. I was taking less damage than our warrior tank, on emperor normal that is, but during titan gases I died. Even though I was also healing myself more. The pug (druid) was assigned to me, we checked wol and during the normal phases I was healing myself for 85-95% of the damage I took. So he didnt have to heal me at all.
But then when titan gases came, my healing was not nearly enough (duh) and he forgot to heal me as he was used to me healing myself.

So the lower tank damage is both a blessing and a cursr for 10 tanks. But also why I feel (omg no statistical data) that the areas around the large damage phases are irrelevant from 10 tank pov.

And yes drumme, as I mentioned way back, I use math to back up logic, but in the end, logic>math for me, which is why 'I feel' is a completely relevant answer for me. As numbers can tell 1 thing but reality another. Math is great to get an understanding or tipping a scale, but in no way is it the sole determining factor.

---------- Post added 2013-04-24 at 12:33 AM ----------

Which again brings back to why I disagree in haste vs exp as i think that the way current math is modelled gives the reliability of expertise far greater value than reality, as in reality, as a tank, I am not even concerned about 80% of a fight since I know that even tanking boss with my back going to the kitchen making a toast, I wont die during those 80% of the fight, it is during the remainder 20% i can die. and for those 20% i am always prepared with 5 hopo regardless of parries in the 'non interesting' part of the fight.

Which gives exp far greater value than the reality (10 man reality? Not sure about 25, maybe entire fight is a danger then).
Also to mention not taking account for rppm procs and dps increase.

In my thoughts, thats a reasonable trade off.

[Fnatick]

Can I ask this question here? I'm about to choose a meta legendary and of course I'd instantly go for the Stamina one as a Paladin tank (10man). I know this is frowned upon but I couldn't resist to check some armory's and see what the pro league players are doing. To my bewilderment Paragons Paladin tank is using the Crit/Capacitance gem?! I think I know half the answer and that is he has enough stats to just go all out dps mode and thus choosing the dps option gem. But let me get this straight, the crit aside since it does nothing for tanks (right?:s) from wowhead comments I see that the proc does: "When Capacitance reaches 5 charges, you will deal a Lightning Strike to your current target for 100 Nature damage." It doesn't look all that impressive to me when I read that (unless it stacks really really fast?). Any thoughts?

[Drummerboy]

I've dropped dead after an Elegon breath, after a dance + gas on Will, a second overwhelming assault on the blade lord etc. Sometimes it's me not hitting a CD; sometimes it's healer inattention; sometimes both.

First don't take the term invulnerable too literally. All tanks die. That said I am told I'm less squishy then my tank partner because of things like HA, DP and the handful of other CD's and self saves I have.

As far as like certain mechanics, the best way to handle those is watching spell timers that DBM puts up. You plan your CD's for these moments. Elegon's breath was a great example where DBM put up a timer for when the breath came off CD and you could make sure to have a CD available. And for Will Guardian worked very nicely for every gas. I will say Blade Lord was a prick in that assault wouldn't hit consistently based on the spell timer and you take an unmitigated 400K hit because you hit Shield 4 seconds too soon.

Currently council (on farm) is still a pain to me because of our strategy. I tank all frost stacks and sometimes the healer nods off when I get frozen and then I'm eating ground. I hit my own CD before I freeze and even go as far to call out each healer CD on me to chain them but I still die once in a while. Since the fight is on farm we don't really want to change the strat (well I do because I'm so sick of dieing) but I guess me dieing is now part of the strat. :P

As you pointed out though you could be the best skilled tank but if you don't have a skilled healer to keep you up then you're still going to die. Paladins can save themselves from at most 3 near death events (LoH, AD, and 5 stack WoG). Once used up if you still die though despite proper CD usage and self-saves then it's all on your healers.

One difference between 10 and 25 is that in a 25 it is more likely to have a healer class variety where as with 10 you get 3 and its not unheard of to have 2 of those 3 being the same class. 10's are much more sensitive to raid make-up and that can vastly affect strategy as well. In your case it could be that because of your raid make-up tank healing isn't as inheritly strong so double-stam trinkets are necessary.

All that said back to the avoidance thing, granted a dodge can save. Think though. You're relying on RNG to save you but look at how much it fails you when you miss when not capped or when you use your Mogu Rune. There is no way to know if RNG will help you when you need it, but you know keeping Shield up will help.

[Ansekh]

Can I ask this question here? I'm about to choose a meta legendary and of course I'd instantly go for the Stamina one as a Paladin tank (10man). I know this is frowned upon but I couldn't resist to check some armory's and see what the pro league players are doing. To my bewilderment Paragons Paladin tank is using the Crit/Capacitance gem?! I think I know half the answer and that is he has enough stats to just go all out dps mode and thus choosing the dps option gem. But let me get this straight, the crit aside since it does nothing for tanks (right?:s) from wowhead comments I see that the proc does: "When Capacitance reaches 5 charges, you will deal a Lightning Strike to your current target for 100 Nature damage." It doesn't look all that impressive to me when I read that (unless it stacks really really fast?). Any thoughts?

First of all I don't believe gearing up as top end tanks do is the right way, as their group is undergear for the content they are doing. Requiring much more health or dps than casual guild's tanks do. They know the reason why they are doing it, but for some tanks that are mindlessly coping the top guild tanks, bigger chance it's a disaster than helpfull to gear the same way. All the top prot paladins in 10 man go for the dps gem, if I'm correct.

also i believe the damage scale's with attack power. I have read in this thread that the gem is 11% dps increase.

Beside this all, the stamina gems can proc at times when you don't need it. As firefly explains a couple of post back, 80% you don't die, for the other 20% you need use StoR or cd's. "Yeah I got a 10% less damage intake, but the boss is doing any noticable damage to me" /sarcasme.

[Xs]

Can I ask this question here? I'm about to choose a meta legendary and of course I'd instantly go for the Stamina one as a Paladin tank (10man). I know this is frowned upon but I couldn't resist to check some armory's and see what the pro league players are doing. To my bewilderment Paragons Paladin tank is using the Crit/Capacitance gem?! I think I know half the answer and that is he has enough stats to just go all out dps mode and thus choosing the dps option gem. But let me get this straight, the crit aside since it does nothing for tanks (right?:s) from wowhead comments I see that the proc does: "When Capacitance reaches 5 charges, you will deal a Lightning Strike to your current target for 100 Nature damage." It doesn't look all that impressive to me when I read that (unless it stacks really really fast?). Any thoughts?

As with all the Legendary Metas, the difference it makes is absolutely huge, and in our case the DPS meta scales incredibly well with Haste stacking (if you're already doing that) and as such you'll see a crazy DPS increase (around 10% is the average reported) which comparatively to the tank meta, is pretty much a no brainer

[Lethora]

First of all I don't believe gearing up as top end tanks do is the right way, as their group is undergear for the content they are doing. Requiring much more health or dps than casual guild's tanks do. They know the reason why they are doing it, but for some tanks that are mindlessly coping the top guild tanks, bigger chance it's a disaster than helpfull to gear the same way. All the top prot paladins in 10 man go for the dps gem, if I'm correct.

also i believe the damage scale's with attack power. I have read in this thread that the gem is 11% dps increase.

Beside this all, the stamina gems can proc at times when you don't need it. As firefly explains a couple of post back, 80% you don't die, for the other 20% you need use StoR or cd's. "Yeah I got a 10% less damage intake, but the boss is doing any noticable damage to me" /sarcasme.

Here's log from our 25man Horridon Normal kill tonight. We're fairly slow on progression (1/12H) since we raid only 10 hours/week, but you can see that dps meta is really shining.
It made up a whooping 12.5% of my damage.
http://www.worldoflogs.com/reports/r...?s=4182&e=4763

[Deleted]

All that said back to the avoidance thing, granted a dodge can save. Think though. You're relying on RNG to save you but look at how much it fails you when you miss when not capped or when you use your Mogu Rune. There is no way to know if RNG will help you when you need it, but you know keeping Shield up will help.

Absolutely, no disagreement there. I reforge all the avoidance away that I can and try to keep my shield up when I need it. It's just I think the "avoidance = RNG = 0" argument is over done.

In my experience, raid wipes due to tank deaths often feel like bad RNG. You make a mistake, or your healers do. On 10 normal, being well geared, you#ll usually be ok even with such mistakes. You've got a sizable buffer of hit points and two healers with mana who should be able to rescue you. Yet bad stuff happens, so sometimes you die. And when you die there's that disappointed sound of surprise over Ventrilo or Teamspeak "oh look, the main tank's dead, wipe it...". You got caught out for the mistake this time. The combat rez has already been used. The raid leader says: "Unlucky. Let's do it again and focus more this time."

Having a bit more stamina reduces such bad RNG a little. Statistically a bit more avoidance might too. A little bit of good RNG might offset the bad RNG. The fact that avoidance relies on RNG is a disadvantage but doesn't mean it should be dismissed completely. Ask a bear or a brewmaster about their active mitigation. Remember why we wanted Moros' pocketwatch to drop in Karazhan. The effects will be small, but when we are considering gear upgrades, the effects even of the best stats are incremental and small. The benefits of avoidance will be smaller than other more reliable, survivability stats, no disagreement there. Small but not zero, is all I was trying to say.

There is a trade off between a little more survivability or more dps. Skilled tanks will need less surivability to clear any particular content. On normal modes and/or 10 mans, they'll often trade a little survivability for a lot of dps. (Although it's interesting with ToT to see an increasing number favor survivability for heroic modes and/or 25 mans.) However as a slowly progressing tank, I think I'm probably better advised to err on the safe side. My reasoning is that by going all out for survivability, I give the raid more chance to learn the encounter. We're not pushing the envelope - we should progress if we get the dance right.

One feature of being a slowly progressing guild is probably that we do things slow: raid just once a week, often kill a later boss months after others have it on farm, but with relatively few attempts. So it's good to try to make those limited attempts count - after a couple of weeks wiping on the same boss, there's a real risk of signups slipping and not being able to field 10 people the next week.

If there's an Ultraxion type fight with a dps check that bites on normal, then sure I would try to maximise dps for the minimum necessary survivability. But Blizzard have decided against such checks in normal. Just look at Jin'rokh live compared to on the ptr - by all reports he was a dps check for 502 geared raids on the ptr but when he was brought to live, his hitpoints were cut by a third. Instead normal fights are more about everyone executing their roles well. The fewer sounds of sad disappointment that I hear directed at me over my microphone, the better.

[Nairobi]

While I can see some of your points here, I'd be remiss if I didn't comment on the monk/bear points, as well as the BC gearing strategies:

Leather tanks, for better or worse, have always been the innately higher "avoidance" tanks due to the nature of dodge. It was not until Cata tha STR provided any real benefit to tanks, other than damage. Yes, through BC there was a [terrible] conversion ratio of STR:Block Value, but with the removal of block value items (which I am still mad about!) that faded away. Likewise at this time, there was no such thing as Mastery or any of the fancy AM items we have today. Warriors had Shield Block (albeit very different), and Paladins had "stand there and get hit for mana" tanking. D/P was the only thing TO get; not making it good, just making it all there was. And even then, stamina was still better.

Now, I'm not trying to compare the "good ol days" to now, just saying that you CAN'T compare them. Monks and druids now still DO focus on avoidance (via gearing innately and by AM tools), but they have the supporting cast of toolkits to prop up this "less reliable" method. 1) Their avoidance levels are nearly an order of magnitude higher than ours, removing a lot of the RNG by placing avoid chance in the majority, rather than minority. 2) They have other, reliable passives (Stagger for monks, huge armor and magic redux for bears) that prop up the times that RNG DOES strike. And guess what; they STILL get flattened occasionally due to bad streaks.

I'm not saying that avoidance is a bad stat, or that it will make you less likely to survive. Of course, 1% more chance to parry means that you can completely avoid an attack, less damage taken, looks good on paper. The issue is that we're designed as stable tanks; we don't take the least amount of damage overall, but we take the lowest variance, which is pure gold to healers. Things that reduce that variance like haste, mastery, and accuracy will obviously be better than things that jeopardize stability for potentially more TDR (avoidance). That alone should be enough reason to avoid avoidance, since any stat allocations you have in the avoidance category will be directly related to less allocation to haste/mast/acc.

So am I saying avoidance is garbage? No. I still have one piece of dodge/parry gear, simply because it's ilvl is so much ahead that it made no sense NOT to use it. I'd suggest that anyone else in a similar situation (<522 haste piece vs 535+ D/P piece) make the same choice, or at least give it consideration. Additionally, I'm using my 2 DPS trinkets that provide large STR buffs in part BECAUSE of the avoidance they give. However, this is due in large part to the massive bolus of avoidance given during proc. ~8500 and ~18000 STR procs mean large amounts of parry (for free) that ARE noticeable. Using these proc's as mini-cooldowns allows me to even change my normal cooldown use sometimes (thanks to haste/RPPM interactions, the uptime on these for prot is quite high and pretty reliable).

All of that is to lead me to my last point on this, which is that whether everyone likes it/agrees or not, tank DPS does matter in 10m. It matters a lot. Normal or heroic, a good tank can pull equal to or higher than a lot of DPS, when played appropriately. Key word here, though, is good; if you're not experienced or familiar enough, you'll just be a sponge/corpse. But once you are familiar, you become even more fundamental to raid success. I regularly rank on WOL on pretty much any fight in the top 50 (not bragging, just reference), and finish in the top 3-4 on damage in our 10man. I'm also between 1-4 on healing, with the way that our damage and haste scale with our self/raid heals.

Those contributions allow us to drop a healer (or gain a healer, depending on need) to alter strats or bypass a challenging part of the encounter, FAR more than any amount of stamina or avoidance ever could.

[Drummerboy]

There is a trade off between a little more survivability or more dps. Skilled tanks will need less surivability to clear any particular content. On normal modes and/or 10 mans, they'll often trade a little survivability for a lot of dps. (Although it's interesting with ToT to see an increasing number favor survivability for heroic modes and/or 25 mans.) However as a slowly progressing tank, I think I'm probably better advised to err on the safe side. My reasoning is that by going all out for survivability, I give the raid more chance to learn the encounter. We're not pushing the envelope - we should progress if we get the dance right.

I think you're missing that Haste/Control and Mastery/Control is about controlling your survivability, not having more dps (that's just a perk). More haste = more HoPo = more shield uptime. The rest is icing.

And I don't know about the generalization of trading survival for dps in 25, especially since I raid 10. All I know is when I have Holy Avenger and all my CD's ready and I have to tank the boss for the next minute, that boss better bring all he's got (other than freezing me) or the healers will be spending very little.

I will say this. If you are hearing "bad sounds" over vent because you died then check all your CD's. Did you miss using one and the hit that killed you was predictable? Then you probably could have done something to stay alive. Otherwise talk to your healers and find out why they're struggling to keep you up.

[Nairobi]

I will say this. If you are hearing "bad sounds" over vent because you died then check all your CD's. Did you miss using one and the hit that killed you was predictable? Then you probably could have done something to stay alive. Otherwise talk to your healers and find out why they're struggling to keep you up.

A good point; outside of standing in 1-shot mechanics like Durumu beams or LeiShen decaps at 0yards, normal 10mans really have no imminent danger to the tank. Especially a prot paladin, assuming at least a modicum of timing in using ShotR. If you macro ShotR to CS or something, I can see you having issues, but by and large if you are using it around the big hits (Trip Punc, Snap Bite, Talon Rake, etc) you should have no issues. Additionally, your self healing should be about even with (or greather than) the healers healing on you. Prot's are perhaps the most self sufficient tank, allowing the healers to focus on the raid rather than tank damage.

If you died despite having ShotR and/or CD's up, like getting whittled down over 8-10 seconds, that's not your fault at all.

If you died on JiKun cause you had about 6 too many beers during trash and ate a second talon rake without ShotR after a pushback...that may be your fault.

Not that I'd know anything about that...

[Deleted]

I regularly rank on WOL on pretty much any fight in the top 50 (not bragging, just reference), and finish in the top 3-4 on damage in our 10man. I'm also between 1-4 on healing, with the way that our damage and haste scale with our self/raid heals.

Those contributions allow us to drop a healer (or gain a healer, depending on need) to alter strats or bypass a challenging part of the encounter, FAR more than any amount of stamina or avoidance ever could.

Yeah, good points. But while you wouldn't rank on WoL, you might still be high up in your raid group for damage and healing even if you favored stamina above haste. It's just how good protection paladins are now, at least on some fights.

For example, if I were to gem and trinket for haste rather than follow theck's control/haste weights in AMR, which favor stamina, then I would lose about 9% health and gain 8% haste. So my dps would be roughly 8% higher. So raid dps would be about 1% higher. That wouldn't allow my raid to drop or gain a healer, or even bypass mechanics.

For an experienced raid team, going for the dps can make sense. But so long as my deaths cause a wipes, I'm inclined to play it safe. As I said, my focus is on staying up long enough for us to learn the fight - if we hit an enrage, then I'd start to rethink.

[Nairobi]

I guess, but if you're not having trouble staying alive at the moment, adding more survival has a net benefit of zero whereas haste has constantly increasing benefits to both damage and healing output (up to 50% anyway).

But yes, if you ARE having trouble staying alive, this is clearly not the advocated method of gearing (for now).

Not to nitpick about %'s though...but I'm going to. 8% haste in your example is actuall more than 8% dps increase, in practice. It's 8% more melee damage, for sure, but the effect on things via SoB yields higher returns, since HoPo generation goes up (as well as HoPo spending, via ShotR), CD goes down, and items that function off of RPPM (like trinkets, meta gem) increase with haste. Synergistically, youre looking more at 10-12% damage increase for that 8% (as well as associated healing and ShotR uptime). Further, I'm usually ~12-15% of the raid's damage, so upping that is more on the lines of 1.5% overall damage for the RAID, by swapping gems. Certainly hefty for 9% less stam, if you can afford to.

Will that 8% haste change allow you to drop a healer every time? Nah, probably not. But, it may/will allow you to push a breakpoint that you couldnt before, like 2 breaths per head on Meg instead of 3, 1 windstorm on IQ instead of 2, etc.

But again, if you can't stay up, this doesn't really matter

[Deleted]

I will say this. If you are hearing "bad sounds" over vent because you died then check all your CD's. Did you miss using one and the hit that killed you was predictable? Then you probably could have done something to stay alive. Otherwise talk to your healers and find out why they're struggling to keep you up.

No disagreement there, recall the premise of my argument:

In my experience, raid wipes due to tank deaths often feel like bad RNG.You make a mistake, or your healers do.

We die when we make mistakes. If you and your healers play flawlessly, then you'll kill the boss. That's how the game's tuned. If you play flawlessly, you can kill the boss in rubbish gear. You can even kill the boss if you make quite a lot of mistakes. That's why I said it feels like bad RNG. It's not purely bad RNG. You did a bad thing - you made a mistake. But this time, you were unlucky and got caught out for it.

All raid teams wipe and they wipe repeatedly - they wipe while learning the fight. And the process of learning the fight inevitably involves making mistakes. You gear for progression, in the knowledge that people will make mistakes. Gearing for perfect play is like gearing for farm content - rather irrelevant.

I think you're missing that Haste/Control and Mastery/Control is about controlling your survivability, not having more dps (that's just a perk).

I follow the haste/control gearing but with the AMR/theck weights that prioritize stamina over haste. I suspect stamina is the bigger boost to survivability than haste. (Theck's sims imply bigger by a considerable factor, perhaps 5:1 in his latest sims.)

[Deleted]

To reiterate the point: whilst stamina may be the best survivability stat, it certainly becomes a waste after a certain threshold, particularly in a 10 man environ. THe tricky bit is determining what that level is for the encounter. On the flip side, haste is never wasted.

It is, and always will be, finding the right balance for you, your healers and your raid.