Paladins are the worst tanks?

Anything, including off-topic posts

Moderators: Fridmarr, Worldie, Aergis, Sabindeus, PsiVen

Re: Paladins are the worst tanks?

Postby Fridmarr » Sat Mar 31, 2012 7:41 am

...and this is why I didn't want to get into this, so this will be my last post on the matter. We've had discussions on this before, this is all just a rehash of what's already been said, nothing new. If you want to continue on this, just make a new thread or find one of the old ones.

The reason gear and nerfs are different is because a nerf is a change to the baseline of the encounter. If the boss' damage is reduced by 10%, it's 10% for everyone. If you get a new piece of gear that should increase your hps/dps by 10, that's only a potential number. Excellent players will see the 10 hps/dps jump, bad players may only get 3. The point being that gear is accounted for by the skill. Further, just as a nature of a comparison, it's silly to compare the affects of class balance for an encounter, when the encounters not the same. I'm not saying that class balance matters as much on a nerfed encounter as it does on a non nerfed one, so there is no reason for you to keep harping on that since no one disagrees.

The reason why wipes don't matter is because they don't include a crystal ball which somehow allows the fight to continue to show you all the ways you would have wiped. If someone blows up the raid and causes a wipe, it hides the fact that you weren't going to beat the enrage timer anyhow, so to somehow statistically lump that in with "people doing stupid things", and not "dps was too low" yields the misleading result that is allowing you to reach your false conclusion. Aggregating those wipes into categories and then extrapolating from that is only compounding the problem.

Because less skilled raids tend to not have the reaction times, situational awareness, HPS of a higher skilled raid, a tank taking more damage is a huge amount of stress because healers don't have the bandwidth. Taking your examples, when people do stupid things to take damage, that occupies the healers focus, when the tank is also taking more damage, that leads to a huge problem. It can even look like the wipe occurred because someone did something stupid, or too many people died and resulted in DPS being too low, when in reality healers couldn't address those problem because they had to heal the tank instead of keeping people up who stood in a fire a little too long, leading to further bad analysis if you try to categorize wipes that way. Class balance particularly on a tank, can matter just as much for them as anyone else.
Fridmarr
Global Mod
 
Posts: 9667
Joined: Sun Apr 08, 2007 1:03 am

Re: Paladins are the worst tanks?

Postby theckhd » Sat Mar 31, 2012 9:10 am

I don't see why this needs yet another thread, it's entirely relevant to the topic in the OP, which is asking about relative tank balance.

I don't think it's fair at all to say wipes don't matter. Wipes are everything. Wipes are the learning process, and what highlights the obstacles you need to overcome a kill. Someone blowing up the raid is a legitimate problem that needs to be addressed, even if it does hide the fact that you weren't going to beat enrage. A tank consistently dying to burst is a similar issue. There's no error in analyzing these statistically, because the whole process of succeeding at an encounter is one of probability - minimize the probability of error or failure in any mode in order to increase the probability of success.

If your tank has a 10% chance of dying to burst, then he's a bigger problem than a tank that only has a 5% chance. But as gear improves (or nerfs increase), that chance gets smaller independently of anything else in the encounter. When one tank has a 1% chance of dying and the other has a 2% chance (because nerfs reduce the burst potential, or gear turns fatal burst conditions into survivable ones), it no longer becomes as significant a concern. That will also be reflected in the statistics that Rhi presented.

In addition, the point that keeps being danced around here is one of options. At the bleeding edge, where there's very little room to improve in terms of raw skill, you literally cannot just "play better" to overcome tank imbalances. You do not have that option. For a week-13 guild, where there's a lot of skill overhead to be taken advantage of, you do have that option. Sure, the DK tank playing at 60% capacity might have an edge over the Paladin tank playing at 60% capacity (though again, those edges will be significantly reduced compared to the week-1 guild thanks to gear inflation). But both tanks have the potential to practice, pay more attention to the encounter mechanics, and improve their skill. The paladin tank playing at 70% capacity will always "beat" the DK tank playing at 60% capacity in a week-13 guild. The only case where that won't be true is for things completely outside of player skill control, like the ridiculous 4-piece set bonuses on T13.

Which is why when someone says "my guild won't let me tank because they say paladin tanks are worse than DK tanks," it's completely fair to say "tank balance doesn't matter at your progression level." Because the best advice we can give them is "practice and improve your play, and you will definitely be able to tank the encounter."
"Theck, Bringer of Numbers and Pounding Headaches," courtesy of Grehn|Skipjack.
MATLAB 5.x, Simcraft 6.x, Call to Arms 6.0, Talent Spec & Glyph Guide 5.x, Blog: Sacred Duty
User avatar
theckhd
Moderator
 
Posts: 7950
Joined: Thu Jul 31, 2008 3:06 pm
Location: Harrisburg, PA

Re: Paladins are the worst tanks?

Postby Fridmarr » Sat Mar 31, 2012 9:45 am

Because as I've said, this has been discussed many times before. There is no need for yet another thread on it.

Wipes simply don't tell you the information you need. They can tell you why you wiped, but that doesn't mean that without whatever caused that wipe, you would have succeeded so its very incomplete data. Even then, determining the cause of the wipe isn't always available in the logs with the detail you really need.

In addition, most people literally cannot just "play better" either, otherwise the skill level would always be raising. They can perform better in an encounter, but if they have been playing at 70% capacity, they aren't going to change that number a whole lot. Bleeding edge guilds do that too, they wipe a few times as they get used to the mechanics of an encounter, but they certainly improve their handling of those mechanics as they increase their attempts. They likely do so at much faster rate than non bleeding edge guilds.

It might be fair to say, but it's also not accurate. It's not really different than someone saying, "my guild won't let me tank because my gems and enchants aren't correct", and my guess is that the answer wouldn't be..."your gems and enchants don't matter at your progression level". Honestly, by suggesting that class balance doesn't matter at lower progression levels, it's hard to make a case that much does matter at lower progression levels. It's all just inputs and outputs. The fact that improving play is the best way to increase their success is obvious, but it's not a mutually exclusive property. Just because playing better is going to grant the biggest benefit doesn't mean that class balance doesn't matter.
Fridmarr
Global Mod
 
Posts: 9667
Joined: Sun Apr 08, 2007 1:03 am

Re: Paladins are the worst tanks?

Postby theckhd » Sat Mar 31, 2012 10:08 am

Fridmarr wrote:Wipes simply don't tell you the information you need. They can tell you why you wiped, but that doesn't mean that without whatever caused that wipe, you would have succeeded so its very incomplete data. Even then, determining the cause of the wipe isn't always available in the logs with the detail you really need.

I disagree. They tell you why you wiped, which is why you didn't succeed. It may be one of several reasons you wouldn't have succeeded, but that's still information that needs to be dissected and incorporated back into your strategy. The feedback might be anywhere from "quit standing in fire" to "let's adjust the strategy so this isn't a concern," but either way that feedback loop matters. The fact that you need to deal with all of those issues doesn't make analyzing and dealing with one of them less relevant. You keep eliminating issues one at a time to increase your probability of success until you succeed.

Fridmarr wrote:In addition, most people literally cannot just "play better" either, otherwise the skill level would always be raising.
That's awfully pessimistic. I think most people can just "play better," but they need to make the effort to do so (by practicing, doing their homework, etc.). There are very few cases where someone is inherently limited from improving. Even I see improvement in my play from week to week, both as I learn the encounters and as I try to correct errors or extinguish bad habits.

Fridmarr wrote:It's not really different than someone saying, "my guild won't let me tank because my gems and enchants aren't correct", and my guess is that the answer wouldn't be..."your gems and enchants don't matter at your progression level".

It's a lot different. I'm not sure how you don't see that. You can't change do anything personally to change class balance. You can fix your gemming, just as you can improve your play. That's the point I was making in the last post. Gemming incorrectly is no different from playing at 60% of your capability. There's a lot of overhead there that you can take advantage of to improve performance.

In other words, you can demonstrate that "DKs are better than paladins" is irrelevant at a certain skill level by showing that either class can tank the boss successfully with minimal differences in success rate. You cannot defend poor gemming or poor play in the same way, because there are large measurable differences between bad gems/play and correct gems/play. And the central point is that bad/correct play is a much more significant factor in week-13 progression than tank imbalances.

Fridmarr wrote:Honestly, by suggesting that class balance doesn't matter at lower progression levels, it's hard to make a case that much does matter at lower progression levels. It's all just inputs and outputs. The fact that improving play is the best way to increase their success is obvious, but it's not a mutually exclusive property. Just because playing better is going to grant the biggest benefit doesn't mean that class balance doesn't matter.

Rather than "doesn't matter," let's get specific. Because I'll agree that it's wrong to say that it's 100% irrelevant - obviously class imbalances will affect the outcome of fights, given two equally-skilled tanks under ideal, normalized conditions. But it's not the dominant effect at that play level, because nobody raids in that fantasy environment. In practice, several other factors (gear, player skill, coordination, and teamwork) make or break week-13 progression. Tank imbalances just aren't statistically significant when the variances on all of those other factors are much larger. When you're wiping to tank death once every 100 pulls, and the other 99 wipes are people standing in fire, not switching targets, or not putting out enough DPS, you don't worry about whether your tank is optimal. You worry about dealing with the big problems first. Dropping your tank-death probability from 1% to 0.5% helps, but not as much as dropping your standing-in-fire probability from 50% to 5%. Or, for that matter, by the tank playing better to reduce his "oops-I-forgot-to-cooldown/WoG/trinket" probability from 50% to 5%, or gemming properly to significantly reduce his burst death probability.

So the most accurate statement wouldn't be "tank balance doesn't matter." It would be "tank balance is a very, very small effect at your level of progression, and it is not a barrier that will prevent you from successfully tanking an encounter."
"Theck, Bringer of Numbers and Pounding Headaches," courtesy of Grehn|Skipjack.
MATLAB 5.x, Simcraft 6.x, Call to Arms 6.0, Talent Spec & Glyph Guide 5.x, Blog: Sacred Duty
User avatar
theckhd
Moderator
 
Posts: 7950
Joined: Thu Jul 31, 2008 3:06 pm
Location: Harrisburg, PA

Re: Paladins are the worst tanks?

Postby Fridmarr » Sat Mar 31, 2012 11:04 am

theckhd wrote:
Fridmarr wrote:Wipes simply don't tell you the information you need. They can tell you why you wiped, but that doesn't mean that without whatever caused that wipe, you would have succeeded so its very incomplete data. Even then, determining the cause of the wipe isn't always available in the logs with the detail you really need.

I disagree. They tell you why you wiped, which is why you didn't succeed. It may be one of several reasons you wouldn't have succeeded, but that's still information that needs to be dissected and incorporated back into your strategy. The feedback might be anywhere from "quit standing in fire" to "let's adjust the strategy so this isn't a concern," but either way that feedback loop matters. The fact that you need to deal with all of those issues doesn't make analyzing and dealing with one of them less relevant. You keep eliminating issues one at a time to increase your probability of success until you succeed.
I don't mean to suggest it doesn't matter for every purpose. It's vital information for a guild, but it doesn't show much pertaining whether or not class balance had much of an impact. For instance, you can't easily determine if that person who died in a fire was almost out and would have survived had someone managed to get another heal on him, or why no heals were able to get to him, was it slow reaction, movement, range, or lack of free GCDs.

theckhd wrote:That's awfully pessimistic. I think most people can just "play better," but they need to make the effort to do so (by practicing, doing their homework, etc.). There are very few cases where someone is inherently limited from improving. Even I see improvement in my play from week to week, both as I learn the encounters and as I try to correct errors or extinguish bad habits.
But it's also mostly true. Situational awareness, reaction time, finger twitch, and desire...those things are difficult to improve on after some amount of time playing. That's not to say that there won't be any improvement, but after awhile (and it really doesn't take too long in a game like this) they are starting to reach their capacity.

theckhd wrote:It's a lot different. I'm not sure how you don't see that. You can't change do anything personally to change class balance. You can fix your gemming, just as you can improve your play. That's the point I was making in the last post. Gemming incorrectly is no different from playing at 60% of your capability. There's a lot of overhead there that you can take advantage of to improve performance.

In other words, you can demonstrate that "DKs are better than paladins" is irrelevant at a certain skill level by showing that either class can tank the boss successfully with minimal differences in success rate. You cannot defend poor gemming or poor play in the same way, because there are large measurable differences between bad gems/play and correct gems/play. And the central point is that bad/correct play is a much more significant factor in week-13 progression than tank imbalances.
Guild's can (and often do) change things because of class balance. Using the best class for the job isn't much different than using the best gems. I'm not sure how you don't see that. Just stroll through the strategy section of the forums here, you'll see all kinds of tips on about different tanks performing different tasks on boss fights and the logic behind whatever subtle thing allows those tanks to perform the job simpler. Class balance can also be a large measurable difference sometimes much more so than improper gems.

theckhd wrote:Rather than "doesn't matter," let's get specific. Because I'll agree that it's wrong to say that it's 100% irrelevant - obviously class imbalances will affect the outcome of fights, given two equally-skilled tanks under ideal, normalized conditions. But it's not the dominant effect at that play level, because nobody raids in that fantasy environment. In practice, several other factors (gear, player skill, coordination, and teamwork) make or break week-13 progression. Tank imbalances just aren't statistically significant when the variances on all of those other factors are much larger. When you're wiping to tank death once every 100 pulls, and the other 99 wipes are people standing in fire, not switching targets, or not putting out enough DPS, you don't worry about whether your tank is optimal. You worry about dealing with the big problems first. Dropping your tank-death probability from 1% to 0.5% helps, but not as much as dropping your standing-in-fire probability from 50% to 5%. Or, for that matter, by the tank playing better to reduce his "oops-I-forgot-to-cooldown/WoG/trinket" probability from 50% to 5%, or gemming properly to significantly reduce his burst death probability.

So the most accurate statement wouldn't be "tank balance doesn't matter." It would be "tank balance is a very, very small effect at your level of progression, and it is not a barrier that will prevent you from successfully tanking an encounter."
I never said class balance was the dominant effect, just the opposite, I've repeatedly said that playing better is. I just think ignoring those issues of balance, which the comments I see often do is folly. That logic is often hidden behind the flawed notion that it only matters for top end guilds. That does create a serious inconsistency where folks regularly attempt maximize all sorts of aspects of their character to tiny degrees, only to hand wave over a balance issue that can be quite significant. I get the bias involved, particularly from long time pallys, but it's still flawed logic.

I think a lot of guilds have a few different tanks that are pretty equal in skill. Tanks taking more damage doesn't only manifest itself in tank death, healers able to focus entirely on a tank usually don't have much of a problem. But when the raid takes damage, and they have to move around, there are only so many heals to go around and often the tank is the priority, so raid members get sacrificed. Sure cleaning up the avoidable part of that is the best way to improve, but those are mistakes that are more consistently made as you stray from the bleeding edge level, so reducing damage intake anywhere in an encounter can end up being crucial, and the tank is no exception.
Fridmarr
Global Mod
 
Posts: 9667
Joined: Sun Apr 08, 2007 1:03 am

Re: Paladins are the worst tanks?

Postby theckhd » Sat Mar 31, 2012 12:34 pm

Fridmarr wrote:
theckhd wrote:It's a lot different. I'm not sure how you don't see that. You can't change do anything personally to change class balance. You can fix your gemming, just as you can improve your play. That's the point I was making in the last post. Gemming incorrectly is no different from playing at 60% of your capability. There's a lot of overhead there that you can take advantage of to improve performance.

In other words, you can demonstrate that "DKs are better than paladins" is irrelevant at a certain skill level by showing that either class can tank the boss successfully with minimal differences in success rate. You cannot defend poor gemming or poor play in the same way, because there are large measurable differences between bad gems/play and correct gems/play. And the central point is that bad/correct play is a much more significant factor in week-13 progression than tank imbalances.
Guild's can (and often do) change things because of class balance. Using the best class for the job isn't much different than using the best gems. I'm not sure how you don't see that. Just stroll through the strategy section of the forums here, you'll see all kinds of tips on about different tanks performing different tasks on boss fights and the logic behind whatever subtle thing allows those tanks to perform the job simpler. Class balance can also be a large measurable difference sometimes much more so than improper gems.


Sure, for guilds attempting it in week 1. Not for week 13. Show me a guild where tank balance makes a large, measurable difference in their ability to kill a boss in week 13, and I'll show you a guild that's making 100 other avoidable mistakes that would be more easily fixed than asking their main tank to reroll classes or swapping them out for another tank class of equal-or-lower personal skill.

Again, the point that you can't seem to refute is that in week 13, tank balance might make the difference in 1% or less of wipes. In week 1, it might be 10-15%, or higher. That's a big, important statistical difference that you're hand-waving away, and it's completely incorrect to do so.
"Theck, Bringer of Numbers and Pounding Headaches," courtesy of Grehn|Skipjack.
MATLAB 5.x, Simcraft 6.x, Call to Arms 6.0, Talent Spec & Glyph Guide 5.x, Blog: Sacred Duty
User avatar
theckhd
Moderator
 
Posts: 7950
Joined: Thu Jul 31, 2008 3:06 pm
Location: Harrisburg, PA

Re: Paladins are the worst tanks?

Postby Fridmarr » Sat Mar 31, 2012 1:08 pm

theckhd wrote:
Fridmarr wrote:
theckhd wrote:It's a lot different. I'm not sure how you don't see that. You can't change do anything personally to change class balance. You can fix your gemming, just as you can improve your play. That's the point I was making in the last post. Gemming incorrectly is no different from playing at 60% of your capability. There's a lot of overhead there that you can take advantage of to improve performance.

In other words, you can demonstrate that "DKs are better than paladins" is irrelevant at a certain skill level by showing that either class can tank the boss successfully with minimal differences in success rate. You cannot defend poor gemming or poor play in the same way, because there are large measurable differences between bad gems/play and correct gems/play. And the central point is that bad/correct play is a much more significant factor in week-13 progression than tank imbalances.
Guild's can (and often do) change things because of class balance. Using the best class for the job isn't much different than using the best gems. I'm not sure how you don't see that. Just stroll through the strategy section of the forums here, you'll see all kinds of tips on about different tanks performing different tasks on boss fights and the logic behind whatever subtle thing allows those tanks to perform the job simpler. Class balance can also be a large measurable difference sometimes much more so than improper gems.


Sure, for guilds attempting it in week 1. Not for week 13. Show me a guild where tank balance makes a large, measurable difference in their ability to kill a boss in week 13, and I'll show you a guild that's making 100 other avoidable mistakes that would be more easily fixed than asking their main tank to reroll classes or swapping them out for another tank class of equal-or-lower personal skill.

Again, the point that you can't seem to refute is that in week 13, tank balance might make the difference in 1% or less of wipes. In week 1, it might be 10-15%, or higher. That's a big, important statistical difference that you're hand-waving away, and it's completely incorrect to do so.
Unless there was a change, whatever the measurable difference was in week 1, probably still holds in week 13. And again, that equates to saying the value of gems, enchants, gear upgrades mean less in week 13. So, as you said, you show me a guild whose gems the tank has chosen is making a large measurable difference in week 13, and I'll show you a guild that's making 100 other avoidable mistakes that would be more easily fixed than asking their tank to change his gems.

The notion that swapping tanks is particularly onerous doesn't make much sense to me, since most guilds carry multiple tanks which one would assume are up to the task, it's one of the easiest changes to make. But more importantly, when discussing the situation with some random poster it's nearly impossible to know all the details that would go into that decision for some guild. So, present the information accurately and allow them to make the best choice for themselves. Their tanks may all be the same class, or they may indeed have a significant drop off in quality, then again they may have a tanking corps that's pretty competent and more concerned with getting the encounter down than who the MT is for it.

I have refuted that point, on several levels. First, to my knowledge no real numbers have been presented, but I also think the statistic of aggregate wipe percentages is poor in several different levels when specifically trying to tie it back to the effect of a certain tank on an encounter. As I said earlier a guild can have 100% of their wipes due to standing in a fire, but they may lack the DPS to defeat the encounter (the vampire thing in WotLK comes to mind), so it's not properly capturing the whole picture. Even on the flip side maybe to combat a healing deficiency on raid damage a guild messes around with changing their healing configuration (removing 1 from the tank) and the tank does die, but that's not necessarily reflective of much either even when it looks like a "tank death" wipe.
Fridmarr
Global Mod
 
Posts: 9667
Joined: Sun Apr 08, 2007 1:03 am

Re: Paladins are the worst tanks?

Postby theckhd » Sat Mar 31, 2012 2:04 pm

Fridmarr wrote:Unless there was a change, whatever the measurable difference was in week 1, probably still holds in week 13.

Except there is a significant change between week 1 and week 13. Gear. Your tanks have more health compared to boss hit size, making burst scenarios less frequent and less deadly. Your raid has more DPS and more health, easing healing requirements and potentially shortening dangerous periods that stress healing or tanks. The fact that the read has 10-15% more health means healers can be less precise with healing without putting the tank in danger. That's a huge, monumental change, and you're hand-waving it away. You can't. You simply can't.

Fridmarr wrote:And again, that equates to saying the value of gems, enchants, gear upgrades mean less in week 13. So, as you said, you show me a guild whose gems the tank has chosen is making a large measurable difference in week 13, and I'll show you a guild that's making 100 other avoidable mistakes that would be more easily fixed than asking their tank to change his gems.

I don't know why you're hooked on a tank with bad gems. Nobody said gemming was a huge effect, but it does make a difference, and often a bigger difference than tank choice. Our alt run tank was using parry/mastery gems everywhere a few weeks ago, and healers had a huge problem keeping her up. I helped her regem, and she's now got 30k more health and is much easier to heal. She died a lot less the last clear than she did in previous clears. And more importantly, that change was completely independent of class. That's why I brought it up - it's an example of something where your personal choices make as much or more difference than what class you're playing.

Fridmarr wrote:The notion that swapping tanks is particularly onerous doesn't make much sense to me, since most guilds carry multiple tanks which one would assume are up to the task, it's one of the easiest changes to make.

Most guilds do not carry multiple equally geared tanks, especially if they've been struggling into week 13. We have two paladin tanks that are well-geared. We have one DK that could potentially tank, but his gear isn't as advanced (no 4-piece, still a few T12 items iirc). We have a warrior that has a decent tanking set too, I think. But we don't swap tanks for any fights, because that gear differential is a bigger difference than tank balance on any fight. Not to mention practice - the two "usual" tanks are far more familiar with the fights, which means less wipes to learning timings and getting used to responding to encounter cues.

Fridmarr wrote:I have refuted that point, on several levels. First, to my knowledge no real numbers have been presented, but I also think the statistic of aggregate wipe percentages is poor in several different levels when specifically trying to tie it back to the effect of a certain tank on an encounter. As I said earlier a guild can have 100% of their wipes due to standing in a fire, but they may lack the DPS to defeat the encounter (the vampire thing in WotLK comes to mind), so it's not properly capturing the whole picture.

No, you haven't, at least not to the satisfaction of anybody who knows anything about progression content. While no real numbers have been presented, I can immediately tell you where you're going wrong in your example. If a guild has 100% of their wipes due to standing in a fire, then you don't even worry about DPS. You fix the problem that's causing 100% of your fucking wipes first, then you worry about optimizing DPS to meet enrages. Having the DPS to beat enrage doesn't mean shit if every single one of your wipes is due to mistakes. You correct the problem that exists first, not the one that might exist if the stars align.
"Theck, Bringer of Numbers and Pounding Headaches," courtesy of Grehn|Skipjack.
MATLAB 5.x, Simcraft 6.x, Call to Arms 6.0, Talent Spec & Glyph Guide 5.x, Blog: Sacred Duty
User avatar
theckhd
Moderator
 
Posts: 7950
Joined: Thu Jul 31, 2008 3:06 pm
Location: Harrisburg, PA

Re: Paladins are the worst tanks?

Postby Meloree » Sat Mar 31, 2012 2:20 pm

Fridmarr wrote:The reason why wipes don't matter is because they don't include a crystal ball which somehow allows the fight to continue to show you all the ways you would have wiped. If someone blows up the raid and causes a wipe, it hides the fact that you weren't going to beat the enrage timer anyhow, so to somehow statistically lump that in with "people doing stupid things", and not "dps was too low" yields the misleading result that is allowing you to reach your false conclusion. Aggregating those wipes into categories and then extrapolating from that is only compounding the problem.


Let's start over. In the hypothetical "week 13 guild faces the same boss that the week 1 guild does, but they do it with a higher average ilvl and L2R videos to guide them through the strategy" situation, the week 13 guild will face different challenges. For one, not standing in fire will be less of a concern, because with higher average ilvl comes higher average stamina. Running out of mana will be less of a concern, because with higher average ilvl comes higher regen and better raw throughput per spell. And, frankly, the healers in the week 13 guild probably cast less often on average. The tanks will have higher avoidance, higher armor, and more stamina - they will be naturally less burstable, and that's the cause of every tank death in heroic content.

Here's a couple of other assumptions:
1)It's hard to play at 100% of your gear's potential, but it's easy to play at 80%. The 80/20 rule, if you will. The week 13 guild, with it's 15% higher potential output has a very similar actual output - but much higher survivability.
2)Skill variance will be higher in the week 13 guild. In the week 1 guild, you can expect all players to perform in the 95-100% range for their class and gear. In the week 13 guild, you can probably expect 70-100%.

Therefore, class balance becomes severely deprecated after leaving the bleeding edge, and player skill becomes dominant. As far as I know, nobody disagrees so far.

If you have two equally skilled and geared tanks in this hypothetical, you would prefer to take the better class to any fight, week 1 or week 13. Every single wipe saved is faster progression. However, if you're talking about a 50 wipe fight, and tank death and/or raid wipes due to attrition from healers prioritizing tanks are an insignificant fraction of those wipes, it's not something that's really worth of attention.

Why? Because raids do come with crystal balls. We call them Raid Leaders, and it's their job to analyze wipes and progression. It's their job to know if you were on pace for beating enrage or not - even during a wipe. BQL and Zon'ozz might make it harder, with increasing RDPS as the fight continues, but it's 100% predictable on 100% of pulls. It's the raid leader's job to know when "tank death" is due to "dumb healer" and when it's due to "tank sucks", etc. And, quite frankly, any raid leader that worries about class balance in a week 13 guild is probably doing it wrong, because all they need to care about is absolute performance metrics. It's all that a raid leader cares about in a week 1 guild, too. What does it take to beat the boss? It's just that player skill is the real dominant factor in a week 13 guild, and those raid leaders that run shitty mages when they have amazing boomkins are probably just bad raid leaders.

So, yes - if the difference between two tank classes in a week 13 guild is one wipe out of 50, then even "approximately equal skill" is probably dominated by which tank is drinking Red Bull, and which one is drinking beer. And, honestly, with a tier's worth of extra gear... that's about all the difference there is.
Meloree
Maintankadonor
 
Posts: 1420
Joined: Wed Mar 12, 2008 10:15 am

Re: Paladins are the worst tanks?

Postby benebarba » Sat Mar 31, 2012 3:20 pm

Meloree wrote:<snip> which tank is drinking Red Bull, and which one is drinking beer. And, honestly, with a tier's worth of extra gear... that's about all the difference there is.


Oh man, now I gotta spec beer instead of whiskey. Thanks Meloree! :evil:

:P
benebarba
 
Posts: 2469
Joined: Tue Oct 19, 2010 7:30 am

Re: Paladins are the worst tanks?

Postby Fridmarr » Sat Mar 31, 2012 3:22 pm

theckhd wrote:
Fridmarr wrote:Unless there was a change, whatever the measurable difference was in week 1, probably still holds in week 13.

Except there is a significant change between week 1 and week 13. Gear. Your tanks have more health compared to boss hit size, making burst scenarios less frequent and less deadly. Your raid has more DPS and more health, easing healing requirements and potentially shortening dangerous periods that stress healing or tanks. The fact that the read has 10-15% more health means healers can be less precise with healing without putting the tank in danger. That's a huge, monumental change, and you're hand-waving it away. You can't. You simply can't.
I'm not. I disagree that the raid in week 13 that has yet to clear the boss is putting out much more DPS than the raid did that killed it in week 1. Their lack of skill has corrected for that. Along with DPS, they also have the same problem when it comes to healing, bad choices, slower reaction time, members standing in fire etc again skill being their handicap to make them less effective. That skill offsets a lot of their gear gain, otherwise they would have already beaten the encounter because their sum output would have been greater than the week 1 guild's output. I haven't hand waved it off, I've gone over this many times.

theckhd wrote:I don't know why you're hooked on a tank with bad gems. Nobody said gemming was a huge effect, but it does make a difference, and often a bigger difference than tank choice. Our alt run tank was using parry/mastery gems everywhere a few weeks ago, and healers had a huge problem keeping her up. I helped her regem, and she's now got 30k more health and is much easier to heal. She died a lot less the last clear than she did in previous clears. And more importantly, that change was completely independent of class. That's why I brought it up - it's an example of something where your personal choices make as much or more difference than what class you're playing.
It's not gems specifically, they are just a good example. We pour over the minutia of it all, and then say class balance doesn't matter. It's not consistent. I don't care how you gemmed, you weren't going to outdo the DK's advantages on 3 drakes what's his face, or the pallys blocking advantage on Anub's adds.

theckhd wrote:Most guilds do not carry multiple equally geared tanks, especially if they've been struggling into week 13. We have two paladin tanks that are well-geared. We have one DK that could potentially tank, but his gear isn't as advanced (no 4-piece, still a few T12 items iirc). We have a warrior that has a decent tanking set too, I think. But we don't swap tanks for any fights, because that gear differential is a bigger difference than tank balance on any fight. Not to mention practice - the two "usual" tanks are far more familiar with the fights, which means less wipes to learning timings and getting used to responding to encounter cues.
Gear is certainly a consideration, if the other tank isn't nearly as well geared then it may not be worth it. A lot of guilds have pretty equally geared tanks though if they are regulars. Not precisely the same, but an item or two is generally pretty minor, unless there's a set bonus at stake or something. Regardless, it's a choice for that guild to make.

theckhd wrote:No, you haven't, at least not to the satisfaction of anybody who knows anything about progression content. While no real numbers have been presented, I can immediately tell you where you're going wrong in your example. If a guild has 100% of their wipes due to standing in a fire, then you don't even worry about DPS. You fix the problem that's causing 100% of your fucking wipes first, then you worry about optimizing DPS to meet enrages. Having the DPS to beat enrage doesn't mean shit if every single one of your wipes is due to mistakes. You correct the problem that exists first, not the one that might exist if the stars align.
First, you need to chill out a bit. I absolutely agree that you deal with the problem that presents itself. However that's not relevant to the comparison beceause the logs won't really show that, which is what my point is.

Meloree wrote:Why? Because raids do come with crystal balls. We call them Raid Leaders, and it's their job to analyze wipes and progression. It's their job to know if you were on pace for beating enrage or not - even during a wipe. BQL and Zon'ozz might make it harder, with increasing RDPS as the fight continues, but it's 100% predictable on 100% of pulls. It's the raid leader's job to know when "tank death" is due to "dumb healer" and when it's due to "tank sucks", etc. And, quite frankly, any raid leader that worries about class balance in a week 13 guild is probably doing it wrong, because all they need to care about is absolute performance metrics. It's all that a raid leader cares about in a week 1 guild, too. What does it take to beat the boss? It's just that player skill is the real dominant factor in a week 13 guild, and those raid leaders that run shitty mages when they have amazing boomkins are probably just bad raid leaders.
The problem is, what is the raid leader supposed to do when the metrics tell him that his mates are putting out their potential? Their skill is what it is, it's not going to change much. Certainly practice handling the encounter will cause improvement in dealing with those mechanics, but you aren't going to get much more output out of them. In my experience, raid damage overwhelming healers has been a pretty significant cause of wipes in heroics, frequently a solid portion of it is avoidable, but frequently these guilds will have to deal with it anyhow because they are more prone to those sorts of mistakes. If as a raid leader of these guilds (which I have been) if I were to ignore the fact that another tank provides some sort of benefit, while expecting my raid to play better, that would have been a pretty tough sell. You do your best to maximize what you can, and that includes using the best class when that's an option.
Fridmarr
Global Mod
 
Posts: 9667
Joined: Sun Apr 08, 2007 1:03 am

Re: Paladins are the worst tanks?

Postby Weebey » Sat Mar 31, 2012 6:02 pm

Some of the "facts" in the following are more like stylized guesses; I hope no one who actually knows something about this subject takes too much offense.

Suppose you want to explain how much genetic variance explains the variance in male peak adult height. If you start with, say, all human males over the last 2,000 years, the answer you will get is, presumably, not very much. While there are real genetic differences, of course, they are swamped by differences in diet, nutrition, health, maternal health, and a host of other environmental factors. If you want to have a tall adult--say, two standard deviations from the population mean--good genes are nice, but scoring well in those factors is much more important.

Now, instead of looking at all males, you restrict yourself to "affluent white American males born after 1970." Many of the environmental factors that were so critical in the larger population are going to be levelled out in this subpopulation; while the genetic variation is of course also lower, it is proportionally a much, much higher fraction of the population variance. If you want to pick a 2 sigma adult from this population, you had damn well better pick a guy with "good" height genes.

The analogy is far from perfect, for a number of reasons I've already thought of (and likely more that I haven't), but I think it captures the basic force of the argument in a way that avoids getting too bogged down in the specifics of the situation.
Weebey
 
Posts: 31
Joined: Mon Mar 07, 2011 2:02 pm

Re: Paladins are the worst tanks?

Postby Fridmarr » Sat Mar 31, 2012 7:39 pm

Weebey wrote:Some of the "facts" in the following are more like stylized guesses; I hope no one who actually knows something about this subject takes too much offense.

Suppose you want to explain how much genetic variance explains the variance in male peak adult height. If you start with, say, all human males over the last 2,000 years, the answer you will get is, presumably, not very much. While there are real genetic differences, of course, they are swamped by differences in diet, nutrition, health, maternal health, and a host of other environmental factors. If you want to have a tall adult--say, two standard deviations from the population mean--good genes are nice, but scoring well in those factors is much more important.

Now, instead of looking at all males, you restrict yourself to "affluent white American males born after 1970." Many of the environmental factors that were so critical in the larger population are going to be levelled out in this subpopulation; while the genetic variation is of course also lower, it is proportionally a much, much higher fraction of the population variance. If you want to pick a 2 sigma adult from this population, you had damn well better pick a guy with "good" height genes.

The analogy is far from perfect, for a number of reasons I've already thought of (and likely more that I haven't), but I think it captures the basic force of the argument in a way that avoids getting too bogged down in the specifics of the situation.

LOL, that's why you've got to love Maintankadin.

But I need to add some additional parameters, which just may bog us down in details....

When you are picking from the smaller post 1970s group, you know that diet, nutrition, health, maternal health, and all of the other environmental factors are always about as good as can be, but when picking from the larger population, only diet averages better than the 1970s group (too much preservatives in modern food ;) ) the rest average out lower. Also, the height you are trying to reach is a fixed value, not relative.
Fridmarr
Global Mod
 
Posts: 9667
Joined: Sun Apr 08, 2007 1:03 am

Re: Paladins are the worst tanks?

Postby gronc2 » Sat Mar 31, 2012 8:55 pm

sorry everyone, didn't mean to re-post something, nor stir up shit.

I did appreciate the input, thank you

gronc
gronc2
 
Posts: 52
Joined: Thu Feb 16, 2012 8:59 am

Re: Paladins are the worst tanks?

Postby Shoju » Mon Apr 02, 2012 10:58 am

I have read the arguments / points made by both sides, and I'm firmly in Theck's camp, probably because my guild is dealing with similar issues right now. on H:Ultra, we (were) a little short on the dps to beat the enrage (with the newest debuff, we probably have the dps). There were people who wanted to stop working on it, when our longest attempt showed we were light on DPS.

Even though, it was only 2 minutes into the fight total, because people fail at the button boss. They wanted to stop working on it, or swap people out. Neither of those solve the reason for the wipe. They solved a reason that we "maybe" would have wiped, had we not failed at what killed us.

We are currently working out some issues on H:Yor, that I'm fairly confident would for the most part go away if we made some changes. But, Even those changes don't fix the real problem.

Practice, playing better at your class are far more important to the non bleeding edge than class balance. Especially when there is room to improve.
User avatar
Shoju
 
Posts: 6362
Joined: Mon May 19, 2008 7:15 am

PreviousNext

Return to General

Who is online

Users browsing this forum: Google [Bot], KysenMurrin, Worldie and 1 guest

Who is online

In total there are 4 users online :: 3 registered, 0 hidden and 1 guest (based on users active over the past 5 minutes)
Most users ever online was 380 on Tue Oct 14, 2008 6:28 pm

Users browsing this forum: Google [Bot], KysenMurrin, Worldie and 1 guest