Remove Advertisements

Ulduar and the Paladin Tank

Moderators: Fridmarr, Worldie, Aergis, guillex

Re: Re:

Postby Rasmfrackn » Mon Mar 23, 2009 2:58 pm

skolor wrote:
Kelaan wrote:For a 10-man team, who might not HAVE any classes that can tank other than some paladins (or any other composition)


Seriously? You're trying to argue "What if we only have 1 class that can tank in a sample of 10 people"? If you take a random sampling of 10 people, assuming all classes have equal makeup, then in 98% of all cases you will have two different tanking classes. When you factor in that most guilds usually run with ~20% extra people for fill-ins and the like, you're covering just about 99% of all situations.

No. Don't even try that. Its like arguing that no boss or trash can use a curse, because only a couple of classes can dispell it (even fewer than can tank). As long as its only one tank type that is being disadvantaged, you're covering 98.7% of all situations just by pulling 12 random people out of Dalaran (and gearing them). They shouldn't be tuning fights for the 1% of groups that will be at a serious disadvantage because they chose to recruit with a huge skew a certain way, any more than they should be tuning fights to require a certain class (which is considerably more likely given a random sampling of people).


I'm not sure how you got your numbers, but even picking at random I don't think 98% is right. Are you assuming if there are ANY paladins/warriors/dks/druids that they're going to be able to fill the tank role? What if we pick randomly from all 30 specs, and see how many tanks we expect to end up with? That's still not realistic, but it's probably closer. They had the same issue with replenishment... 3 specs out of 30 (instead of 4 tank specs out of 30) and they deemed it insufficient so they gave it to 2 more specs.

And remember that "random" is streaky... you're usually much more likely to have something other than the mean than exactly the mean. Most 10-mans will not be 1 of every class. (Unless of course you're hand-picking for class balance ahead of any other criteria. This is not truly a random process we're talking about here.)

-- Anecdotally: my guild has 3 tankadins, 1 DK, and 1 prot warrior. It's not too hard to end up with a 10-man that has only tankadins.
Rasmfrackn
Dwarf Paladin
Icecrown Server
Eng/Scribe/Masochist
User avatar
Rasmfrackn
 
Posts: 1014
Joined: Fri Aug 17, 2007 12:11 pm

Re: Ulduar and the Paladin Tank

Postby Panzerdin » Mon Mar 23, 2009 3:10 pm

His maths is correct. The chance of having no non-Paladin tanking classes is 7/10 (the chance that a random class is either unable to tank, or is a Paladin) to the power 10, which is equivalent to 282475249/10000000000. This gives a 0.028247525 chance, or 2.82%.

Obviously, if you refuse to have your other classes respec to do the fight, then you're screwed. But that's your problem.
User avatar
Panzerdin
 
Posts: 5504
Joined: Sat May 24, 2008 2:51 pm
Location: On a picket line, protesting against the changes to Maintankadin

Re: Ulduar and the Paladin Tank

Postby KysenMurrin » Mon Mar 23, 2009 3:14 pm

So raids should be balanced around a raid group that not only has all classes available, but also has all those members ready to swap specs and roles on an encounter-by-encounter basis?
I don't play WoW any more.
Donnan - Nangun - Kysen - Kysen - Mardun - Timkins

Mostly-Book Blog.
KysenMurrin
 
Posts: 6810
Joined: Thu Jun 26, 2008 6:37 am
Location: UK

Re: Ulduar and the Paladin Tank

Postby Panzerdin » Mon Mar 23, 2009 3:20 pm

Yes. Raids should be balanced on the assumption that you have more than one class able to fulfil a given role, and that the classes you have are prepared to do so.
User avatar
Panzerdin
 
Posts: 5504
Joined: Sat May 24, 2008 2:51 pm
Location: On a picket line, protesting against the changes to Maintankadin

Re: Ulduar and the Paladin Tank

Postby Conaan! » Mon Mar 23, 2009 3:51 pm

Panzerdin wrote:His maths is correct. The chance of having no non-Paladin tanking classes is 7/10 (the chance that a random class is either unable to tank, or is a Paladin) to the power 10, which is equivalent to 282475249/10000000000. This gives a 0.028247525 chance, or 2.82%.

Obviously, if you refuse to have your other classes respec to do the fight, then you're screwed. But that's your problem.



the problem with your math is the fact that you put equal weight on all classes, because different servers have different populations and only certain people play during certain times, the population of certain classes fluctuates during certain time periods, because of this 2.82% chance is a skewed number

then of course, you look at what blizz stated, something along the lines of "we want you to be able to pick among the 4 tanks, who to bring", damn i just dont remember "we want to you be able to pick among the 3 tanks and sometimes among that one wannabe, who to bring"

its not hard to make encounters work with all 4 tanks, there are no reasons to have just 1 boss we are disadvantaged on
Image
Lore wrote:JEAPORDY: CREEPY BASEMENT EDITION
User avatar
Conaan!
 
Posts: 1271
Joined: Wed May 14, 2008 3:22 pm

Re: Ulduar and the Paladin Tank

Postby PsiVen » Mon Mar 23, 2009 3:53 pm

It doesn't make any sense to assume that a progression raid will have anyone beyond the 2 tanks they brought in terms of people with adequate gear to tank the boss. Therefore, there are 2 slots and 4 classes that can fit in each. Assuming they are all equally popular, 1/16 of the raids then use two paladins for their tanks and would be forced to dig deep and have someone tank with their off set. 3/16 of the raids have a paladin MT who would be forced into an OT position at best. In serious progression, 3/16 have a paladin OT who would be displaced entirely because a druid would be ideal to OT this fight instead. That adds up to 43.75% of all 10-man groups inconvenienced by such a mechanic. Not happening.

SA functioning properly on this fight will make this argument entirely moot.
Gladiator Psiven, 90 Tankadin
90 Druid, 90 Mage, 85 Monk, 85 DK, 70 War, 70 Pal, 60 Priest, 60 Lock, 64 Rogue
Longtime addict of Space - Glory Through Conquest
User avatar
PsiVen
Moderator
 
Posts: 4363
Joined: Fri Jun 01, 2007 5:28 pm
Location: On a Boat

Re: Ulduar and the Paladin Tank

Postby Karnah » Mon Mar 23, 2009 8:43 pm

Which, while currently untestable, seems to be the direction they want to go. When the other spells mentioned were used, as an aside, they remained to their full effect, much to the glee of our hunter, as they get the best deal in this fight.

As for auriaya, i think the fear is intended to be a horror, as some effects could/could not break it last time round. If its something you have to eat, then one can only assume they would do their best to equalize its effect on everyone.
Image
Karnah
 
Posts: 48
Joined: Tue Dec 09, 2008 8:14 pm
Location: Middlesbrough, England

Re: Ulduar and the Paladin Tank

Postby skolor » Mon Mar 23, 2009 8:58 pm

Conaan! wrote:the problem with your math is the fact that you put equal weight on all classes, because different servers have different populations and only certain people play during certain times, the population of certain classes fluctuates during certain time periods, because of this 2.82% chance is a skewed number


Which, you'll notice, I mentioned, specifically that was an assumption.

Panzerdin wrote:Obviously, if you refuse to have your other classes respec to do the fight, then you're screwed. But that's your problem.


This would be the point of dual specs, to make this kind of situation more viable.

PsiVen wrote:It doesn't make any sense to assume that a progression raid will have anyone beyond the 2 tanks they brought in terms of people with adequate gear to tank the boss. Therefore, there are 2 slots and 4 classes that can fit in each. Assuming they are all equally popular, 1/16 of the raids then use two paladins for their tanks and would be forced to dig deep and have someone tank with their off set. 3/16 of the raids have a paladin MT who would be forced into an OT position at best. In serious progression, 3/16 have a paladin OT who would be displaced entirely because a druid would be ideal to OT this fight instead. That adds up to 43.75% of all 10-man groups inconvenienced by such a mechanic. Not happening.

SA functioning properly on this fight will make this argument entirely moot.


Ignoring the SA argument; I know nothing about the fights, and was reading this to get an idea on how they would work. My issue is with bad maths being thrown around.

As a note with your math, its rather faulty to from "Dig deep and have someone tank with their off set" to pulling in a druid in an offset as an offtank.

I also see this as an intentional thing. Dual specs are in the game for a reason. I hope they do more things like this, encouraging 10 man groups to have, say, 3-4 tanks, 4-5 healers and 8-9 DPS. Too many problems are caused because someone doesn't know the other side of the fight, and thinks the tank, or a healer, or a dps could be doing something differently without having been in that situation. Will it magically make things work better? No, but it will help my healers understand why the DPS does things a certain way, and help some of the DPS understand why the tank does the fight a certain way.

Again, I haven't seen all the fights, and don't know much more than the basics, but it seems very surprising that Paladins are the only classes at a disadvantage on fights. Especially with many of the hard modes (as far as I know) untested, it doesn't seem unreasonable that there will be a fight where a DK/Druid is disadvantaged.
My blog of my experiences leveling a tankadin.
skolor
 
Posts: 177
Joined: Thu Nov 29, 2007 8:56 am

Re: Ulduar and the Paladin Tank

Postby mconeone » Tue Mar 24, 2009 5:46 am

I also see this as an intentional thing. Dual specs are in the game for a reason. I hope they do more things like this, encouraging 10 man groups to have, say, 3-4 tanks, 4-5 healers and 8-9 DPS.


They specifically said they wouldn't be doing this. It lessens the value of the 4 pure DPS classes.
Image
mconeone
 
Posts: 245
Joined: Tue Sep 11, 2007 8:42 am

Re: Ulduar and the Paladin Tank

Postby Senador » Tue Mar 24, 2009 9:44 am

Yes. Raids should be balanced on the assumption that you have more than one class able to fulfil a given role, and that the classes you have are prepared to do so.


I kind of question this...

I mean, they did give Protection Paladins a single target taunt specifically to reduce these kind of issues (Most notably against 4 horsemen; again you COULD work around it, or you could just bring a different tank). I find it odd they would do a precedent like that, then go against it so quickly.
Senador
 
Posts: 453
Joined: Wed Aug 06, 2008 10:55 am

Re: Ulduar and the Paladin Tank

Postby Kelaan » Tue Mar 24, 2009 10:01 am

skolor wrote:Seriously? You're trying to argue "What if we only have 1 class that can tank in a sample of 10 people"? If you take a random sampling of 10 people, assuming all classes have equal makeup, then in 98% of all cases you will have two different tanking classes.


That doesn't make it any better for those small number of raid teams. We each can imagine a 10-man raid team without a warlock, or without a warrior, or rogue, or priests. That variance (pick 10 people from 30 specs) was the entire reason raid buffing synergy was spread across more classes. Also, the 2% number is only across all possible combinations, isn't it? It's not across the raid teams we actually see. No one has all-warlock raids or all-warrior raids, but those would be among the possible combinations of classes.

Panzerdin wrote:Obviously, if you refuse to have your other classes respec to do the fight, then you're screwed. But that's your problem.


The trouble is not refusing to re-gear, it's that some people may not want to play that spec, or may not be GEARED for that spec. E.g., I currently could not contribute well as DPS, as I don't have offspec ret gear. Similarly, we have several DKs (and even bears), but they either do not WANT to tank, or do not have a tanking set suitable to heroics, let alone ilevel 213 encounters.

Blizzard has probably on the order of a million raiders, given that 10-mans are the new market. If we use what I consider to be a conservative SWAG, and claim that they have ~100000 people that are part of 10-man raid teams, 2% is still 2000 people, or ~200 raid teams. Yes, it's a minority, but a mechanic that screws them for not having the right group of friends is a poor design decision. We saw fixes to 4Horsemen already for raid teams with two paladin tanks, after all. (And as a result, I don't think Blizzard is intending to make paladin tanks a Bad Idea on this fight -- something I think we all agree on.)

For the average 10-man raid team, does Blizzard want them to be blocked from progress in new content because they either haven't fully geared off-specs (and thus need to keep farming Naxx) or recruit different people? I don't think replacing 10 or 20% of your raid team is an acceptible solution to making a non-hard-mode encounter possible. Now, for protodrakes? Sure, I guess so. But I don't think your average raid team should need to deal with making sure they have a tank of each type. What if they had an encounter that heavily punished rage tanks, rather than mana-using tanks? The raid teams without a geared DK or Paladin would be screwed. I argue that that goes against Blizzard's design goals, which is "bring the person not the class".

Panzerdin wrote:Yes. Raids should be balanced on the assumption that you have more than one class able to fulfil a given role, and that the classes you have are prepared to do so.

Hard modes should be, but I don't think normal ones should be.

Are all your druids geared enough to heal, DPS or tank? I don't believe most of our trees have good tanking gear (though they could go boom), and I don't think that our cats have good healing gear, for the most part.
Are all your paladins well geared for DPS, tanking AND healing? Ours aren't. We don't have enough farmage of current content to have enough gear to go around for competitive offspecs, in many cases. My healing gear is mostly good, but I have nearly zero DPS pieces from Naxx, for example.
User avatar
Kelaan
 
Posts: 4036
Joined: Thu Jan 03, 2008 12:01 pm

Re: Ulduar and the Paladin Tank

Postby Belloc » Tue Mar 24, 2009 10:23 am

Applying math to this situation is bullshit when the vast majority of tanks currently in the game are warriors to begin with. Math doesn't apply.

All tanks are supposed to be able to tank all content -- that is the design goal. So, it doesn't matter if you bring two paladins or two druids, they need to be able to tank. End of story.
User avatar
Belloc
 
Posts: 3195
Joined: Fri Jul 18, 2008 1:56 pm
Location: Silent Earth

Re: Ulduar and the Paladin Tank

Postby Jasari » Tue Mar 24, 2009 10:38 am

Belloc wrote:Applying math to this situation is bullshit when the vast majority of tanks currently in the game are warriors to begin with. Math doesn't apply.

All tanks are supposed to be able to tank all content -- that is the design goal. So, it doesn't matter if you bring two paladins or two druids, they need to be able to tank. End of story.


I've currently cleared all 10 man content using 2 Paladin tanks, including 4 horseman before we got our single target taunt. The only exception is 3D Sarth where we bring 3 tanks anyway. With our strat, we don't even have the MT (a druid in our case) use any of their CDs except as a last resort if a breath is incoming and someone is out of position, dead, or otherwise unable to use their assigned CD. With the changes in 3.1, I don't see any indication that we won't be able to clear Ulduar with 2 Paladin tanks.
User avatar
Jasari
 
Posts: 2195
Joined: Sun Apr 27, 2008 12:32 am
Location: Boston, MA

Re: Ulduar and the Paladin Tank

Postby skolor » Tue Mar 24, 2009 11:14 am

mconeone wrote:
I also see this as an intentional thing. Dual specs are in the game for a reason. I hope they do more things like this, encouraging 10 man groups to have, say, 3-4 tanks, 4-5 healers and 8-9 DPS.


They specifically said they wouldn't be doing this. It lessens the value of the 4 pure DPS classes.


That has no effect on the pure DPS classes. They are in a raid to DPS, and they get a marginal boost to their dps to cement that. On the other hand, hybrids can, and with dual specs should be prepared to do more than one thing, especially in a 10 man situation. Does that mean every paladin should be carrying around 3 (or 4 or 5) sets of gear so they are prepared for every possible encounter? No, what they should be doing is having a main spec and an offspec, and be prepared to switch between them as a fight demands. The offspec should be by no means on the same level as their main spec, but I don't see why a paladin going into Ulduar not able to have a healing off set made of 50-60% epics. If you've been clearing at least the 4 wings of naxx since the end of January, you're looking at 6 weeks of farming, which seems like more than enough to be scraping together a handful worth of offset pieces.

Kelaan wrote:
skolor wrote:Seriously? You're trying to argue "What if we only have 1 class that can tank in a sample of 10 people"? If you take a random sampling of 10 people, assuming all classes have equal makeup, then in 98% of all cases you will have two different tanking classes.


That doesn't make it any better for those small number of raid teams. We each can imagine a 10-man raid team without a warlock, or without a warrior, or rogue, or priests. That variance (pick 10 people from 30 specs) was the entire reason raid buffing synergy was spread across more classes. Also, the 2% number is only across all possible combinations, isn't it? It's not across the raid teams we actually see. No one has all-warlock raids or all-warrior raids, but those would be among the possible combinations of classes.


So, looking at the math here (Going by random samplings of 10 people, not a raid composition):

Only one class that can tank - 2.8% chance
Chance you don't have a XXXXX (for any given class) - 75.2%


And, since there have been so many objections to me just using a random sampling of 10 people, rather than an actual raid makeup, here's the math on that. Some assumptions: All classes have a minimum amount of gear to perform in whatever capacity they are needed for. This likely means at least Quest blues for that spec, with a couple of epics/emblem pieces. It also assumes there is an exact even distribution between the classes. Yes, this is inaccurate, and the one server that doesn't have a single druid is going to get screwed. This also assumes a raid composition of 2xTanks, 3xHealers, 5xDPS. No differentiation is made between ranged and melee dps. So, the maths for only having only class that can tank:

To start with, all assumptions require both tanks to be the same. Since each tank class has to be calculated individually, there is a 6.25% chance of this happening for each class (Meaning 6.25% of groups have warrior-warrior tank teams, 6.25% druid-druid, ...). That means 1 in 16 raids have two of the same tanks. Now, to compute the rest of things, we have to do two parts seperately, the chance that warrior/DK is the only class available to tank, and Druid/Paladin is the only class to tank.

Warrior/DK:
Probability both tanks are Warriors : 6.25%
Probability all DPS is NOT a DK/Druid/Paladin: 17.8%
Probability all healers are NOT a DK/Druid/Paladin: 12.5%

Total Probability that a Warrior is the only tanking class: .14%*2 (because the math is the same for both DKs and Warriors) = 0.28% chance a DK/Warrior is the only class capable of tanking.

Druid/Paladin:
Probability both tanks are Druids: 6.25%
Probability all DPS is NOT a Warrior/DK/Paladin: 17.8%
Probability all healers are NOT a Warrior/DK/Paladin: 42.2%

Total Probaility that a Druid is the only Tanking class: .47%*2 = .94% chance only a Druid/Paladin can tank.

Total Probability that only a DK/Warrior/Druid/Paladin can tank: 1.22%

When you say "You're discarding all Warrior-only raid compositions" you aren't considering that, given a random sampling of players, only occurs 1 in every 10,000,000 samples.

There are 52,214 guilds on WoWJutsu. Assuming every single one of them has exactly 10 people in it, then there are 637 of them. Some back of the envelope math based on their percentages of guilds that have cleared Naxx 10 and Naxx 25 says that ~80% of those guilds do both 10 and 25 mans, so they aren't applicable here. Meaning, out of all 52,214 guilds that are raiding, 127 of them fall into a situation where only 1 class in their group can tank. That's .2%. There is not a single reason why every fight needs to be balanced for .2% of guilds.


We're not even talking about a fight that is undo-able for .2% of raiding guilds, we're talking about fights that are significantly harder for .2% of guilds. And even that figure assumes that there are only 10 people in the 10 man groups. Most of them are going to have 12-15 people, and some of those people will have level 80 alts they could, if needed, get on to tank a boss. It won't be the best situation, and they'll be undergeared for it, but saying that for ~6% of situations someone needs to have a Tanking offset that is passable for your healers to keep up, and in .2% of situations someone will have to get on their alt for a fight does not seem ridiculous in the slightest.

Belloc wrote:Applying math to this situation is bullshit when the vast majority of tanks currently in the game are warriors to begin with. Math doesn't apply.

All tanks are supposed to be able to tank all content -- that is the design goal. So, it doesn't matter if you bring two paladins or two druids, they need to be able to tank. End of story.


And using hyperbole to assume that an acknowledged bug which makes threat generation for paladins very difficult is bullshit. We're not talking about a paladin with no mana like is on the PTR right now. We're talking about a paladin with non-infinite mana, unlike our usual. While its nothing more than conjecture, I would say that they probably will change the fight so that paladins get enough mana back to use their usual rotation minus consecration.
My blog of my experiences leveling a tankadin.
skolor
 
Posts: 177
Joined: Thu Nov 29, 2007 8:56 am

Re: Ulduar and the Paladin Tank

Postby Kelaan » Tue Mar 24, 2009 2:12 pm

skolor wrote:So, looking at the math here (Going by random samplings of 10 people, not a raid composition).... {low percentages}
....
All classes have a minimum amount of gear to perform in whatever capacity they are needed for. This likely means at least Quest blues for that spec, with a couple of epics/emblem pieces.


For entry level 10-man raiding, sure -- some dungeon stuff and some badge gear is sufficient. I do NOT think that is sufficient to "require" those people people to get in order to be able to hit higher levels of content (tier 8). Sure, many PUG naxx10 now, but some do not. Objectively speaking, it's a weaker design if it has mechanics which specifically exclude certain makeups, even if they are a very small minority. I do NOT think that pre-Naxx tanking gear will be sufficient for the average 10-man raid team to have a good time in Ulduar, which is basically what some off-spec people will have.

As a more concrete example, I have nearly zero dps pieces, and the only healing ones I have are ones I got before our paladin healer got to 80 -- I'll not be a competitive healer or DPS player in Ulduar, at ALL. Tank loot distribution has been similar in our guild, though it's somewhat more concentrated as only a few DPS players are interested in tanking. We have TWO people (out of ~30) whose off-spec tanking gear is sufficient for tanking even Sartharion10 or Malygos. It's not at all been a goal for those people to get tanking gear, and thus they haven't pursued it. People spent their badges on primary role upgrades.

There are 52,214 guilds on WoWJutsu... out of all 52,214 guilds that are raiding, 127 of them fall into a situation where only 1 class in their group can tank. That's .2%. There is not a single reason why every fight needs to be balanced for .2% of guilds.

I don't think it needs to be tweaked for it, but I also don't think it should be explicitly designed to screw those. We don't have any fights that unduly punish rage users, and I doubt we plan to.

You're looking at this from the perspective of a fairly-progressed guild who has Naxxramas on farm. You (and I) have a pool of people who can re-gear, or may have tanking alts that they like playing also. However, there are still guilds who are NOT as progressed, who still struggle seriously with parts of Naxxramas (as evidenced by requests for advice about 4Horsemen, Heigan, and Patchwerk), and who do not have that pool of people with enough spare time to have multiple alts and enough badges/heroics to have multiple gear sets. The smaller the sample size (10 vs 25), the easier it is for some key factor like that to be missing.

You could take our 25-man group and partition it into two raid teams for 25-man; some of us play more than others and thus could bring alts or alternative gear sets, but some cannot. We just happen to be fortunate that (a) I have healing gear and (b) some of our DPS like to play tankish alts. Not all 10-man guilds are so fortunate.

Expecting your players to be geared for two roles is something that's fine for hard modes, but may be a bit much to ask of some guilds. Then again, Ghostcrawler HAS said that Ulduar is where we have to bring more to the table, so perhaps this is already part of their design goal. I'm skeptical, though, that screwage of two-paladin 10-man guilds is intentional.
User avatar
Kelaan
 
Posts: 4036
Joined: Thu Jan 03, 2008 12:01 pm

PreviousNext

Return to T8: Ulduar / Emalon

Who is online

Users browsing this forum: No registered users and 1 guest


Remove Advertisements

Who is online

In total there is 1 user online :: 0 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: No registered users and 1 guest