Balancing WAR: A Game of Graphs

[Gamemako]

As a continuation of my suggestions from the How to fix WAR thread, I thought I'd generate a complete report on what went wrong with WAR, and how to improve it without fundamentally redesigning the job.

SUMMARY: Warrior is an ineffective tank at endgame damage levels due to design issues with Wrath. Inner Beast is WAR's burst mitigation. It is mostly sufficient for the task, even in more difficult content. WAR's health pool is too small to take appropriate advantage of it in some situations. The primary cause of WAR's later tanking troubles are the design decision to tie continuous mitigation (i.e. healing bonus) to burst mitigation (Inner Beast). Solutions are suggested: place the healing bonus on Defiance rather than Wrath, increase the potency of the healing bonus, and increase the maximum health bonus on Defiance. Suggestions are supported with mathematical outcomes.


Part 1: Tanking for dummies; or, dummies for tanking

This is WAR at current in one graph.



This is a graph of the total unique mitigation between WAR and PLD as a function of the rate of damage delivered to a WAR. The x axis is the damage delivered to the tank after armor reduction and dodge chance; these two effects are identical between the two jobs. The y axis is the average portion of healing requirement reduced after blocking differences, damage reduction, self-healing, and healing bonuses. I have included 5 cases here. The first is PLD, which is fairly straightforward. It includes blocking, Rampart, Sentinel, Hallowed Ground, Convalescence, and Bloodbath. The second case, Wrath-only, is a case for WAR where the player does not use Inner Beast at all, instead sitting on Wrath stacks for the healing bonus. Other abilities are still used, including Thrill of Battle, Second Wind, Berserk, Bloodbath, and Convalescence. The next two cases, WAR-low and WAR-high, include repeated use of Inner Beast as well as all abilities from the Wrath-only case. Both of the two cases involve the use of 3 Inner Beast per minute buffed by Maim and Storm's Eye, with a 20% average overheal and 1/4 chance of being used in Berserk. Due to the use of Inner Beast, average Wrath stacks are 3, for an average bonus of 9% healing. The WAR-low case is a VIT-heavy ilvl70 setup with Garuda weapon (about 1133 heal per IB). The WAR-high case is effectively an all-strength setup with a Bravura+1 and ilvl90 gear (approximately 1600 average heal per IB). The final case, WAR-hold, is a hybrid between the Wrath-only and Wrath-high cases, where Inner Beast is only used in conjunction with Infuriate. This means that Wrath is always at 5 stacks, but the player still uses 1 Inner Beast per minute.

Right now, in terms of total unique mitigation, WAR is half the tank that PLD is. This culminates in an ideal endgame strategy which involves completely forgoing the burst abilities of the job to maintain a gimped healing bonus. As I will show, this is primarily the result of some dodgy design decisions related to Wrath as a resource.

First, a preface:

A tank has three important tasks:

1. Hold aggro, keeping squishies safe from slaughter
2. Reduce healing need so that healers don't end up out of mana MP.
3. Not die.

The first is easily accomplished by either tank. Any issues here are purely the fault of the player. I'm not going to get into the math here because it's not relevant and certainly doesn't need attention one way or the other. The second is exemplified by the opening chart: all told, WAR is just not getting the job done. The third is a question of total reserve, and the story is told about like this:



These are the effective HP pools of WAR and PLD, as well as when using their abilities. Hallowed Ground is not included because infinity is a really big number. A WAR has 85 more base HP than PLD. WAR's HP is increased by 25% from Defiance, while PLD's effective HP is also increased by 25% from Shield Oath (this is not a typo -- 1/(1-0.2) = 1.25). WAR has only Thrill of Battle to increase effective HP pool from there by 25%. However, PLD will use Rampart to increase effective HP pool by another multiplicative 25%, or Sentinel to increase it by 66.7%, or both together for a whopping 108.33%. In essence, a WAR goes down faster than PLD due to having a smaller effective health pool.

This is where we stand right with WAR right now. Doesn't take a doctorate to pick which tank is flat-out better. Let's look, then, at how we got here.

Part 2: Burst as a really bursty thing

When talking mitigation, there are three types of mitigation which are important to balance. The first is continuous mitigation. In normal functioning, you will reduce some portion of damage, either by preventing it, healing it, or handing it off to the next guy and letting him eat it (sucker). Continuous mitigation is the stuff you have all the time use all the time, and maintain regardless of the opponents you face. This is your block, parry, and continual bonuses like Shield Oath and Wrath's healing bonus. The second is burst mitigation, the stuff you use to get out of the bad spots like Mountain Busters. Burst mitigation are active abilities like Rampart and Thrill of Battle. The final is the sum of the first two: continuous mitigation plus the time-weighted effect of burst mitigation is total mitigation. Total mitigation tells you the net outcome of all effects. This is important because of discrepancies in individual timing.

Inner Beast is not continuous mitigation. It is not used to maintain WAR in normal functioning. Every time you compare Inner Beast to Shield Oath, Althyk kills a puppy. Please, think of the puppies. Inner Beast is your primary, secondary, and tertiary burst mitigation. Almost every ability WAR has will support use of Inner Beast for burst mitigation: Berserk rolls up damage for more drain, Thrill of Battle gives you more breathing room for healing hits, Maim and Storm's Eye increase damage for more drain. And all told, Inner Beast is actually not so bad. Let's look at the benefit given by burst mitigation abilities in a few cases. I have assumed 20% average overheal -- this will depend on player and healer as well as another issue I will address later.



This is a simple treatment of two burst abilities compared: WAR will use two Inner Beast, one under the effect of Berserk, while PLD will sit on Rampart. The effect of Inner Beast is averaged over the 20s interval. Note that I chose Berserk over Infuriate because it has an identical cooldown to Rampart. You could also use Infuriate here for similar results -- it really doesn't matter.

You could also compare peak burst ability between the two jobs.



In this case, we compare a WAR or PLD throwing everything they have up over 20 seconds. I have two cases here for PLD due to Hallowed Ground. The first considers using only Rampart, Sentinel, Convalescence, and Bloodbath, without use of Hallowed Ground. The second uses Hallowed Ground as well. This is due to the 7-minute cooldown on Hallowed Ground and general insanity of the ability. WAR will use a 7 cooldowns here to match up: Bloodbath, Berserk, Second Wind, Thrill of Battle, Infuriate, Convalescence, and Vengeance. These will support the use of 3 Inner Beast over 20 seconds, two under the effect of Berserk. Obviously, WAR cannot compete with Hallowed Ground, but that was never going to be a fair fight. Finally, you can compare the total time-weighted effect of burst mitigation:



In most cases, WAR ends up a bit shaky at the outer limits of damage. This is the messy effect of unscaled healing. You're always going to be more powerful when you have fewer healers, and that is unavoidable using unscaled drain setups. Nevertheless, this difference isn't the death of WAR. It's actually very little of it. Compared to using both Rampart and Sentinel at the same time, WAR keeps up all the way to 1000 enemy DPS. Imagine if WAR and PLD both had shield oath and nothing else. This is what they would look like over time:



At the very limits, PLD would be just 17-25% better at the edges of healing. Unfortunately, WAR doesn't have a comparable reduction to Shield Oath. Which leads us to where the breakdown occurs:

Part 3: Bringing an axe to a shield fight

Continuous mitigation is the lion's share of how WAR falls behind. There are three primary contributors: the coupling to Wrath, the bonus magnitude, and blocking. We will address each in turn.

Right now, WAR has a shared resource for continuous and burst mitigation. Using your Wrath stacks on burst mitigation (i.e. Inner Beast) reduces your continuous mitigation (Wrath healing bonus). This mechanic actively drags down burst mitigation. For that, consider the total mitigation of burst periods. WAR is clearly behind when you account for the impact of continuous mitigation on total burst mitigation.



This should look pretty familiar now, because it's basically the same as the very first chart with blocking omitted. This is how the molehill becomes a mountain. As we saw in the very first chart, hammering Inner Beast falls behind just sitting on Wrath at high enemy damage rates. This is because healing rates scale with content while burst self-healing does not. [b]There is no way to balance this system of flat heals and dependent scaled recovery.[b] If you increase Wrath's healing bonus, then using WAR's burst mitigation becomes increasingly disadvantageous -- WAR will just be pressured even further to sit on Wrath and never use it. If you increase the burst mitigation to avoid this, then you further unbalance WAR in 4-player content with minimal impact on 8-player content. The healing bonus and Inner Beast must be decoupled. Simply adjusting numbers will invariably fail to balance WAR.

With that out of the way, the magnitude of the current healing bonus is also clearly too low. PLD has an effective healing bonus of 25%. This is easily shown by considering the effect of healing reduced damage. Consider this example: PLD is struck for 1000 damage. Shield Oath reduces this to 800 damage. A healer then heals PLD for 800, filling up the PLD's health bar. The healer healed enough to withstand another 1000 damage -- that is, the healer effectively healed 1000 damage. 1000 damage healed / 800 healing delivered = 1.25 HP healed per point of healing delivered -- a 25% bonus. WAR's bonus just works on top: if the WAR is hit for 1000 damage, a healer has to heal for 870 points with the 15% healing bonus to recover the full 1000 damage delivered. It's pretty obvious here that WAR's bonus is just wrong. If you believe that WAR's burst mitigation was making up the difference, you can scroll back to the previous section and completely dispel that notion in a hurry.

The third contributor here is blocking. This is a no-brainer. PLD has a shield. PLD blocks things. PLD takes less damage. This is only really applicable on a continuous basis, where you're getting hit often enough for the randomness to wash out. It also only applies to physical damage taken from the front or sides -- magical attacks and attacks from behind cannot be blocked. Still, blocking is rather powerful, adding an additional ~5% damage reduction in melee damage. This one is pretty tough to balance, though, because it both scales with enemy damage and has no reliable impact on burst survivability. Countering it is more a matter of applying comparable utility or reduction that is not reliable for burst situations.

Of course, there are a few other spots which have impact. Rage of Halone, on-demand stun, pacification, silence, plus the odd Stoneskin are also tipping the scales in PLD's favor. I'm not specifically going to address these here specifically -- the goal is to address the core issue.

Part 4: Cure is the cure for what ails you

This is the easiest section, because I've already shown the problems and touched on how to address them, plus given an example of what happens once the solutions are implemented. In the second section, I gave an example of what WAR would look like if it had Shield Oath. Here's what it looks like from a total standpoint:



Now, remember that the effective healing bonus from Shield Oath is 25%. Place WAR's bonus from Defiance (no longer Wrath, as the two must be decoupled) and the chart above is the outcome. The great part about the healing bonus is that even when it's not coupled to Wrath, it's already at war with itself. Every amount you heal with your abilities is an amount not affected by the bonus. As usual, this is easily understood graphically:



When you fall to lower damage levels, self-healing takes over and the healing bonus falls in value. The bonus affects a smaller portion of total incoming DPS -- you can't heal the same damage twice. This normalizing effect helps keep it from being overpowered in 4-man content, while adding maximum benefit to 8-man content. The net buff to mitigation ability across damage ranges looks like this:



Where the WAR needs help most, this buff would give maximum effect. Where WAR doesn't need it, the effect is minimal. Since it does not increase self-healing at all, the buff would have no impact on solo balance.

Part 5: Glass ceilings and glass jaws

We've now covered how to address the weakness in the second duty of a tank: reducing healer load. Now, on to the third. Inner Beast is burst rather than continuous mitigation, there's a glaring design flaw with current design that is reflected in the second chart of the first section. When PLD uses burst mitigation, that effect occurs concurrent with damage. For WAR's effect to occur, it must be enacted after damage. WAR has to eat damage before it can be healed. Let's go back to the chart for an example.


(Note that blocking is omitted here. While blocking does have an impact, it is not a reliable eHP increase and cannot be counted without placing a job in a position of randomly dying. I have also omitted Stoneskin, which is an effective 10% increase for PLD, but which is not always usable.)

WAR's burst abilities (ToB excepted) do not extend the effective HP pool, while PLD's do. WAR and PLD at base have identical eHP pools, so they take hits just as well when using no abilities. In a nutshell, this means that WAR is simply easier to kill in every single case than PLD -- if a WAR can survive it, so necessarily can PLD. The problem that arises is that in burst situations, WAR still has a small eHP pool while PLD has a much larger one. This is where WAR goes wrong on the third role of a tank: not dying.

If WAR is to be able to use its burst mitigation ability, it must have the extended HP pool needed to take the hit before healing it. This is necessarily larger than PLD's base pool, because PLD has an HP pool for burst uses which is 25% larger. You can view the correct balance point in two ways, and they are almost identical in effect. The first is to key WAR's eHP to the bread-and-butter eHP of PLD in burst situations, which is Rampart. Rampart increases PLD's eHP pool by an additional 25%. For a comparable bonus on Defiance, this would give an HP bonus of approximately 54% compared to the current HP boost of 25%. The second way is to look at the base eHP value of PLD and add on top of that one use of Inner Beast (roughly 1600 HP). This would come to a bonus value of about 49.55%. Either way, the prescribed change is to move the HP boost on Defiance up to 50% from 25%. At that point, the comparative eHP pools would look a bit like this:



This allows WAR to live long enough to use its mitigation abilities.

Part 6: Fixing what ain't broke

For the sake of completeness, let me hammer out a few of the common suggestions for WAR which will not fix WAR. First is buffing WAR's damage output in any way, such as via reducing Defiance penalty or buffing Unchained. This one is terribly obvious, because damage output has nothing to do with why WAR can't tank. Sure, Unchained sucks, but it's not why WAR sucks at tanking Dreadnoughts. Second is buffing Inner Beast's use rate or healing amount. This one should be obviously ineffective to anyone who realizes that Inner Beast is burst mitigation. Still, let's create a case where we buff Inner Beast to the point where it's competitive with PLD at 1000 DPS.



WAR is still behind at endgame, just as much as with the recommended buffs. However, it's also rendered godlike in 4-man content. And for how silly-huge this buff would be, you could also replace the description with "restore HP to max", because that's what it's doing. 5650 average heal in the high case. Overheal alone makes this magnitude of buff totally impossible, and it's still behind. Hard though you may try, there is no way to fix WAR by buffing flat heals.

Another common suggestion is adding an overheal ability to Inner Beast, like Scholar's Adloquium. This is just effectively increasing flat heals again -- you've reduced overheal, but still just have flat healing. As described above, flat healing can't fix WAR. It isn't wrong, per se, but it can't be the solution WAR's problem. Similarly, swapping methods of flat healing from damage-based to percentage of max HP won't do any good either. You'd make one stat better, but you'd still just have flat heals with the same problem. To address WAR, you have no choice but to use some form of scaled mitigation.

Furthermore, not all scaled mitigation is created equal. Consider a case where WAR, instead of using a flat-healing 200% damage, heals 20% of damage received in the past 20 seconds. While this looks good on paper, it runs into the same problem as before: endgame opponents will deal more damage in 5 seconds than you can fit on your HP bar. Due to the damage rates encountered, you would have to couple any such scaled healing to an increase in Wrath generation, or the result will simply be a truckload of overheal. Worse, with that ability still coupled to Wrath, Shield Oath alone is an additional 12% mitigation on a constant basis. Add in blocking and you could drain 20% of all incoming damage and still be permanently behind.

Part 7: More than one way to skin a Coeurl

Buffing the healing bonus isn't the only way to balance WAR. As addressed in the previous section, all options based on flat healing are ineffective. However, a healing bonus is not the only way to improve WAR. I'll address a few genuine possibilities, as well as explain why I don't believe they are good for ARR:

Some players have suggested going full-WoW with WAR and just scaling Inner Beast. This is technically feasible: have Inner Beast heal ~30% of damage received in past ~5 seconds and cost only 1 Wrath stack. That would fall out to a 24% mitigation boost with an average 4 Wrath stacks. This would be a continuous 32% total mitigation, or around where PLD is now after. However, this solution would also ruin the use of Berserk as a mitigation boost and would make WAR pretty damn one-button-wonder with the spam of Inner Beast. You would actively mash Inner Beast 10 times per minute or more -- more than every third GCD would be hitting Inner Beast yet again. It would also be totally copying the hell out of Death Strike, right down to the numbers -- 30% (20% + 10% shield) of damage from the last 5 seconds is exactly what Death Strike is.

Others have suggested adding damage resistance to WAR in some form or fashion, either by adding a damage reduction buff to Inner Beast or by just dumping it on Defiance. The effect is the same: you give WAR damage resistance. It should be glaringly obvious that if a healing bonus works, a damage reduction will do the same job. However, there's a particular difference between WAR and PLD that I described in the fifth section: PLD acts before damage comes, reducing the damage taken; WAR acts after the damage comes, undoing it. This applies to both the burst and continuous mitigation for the two jobs. Applying a damage reduction to WAR would not only reduce the uniqueness of the two tanks, it would also make the class more difficult to balance. WAR's reactive burst mitigation requires a higher health pool, such that the mitigation can be enacted. Adding damage resistance would further extend the effective health pool as a function of damage received. You will also extend the value of flat heals (which are unaffected by healing bonuses), improving returns on strength and muddling the necessary HP pool for any new content. From a balance standpoint, it's unnecessary complexity; from a design standpoint, it's encroaching on PLD territory.

Part 8: The road not traveled

Yes, there are a lot of other issues with WAR's design right now. Steel Cyclone and Unchained totally suck. WAR doesn't have good CC/interrupts. WAR lacks comparable reduction tools like Flash's blind or Rage of Halone's debuff. WAR has to wear hideous tin can armor. Yes, all of these hurt WAR. However, they are not what has shoved WAR off the viability cliff. WAR needs, first and foremost, to have its core issues fixed. I'd love to take Unchained off Wrath and make Steel Cyclone a hardcore CC ability, but none of it will do any good until the job's scaling is fixed.

TL;DR Wrath mechanics suck.

[Gilthas]

You wrote way too much.

Most of your graphs are incomprehensible.

This doesn't help much.

GJ OP.

Ignore the first two responses, some people have brain damage.

Sorry, but reading the first post was like reading a scientific article where the results don't come with any explanations. The graphs are pretty, but they come with no explanation of how they were constructed, which makes them pretty useless if you're trying to be informative.

I don't have a problem seeing that Warrior's have a problem with endgame. Their damage mitigation evidently sucks when compared to the abilities that Paladins have. And it also doesn't help that Defiance's healing bonus is tied to Wrath. Whatever damage mitigation that Defiance comes with needs to be separated from Wrath. But like I said, if the objective of this topic is to be informative, I don't think its helping much.

Daily posting limit reached so time for some edits that no one will read:

If the point were merely for you to do a number of attacks before hitting IB/CS/unchained then there wouldn't of been any benefit intrinsically attached to Wrath.

Wrath would still give you a bonus to critical hit if that's the problem.

I haven't actually played endgame Warrior, so most of what I can do is just guesswork but I don't really see a problem with Wrath only being a resource to use other abilities.

I do see an enormous problem with Defiance giving WARs 25% bonus to HP when most healing in the game is static rather than percentage based. That's why Defiance should also give a static bonus to healing instead of tieing it to Wrath stacks.

[GinTama]

I call bullshit on this anyways, you just cant play or dont get war.

[Sybreed]

GJ OP.

Ignore the first two responses, some people have brain damage.

[Darthvoltrius]

Good Job, hopefully it will get some serious attention. If I may copy/paste some ideas I posted a bit ago on another thread on possible ideas on how to improve warrior.

Steel Cyclone: this used to be like our bread and butter back in the day in 1.0, now its pointless use of wrath, to make it more useful/engaging change it to the following options:

2 wrath stack: 150-200 aoe dmg like it is right now with an enmity modifier (our version of flash so to speak)
3 wrath stack: 200 aoe dmg with +5% crit or dmg bonus for a duration
4 wrath stack: 250 aoe dmg with a 50%? heal according to dmg
5 wrath stack: 250-300 aoe dmg & -20% dmg taken for a duration

Just a random idea and it may take some tweaks and/or reverse the orders or w/e but at least this way there's more options according to situation and steel cyclone is useful in current situation as Wrath is precious to us right now.

Or like someone mentioned:

Heavy Swing > Overpower > Steel Cyclone Combo would be great.

[ASaturnus]

First a technical comment. You're graph descriptions are confusing and partly unexplained (what's WAR-low and WAR-high). You're y-axis says percentage, but it seems to be a ratio (a mitigation of 0.35% vs. 0.20% would be irrelevant). However, some of these ratios go over 1. More than 100% reduction?
The more general problem with all this numbercrunching I see on the forum is that no one explains basic game mechanics. Don't armor and resistances also provide mitigation? Is the PLDs mitigation additive or multiplicative? Is all damage mitigated?

My theoretical complaint with this kind of reasoning is that looks only a part of the role a WAR or a PLD has. Because contrary of what you claim, a tank's job is not to survive and keep others alive.
A tank is part of a group and only the group as a whole can fulfill a purpose. And a group's purpose in any MMO I know of is do damage. Damage is all what counts because only damage turns bosses into loot and progress. Other things like tanks and healers are only secundary functions in order to keep the group able to do damage.
Thus, any claim that it is not a tank's role to do damage is absurd, because it is the only role of any group. Therefore doing damage is integral part of tanking.
Of course, keeping the group alive is more important, because without that damage falls to 0. But if keeping the group alive is achieved, all what matters is damage.
And that's the problem, because contrary to what everybody says, we do not have the means to calculate or reliably measure the difference in damage between a PLD and a WAR. Thus, we can not conclusively compare both jobs.

tl;dr Numbercrunching alone proves nothing

[Hundred]

1. Which area's are creating the DPS you are discussing.
For instance your graph has 250, 375 and 500 as DPS values which instances are generating 250, 375 and 500 DPS at level 50?

2.

The healing bonus and Inner Beast must be decoupled. Simply adjusting numbers will invariably fail to balance WAR.

That is the Warrior though. When you use Inner Beast, Steel Cyclone or Unchained, you trade sustain and crit rate for their effects, burst heal, aoe, DPS. If the point were merely for you to do a number of attacks before hitting IB/CS/unchained then there wouldn't of been any benefit intrinsically attached to Wrath.

3. You mention increasing the HP increase from 25% to 50%. Seems silly straight out. Because of below...

4. The thing is 50% HP to use what abilities? The only signficant heals are Mercy stroke (adds) and IB. Since ToB already increases your EHP anyways it doesnt overheal. IB is for mitigating burst, not for sustain and it can do that even with the 25% ehp increase. SP is something like 9x of an IB, and Bloodbath requires a bucketload of damage to equal 1 IB. Which is wonky because sustain should beat out a burst in a reasonable period of time but SP never beats out IB, let alone the 10% healing difference, and BL doesn't really do it either, let alone that PLD has access to it for half as long. You don't really have any abilities other than IB that heal anything to justify 50% increased HP just for the sake of more time to IB which is outdone overtime anyways (if we're giving more time before you need to use IB) by the 15% bonus.

[Gamemako]

You're graph descriptions are confusing and partly unexplained (what's WAR-low and WAR-high).

I explained in the text, but suppose I should be more explicit about it. I will try to improve that.

You're x-axis says percentage, but it seems to be a ratio (a mitigation of 0.35% vs. 0.20% would be irrelevant). However, some of these ratios go over 1.

Yes. I could multiply out by 100, but I thought that was pretty self-explanatory. Yes, some do go over 1, meaning you actively return health to the player faster than it can be delivered. PLD generally can't do this (unless you're a extremely low damage rates), but WAR sometimes can. Flat healing does that.

The more general problem with all this numbercrunching I see on the forum is that no one explains basic game mechanics. Don't armor and resistances also provide mitigation? Is the PLDs mitigation additive or multiplicative? Is all damage mitigated?

Armor values are not included because they are identical. They do reduce damage, but there is no difference between the two jobs. Same with evasion. Most bonuses are actually multiplicative -- for example, Shield Oath, Sentinel, and Rampart combine to reduce damage by 61.6%. If it were additive, the three would add up to 80% -- this does not appear to be the case. I could include that detail, but it's a horribly long post without going into extreme detail about mechanics.

My theoretical complaint with this kind of reasoning is that looks only a part of the role a WAR or a PLD has. Because contrary of what you claim, a tank's job is not to survive and keep others alive. A tank is part of a group and only the group as a whole can fulfill a purpose. And a group's purpose in any MMO I know of is do damage. Damage is all what counts because only damage turns bosses into loot and progress. Other things like tanks and healers are only secundary functions in order to keep the group able to do damage.

You are conflating the group's overall goal (defeat opponents) with an individual role. A healer heals. It may often be efficient for the group for the healer to also deal damage, but if the healer cannot fill the role of healing, then the group is doomed to failure. This is the same with tank. If the tank did no damage at all, that would still fill the tanking role. It's a bit like VIT: if you don't have enough, you lose by default; if you do have enough, then it's you're free put efforts elsewhere in hopes of winning. The problem is that WAR is either incapable of filling the role or requires far more resources to do so.

And that's the problem, because contrary to what everybody says, we do not have the means to calculate or reliably measure the difference in damage between a PLD and a WAR.

Er... parses?

//EDIT:

1. Which area's are creating the DPS you are discussing.

Not getting into that at all. You can often create situations for it by stacking mobs together if you really want. I'm just expressing the outcomes as a function of enemy damage output.

That is the Warrior though.

I'm not going to argue about your vision of how WAR should play. From a balance standpoint, WAR will always suck as long as this design is used. It may interest you, but it's not a viable path.

The thing is 50% HP to use what abilities?

I'm not sure how I failed to communicate this. Inner Beast is WAR's burst mitigation ability, and it is a reactive ability. In order to make use of it, you must have taken damage. PLD does not have this weakness: its burst mitigation applies up front. PLD is therefore able to take a bigger punch than WAR in every case. WAR cannot out-think this; Inner Beast cannot be applied beforehand and WAR will always just die. This is a design flaw.

You don't really have any abilities other than IB that heal anything to justify 50% increased HP just for the sake of more time to IB which is outdone overtime anyways (if we're giving more time before you need to use IB) by the 15% bonus.

I believe I addressed the end of the Wrath coupling mechanic already. If you are sitting on Wrath for a 15% healing bonus, you've already acknowledged that you are hopelessly gimped.

//EDIT 2:

You wrote way too much. Most of your graphs are incomprehensible. Sorry, but reading the first post was like reading a scientific article where the results don't come with any explanations.

I will try to improve this.

[Telepathic]

I don't think it changes the conclusion, but if a WAR is using all their CD's for a big burst they should be getting 3 Inner Beast during Berserk. They should also be able to get two IB's off every other Berserk.

[Gamemako]

I don't think it changes the conclusion, but if a WAR is using all their CD's for a big burst they should be getting 3 Inner Beast during Berserk. They should also be able to get two IB's off every other Berserk.

If you have a good skill speed bonus, you can. Without one, you will have 5 stacks and be ready for that last hit just as the effect ends and you get pacified, heh. That would give you an additional ~8% HP mitigated in that 20s interval. Regardless, it would just reinforce the conclusion that WAR's burst ability with Inner Beast is fine, and that the problem with WAR is elsewhere.