real easy way to measure expertise in defense

[Astarica]

There's usually a lot of argument about how good is mitigation defense versus more HPs, and then it just occurred to me. The easiest way is hover over your death counts in a WZ, which tells you the damage taken. For my last WZ, I took 163949 damage and had 4 deaths. This means it took on average 40K damage to kill me.

 

This means 1% damage mitigation is worth 400 HP even though my HP bar reads 18000. It is not a theory thing. This is exactly the amount of damage I took before I actually died. There's no point to talk about how good HP or expertise is defensively if you never died, so only the the number of deaths matter.

 

Of course this number also varies depending on which WZ is in, and sometimes you might even have 0 deaths (though in this case we don't actually care too much about whether any kind of defense matters because you never died).

 

So if you're wondering whether more HP or more expertise is good for you, just look at your next 5 WZs or so. That way it will tell you exactly how much HP you really have before you died. This number automatically factors in any stuff like self heals, healers (or lack there of), and so on. No need to theorycraft at all.

[iRedRIng]

I think this is flawed unless in every warzone you get no healers and you never drop combat.

[Talarchy]

There's usually a lot of argument about how good is mitigation defense versus more HPs, and then it just occurred to me. The easiest way is hover over your death counts in a WZ, which tells you the damage taken. For my last WZ, I took 163949 damage and had 4 deaths. This means it took on average 40K damage to kill me.

 

This means 1% damage mitigation is worth 400 HP even though my HP bar reads 18000. It is not a theory thing. This is exactly the amount of damage I took before I actually died. There's no point to talk about how good HP or expertise is defensively if you never died, so only the the number of deaths matter.

 

Of course this number also varies depending on which WZ is in, and sometimes you might even have 0 deaths (though in this case we don't actually care too much about whether any kind of defense matters because you never died).

 

So if you're wondering whether more HP or more expertise is good for you, just look at your next 5 WZs or so. That way it will tell you exactly how much HP you really have before you died. This number automatically factors in any stuff like self heals, healers (or lack there of), and so on. No need to theorycraft at all.

 

Does not take into account at all the 30k of heals you got before each death.

[LancelotOC]

I think this is flawed unless in every warzone you get no healers and you never drop combat.

 

Yea maybe if you subtracted total heals/players and maybe even total protecion per player your number would be closer to total damage per death. Also you are assuming total damage taken is unmitigated damage. It probobly represents mitigated damage or the total of actual damage after your damage was reduced by your defence.

[Astarica]

Um, the whole point of Expertise is that anybody who protected ou/healed you make use of your extra mitigation. I.e. if I take half as much as you do, then any heals on me is essentially twice as good. Subtracting that number would completely defeat the point of trying to understand what's the true value of Expertise.

 

Of course some games you'll have higher values than others. The next game I have was 85000 damage taken 4 death, down to 21250 HP per death. There probably isn't a healer in that game, so it took considerably less damage to lose one life compared to before. But I still have my own self heals, med packs, and so on, and those still count.

[Ahhmyface]

Astarica is correct. Big surprise that the forum rabble lacks the reasoning skills to comprehend why. Edited June 14, 2012 by Ahhmyface

[LancelotOC]

Excuse me as it's very late but I'm not grasping this. Are you saying with more expertise the damage taken per death number would go down and by comparing it to previous average numbers you could determine what additional benefit this additional expertise gave you?

 

Wouldn't a number like average damage taken per death be more indicitaive of how well the enemy was assisting on you rather than the value of the additional expertise? The longer you live the more defensive cooldowns in that span, the more heals received, the more health nodes would respawn to be grabbed, the more opportunities to get out of combat and self heal.. etc.

[Astarica]

Excuse me as it's very late but I'm not grasping this. Are you saying with more expertise the damage taken per death number would go down and by comparing it to previous average numbers you could determine what additional benefit this additional expertise gave you?

 

Wouldn't a number like average damage taken per death be more indicitaive of how well the enemy was assisting on you rather than the value of the additional expertise? The longer you live the more defensive cooldowns in that span, the more heals received, the more health nodes would respawn to be grabbed, the more opportunities to get out of combat and self heal.. etc.

 

It's pretty clear that expertise benefits with any amount of healing on you, since any HP replenished is used by the protection granted by Expertise, while extra HPs don't benefit from additional heals.

 

Now it is difficult to say just how much heal you expect to get. Some people will argue they never get any heals, so a full life bar should be treated as your full life bar. Let's say it's 17K, then 1% mitigation has a value of 170 HP and thus if you can get more than 170 HP for 1% of expertise, you should go with the HP, at least for defensive purposes.

 

However if you operate in a world where you never get healed, all the HP in the world won't do you any good. In realty, people DO heal you. If you take the number of damage you took divided by death, this basically factors in whatever healing you may have received. Yes it'll also catch random things like going out of combat or any % based heals (which is as good as Expertise since it's % based). But, this is about as close as you'll get an idea of how much you actually get healed.

 

In the first example I was in a game with pretty decent healers. You can see that on average it took 40K damage before I died once. This means my lifebar is effectively 40K HP with a healer around.

 

The second game it was a standard 'no healer' game. In that game you can see it took 21500 damage before I died, but it's still more than the 18000 HP I started with.

 

I logged on a level 18 alt, I had 11 death and took 198000 damage. That's 18000 HP per death, compared to her base HP of 12000 HP.

 

People always greatly underestimate the amount of heals they receive. Even in totally PUG games sometimes you still get healed. Now, using this number doesn't necessarily mean Expertise is better, but it makes Expertise a lot more favorable compared to HPs. Though I think it'd be more fair to say the standard 'no heals' assumption for Expertise versus HP greatly discounts the value of Expertise, not to mention it's a pointless scenario since in a 'no heals' situation, you're already dead.

 

In the case of my alt, if she was level 50, at 18K HP per death that means 1% mitigation = 180 HP. It is definitely possible to beat this ratio on some slots (maybe Campaign relic versus PvP relic), so if she's somehow level 50, I probably should take a Campaign relic instead, at least for defensive purposes.

[Funick]

my favorite is healers who try to argue with me that they should use 10% more health over 10% damage reduction... you're a healer, if you get even one heal off the damage reduction will do far more to increase your survivability.

[Sabredance]

Pretty flawed conclusion.

 

I´d love to see a Healing received stat tho.

Edited June 14, 2012 by Sabredance

[LancelotOC]

Yea very good. It would be your personal number. More reliable for 8 mans than currently. As the number would also rate how well your enemies did against you or it would indicate what type of classes/abilities you faced or it would indicate how well your healer and tank played. Rather a mix of these.

 

The flat 1% of hitpoints vs. 1% of mitigation ignores all these things. To valuate these more elusive things differntly would still probobly bring you close to the same place.

 

I think though, in your pursuit of this elusive number, you should be very conservative. As the times you would get 400 per 1% could be just situations where you completly outplayed your opponents or significantly outgeared them or had more healers etc. .

 

It would be better to only use for averaging the numbers aquired in challenging warzones.

 

One percent of expertise may be worth more than 1% of hitpoints for defensive purposes . My opinion is you should toss out your better numbers when analyzing since ultimatly you are only min/maxing to do your best against the challenging situations. Min/maxing to do great in situations where you'd be predominatly succcessful otherwise is overkill.

 

Oh and I called it a presonal number because of my experiance with my friend. He and I are specced almost identically. We are geared almost identically. I kite and LOS much more and live longer on average. Common enemies accept that I'm more slippery and when given the choice of focusing him or me almost always focus him. From an empirical point of view our hitpoints per death should be the same and the value of expertise should be equal for us. The truth though is expertise is worth more to me and hitpoints is more precious to him than me.

[Revelationjp]

Yeah, you're not taking healing into account at all. Plus, I may be mistaken, but damaged RECEIVED is after mitigation anyway, so you're not measuring mitigation at all, you're ONLY measuring healing effect on that battle.

[LarryRow]

Is it just me, or does this also ignore the fact that expertise increases damage dealt as well.

[SinnedWill]

Problem with monitoring total damage taken/death count is that it gets skewed. Healing received from medpacks/healers/wz heal powerups will result in SIGNIFICANT increases to damage taken per death (as will enemies that tend to deal sustained damage and/or DoT's). Inversely proportional to that would be situations where you cannot use medpacks/have no healers/can't get the heal powerups between deaths (as a result of stunlocks + burst damage dealing/focus fire) - thus, you'll have significantly less damage taken per death in this fashion. Sorc/Sage bubbles will also throw damage taken/death values off (damage dealt by enemy team to a bubbled target is negated and only contributes to healing values for the sorc/sage that casts the bubble) - as will defensive abilities, mitigation values, etc.

 

The fact that damage taken per death varies significantly depending upon these factors makes it consistantly unviable to apply (unless in a controlled situation/circumstance) as a constant to be proportional to expertise. However, it CAN be done in theory (i.e. if all players were the same class/spec, using the same abilities on each other in 1v1's for the entire duration of a match - and none use any activatable items, none use defensive abilities, none use heals of any sort) - this would get you the ACCURATE damage taken/death values you're looking for, but will never happen in actual application (let alone, everyone would also have to change out of pvp gear following that first match and equip the identically equivalent pve gear and compete with the same identical restrictions as the first match).

[Astarica]

This doesn't factor into offense because it's a whole different dimension. Some characters might only care about defense, and certainly it's not reasonable to assume to everyone cares about offense/defense equally. So this is a strictly one dimensino comparison.

 

For those of you talking about healing skews the numbers, way to miss the entire point of expertise. Let's say you can get 50% expertise for defense only, or 100% more HP. I hope you take the 50% expertise for defense only over 100% HP. Almost certainly with this amount of expertise you'd never die with any halfway decent healer, getting you an infinite number for damage taken divided by death. On the other hand the guy with double the HP is unlikely to have no death at all since it's not hard for a few good DPS to beat the HPS of a single healer so you'd still die eventually. This is because Expertise and healing is built to work together. So of course healing will greatly skew the damage/death number because that's what healing is supposed to do: magnify any advantage in mitigation you already have.

 

And of course this is a personal number. If you always play very conservatively by running away from all engagements your damage/death is going to be higher than someone who stands his ground. Someone who never runs away may indeed have a damage/death number that's very close to his current HPs. But it's far more practical to actually try to measure this number then just say, "I think no one ever heals me because they all hate me". You'll be surprised who heals you out there and if not, it's at least for a way for you to know for sure rather than just theorizing. If you found that your damage/death number is consistently close to your total HP, then maybe no healer wants to heal you and in that case you'd be better off having a bigger life bar than more Expertise. Hopefully your situation isn't that grim, but it's possible.

[LancelotOC]

You are definatly closer to quantifying this than other attempts I've seen. I think a fair way to push foward would be one of two ways.

 

You could wait for 8 man's then use your hpts/death from those matches as you would both be consintently playing with the same players and consistenly playing against other determined 8 mans.

 

Or you could set the criteria for an average group of allies and an average group of enemies. For example if most times in your personal view you face groups consiting of 3 healers 2 tanks and 3 DPS with mostly groups of 2 healers 2 tanks and 4 DPS then only record your hpts/deaths for those WZs. Then find your magic number of what 1% defence equals in a hitpoints sence for your set of circumstances.

[SinnedWill]

For those of you talking about healing skews the numbers, way to miss the entire point of expertise.

Having lots of healers and/or a dedicated pocket healer (whose primary concern is to follow one specific player around to keep alive the entire warzone - which tends to be the primary objective given to a healer in a pre-made when one particular member of the group is a glass cannon dps capable of doing something exceptional enough to warrant it) will ALWAYS skew the damage taken/death value compared to being in any other circumstance - particularly when this glass cannon dps class can take 0 deaths (with over 300k damage taken) in favorable situations and 8 deaths (with ~ 160k damage taken) in others.

 

If you found that your damage/death number is consistently close to your total HP, then maybe no healer wants to heal you and in that case you'd be better off having a bigger life bar than more Expertise. Hopefully your situation isn't that grim, but it's possible.

 

Trade damage dealt/received/healing value changes for more HP? I can only see flat-line %-based healing received classes POTENTIALLY benefitting from this (i.e. madness sorcerer, annihilation marauder, juggs, p-techs, etc.), but you still tend to lose a good bit of damage dealt for increased heal-valuse which might not even guarantee an increase to survivability - certainly doesn't if you're taking increased damage and the enemy team has significant burst-damage to drop someone quickly.

 

Also, the presence of healers isn't the only thing that skew the damage taken/death value (though they are the most significant variable) - many things will do this as well (i.e. juggs popping endure pain despite the fact that they'll still die will increase damage take/death; marauders popping undying rage will REDUCE it if they still die due to the HP sacrifice; crit-procced reactive heling and self-healing without being a healer as mentioned above will increase it; usage of medpacks, obtaining heal powerups, etc.).

Not to mention that if you're facing an enemy team whose class composition is very burst DPS heavy, your entire team is more likely to have a lower value of damage taken/death compared to facing an enemy team that has less burst-damage regardless of the presence of healers, self-healing, etc.

 

Simply put: there are too many variables that can factor heavily on damage taken/death values to make it inaccurate and when you're trying to prove something mathematically, you want as few inconsistancies/inaccuracies as possible.

[SinnedWill]

This doesn't factor into offense because it's a whole different dimension. Some characters might only care about defense, and certainly it's not reasonable to assume to everyone cares about offense/defense equally.

One thing you cannot argue: (offensively) if group of players "x" can kill group of players "y" more quickly/efficiently (happens often, particularly in novare coast zerg-fests at south), then it's safe to assume that group "x" will have a higher damage taken/death ratio than group "y" since they can use out of combat regeneration abilities between their opponents' respawns and the fact that group "y" will have more deaths.

Therefore: The whole "offense is the best defense" quote rings true and certainly DOES play a factor in that situation.

 

Now I do agree that not everyone is wanting to focus on both evenly (usually tanks that want to play the role of a tank are the most common proof of this) - but when a fully WH tank-geared tank gets lit up and dies more often than a fully WH dps-geared tank of the same identical class/spec does (which is almost always true), then this further goes to show the result of what happens when someone decides to purely focus on ONE aspect (right now, it's safe to say that offense significantly trumps defense - 1.3 might change this completely around or at least reduce the gap, however).

Edited June 14, 2012 by SinnedWill

[LancelotOC]

I dont think the guy is saying he thinks defense has merit over damage or vica versa. He is just saying you can quantify all these variables within acceptable deviations by using deaths/damage.

 

He is simply pointing out that regardless if there are a gazillion variables, the sum of their effects neatly closes into your deaths/damage for that specific warzone. So there is a readily available way to weigh it if you care for such.

[Revelationjp]

I dont think the guy is saying he thinks defense has merit over damage or vica versa. He is just saying you can quantify all these variables within acceptable deviations by using deaths/damage.

 

He is simply pointing out that regardless if there are a gazillion variables, the sum of their effects neatly closes into your deaths/damage for that specific warzone. So there is a readily available way to weigh it if you care for such.

 

But even saying that is wrong. Theoretically, lets say you have one attacker, one healer. and one person receiving the attention of both. Then let's say that the healer can slightly outheal the damage done. Again, theoretically, you would have an infinite damage to death ratio. If you don't factor out healing, defensive cd's, stims, etc... first, every bit of data you extrapolate is meaningless. And since factoring out those things would be near impossible, this whole subject doesn't make much sense. Now show me your damage to death ratio after factoring out all those variables, and I'll happily discuss what the data shows us then.

[SinnedWill]

I dont think the guy is saying he thinks defense has merit over damage or vica versa. He is just saying you can quantify all these variables within acceptable deviations by using deaths/damage.

 

He is simply pointing out that regardless if there are a gazillion variables, the sum of their effects neatly closes into your deaths/damage for that specific warzone. So there is a readily available way to weigh it if you care for such.

 

Sorry, mate, but as I've pointed out - a glass-cannon dps class with DEDICATED HEALING from a premade group will go a full match without any deaths (this is NOT theory, this is how one of our premades rolls) - this causes 300k+ damage taken, 0 deaths. When the healers aren't logged in, that glass-cannon will usually have 7-8 deaths against the same (or most of the same) enemies and only takes ~160k damage.

 

Mathematically, round 1 gives him a value of infinity (300,000 / 0) - round 2 gives him a value of 20k (160,000 / 8). The difference of these values (based on the variable of whether or not dedicated healing was present) would result in the deviation of the values to be mathimatically rejected for inconistencies based upon the presence or absence of such a significant variable.

 

The math could work in a completely controlled environment, but not in practical application.

[LancelotOC]

Think about it more broadly. We are devoid of any way to quantify all these variables. We can name countless ones.

 

Right now we could all say that 1% of mitigation is equal to 1% of hitpoints. The truth is it's worth more because of healing, defensive abilities, your teammates play, your enemies play, your team's class mix, the enemies teams class mix.. and on and on.

 

Astarica is right in that the actual numbers you get in deaths/damage is what your mitigation is worth to you as the sum of all variables.

 

Of coarse it's only worthwhile in a controlled enviorment. Well what about 8 man ranked warzones? It will be controlled in that you will mostly always have the same 7 people with you, who will be specced practically teh same way, using the same reactions to situations they normally would. It will be controlled in that your enemies will also always be similar balanced teams following similar likewise behavioral patterns.

 

Yes these constraints don't amount to much but if you recorded Astarica's indicator from enough warzones you should get a somewhat normal distribution of values. The average of these being a better indicator of what mitigation is worth to you personally than simply 1% mitigation equals 1% of hitpoints.