Jump to content


Member Since 07 Feb 2010
Offline Last Active Aug 26 2015 01:20 PM

#3758088 5.4 PVP Power and Resilience Guide

Posted Eldacar on 03 September 2012 - 01:00 AM

***With the release of Patch 6.0 leading up to Warlords, this guide is fully out-of-date, a full re-write is in progress***

Eldacar's Guide to PVP Power & Resilience

Hello PVP'ers! For those of you that don't know me I go by Eldacar and I am a long time PVP'er and PVP Theorycrafter, as well as a member of Blizzard's forum MVP program. I write PVP focused guides and do everything I can to help the PVP community grow and prosper, which is why I wrote this PVP Stats Guide. You can find me on twitter as @EldacarJS and on the US Official Forums as [email protected] This guide goes into a lot of detail and gets into some complex stuff in a few places, if you have questions leave them in the comments and I will answer them as best I can.

- The base damage reduction of Resilience was increased to 72%.
- The Resilience rating to damage reduction conversion formula was altered.

Summary & Key Points:
-Resilience has linear returns, +100 Resilience rating increases your effective health by 1.235% relative to displayed health.
-PVP Power has linear returns, +400 PVP Power gives you +1% damage or healing in PVP (before spec modifiers).
-The latest season's PVP gear is generally the best gear for instanced PVP, however in world PVP players with heroic raiding gear will have an advantage.
-PVP Power does not cancel out a target's Resilience but it will still help you hit them or heal them harder.
-It is generally ineffective to gem for PVP Power, gemming for primary stats is more effective in most cases.
-The PVP trinket set bonus offers roughly 8.25% effective damage reduction and will always increase your effective health by 32.11% of your displayed health.

Table of Contents:

Section 1 - Resilience
1A - The Exponential Returns of Damage Reduction
1B - The Diminishing Returns of Resilience Rating
1C - Baseline Damage Reduction
1D - Effective Health
1E - The Full Scale of Resilience
1F – Resilience on Items

Section 2 - PVP Power
2A - The Absolute vs Relative Returns of PVP Power
2B - PVP Power's Interaction with Resilience

Section 3 - Gemming for PVP

Section 4 - Closing Thoughts and Remarks

Section 5 - Appendix
Appendix A - Valuation and Frame of Reference
Appendix B - Formulas

Section 1 - Resilience

Everyone reading this likely already has at least a general understanding of how Resilience works; its fundamental purpose is to provide percentage based damage reduction against all damage done by players, the more Resilience you have the less damage you take. That is fairly straightforward and easy to understand; however understanding how the stat scales and all the factors at work is more complicated. There are three main factors that go into how Resilience scales, first is the exponential returns of percentage based damage reduction, second is the diminishing returns of Resilience rating, and third is the baseline 72% reduction that all players have in PVP.

Section 1A - The Exponential Returns of Damage Reduction

The effects of percentage based damage reduction scale exponentially, the more you have the more valuable additional damage reduction becomes. For example, let's say someone is hitting you for 100 damage, if you have 0% damage reduction and you add 1% that 100 damage is reduced to 99 damage, a 1% effective reduction. However if you already have 90% damage reduction and you add another 1% that 100 base damage which was already reduced to 10 is now further reduced to 9. That change in incoming damage from 10 to 9 is a 10% reduction in actual damage taken by adding just 1% of damage reduction.

Here is a graph that shows how the value of damage reduction increases as you gain more:

Posted Image

As you can see at 50% damage reduction additional reduction is worth twice as much as normal, at 90% its worth ten times as much as normal. This kind of scaling isn't unique to Resilience, armor and any other percentage based damage reduction (even in other games) function the same way. Games control the overall scaling of these mechanics by manipulating how fast you are awarded the damage reduction.

One additional note on this, in World of Warcraft different damage reduction mechanics have multiplicative relationships NOT additive, what that means is that the value scaling for any one of these mechanics is only accurate within that one mechanic. At 50% damage reduction from Resilience an extra 1% from Resilience is effectively worth 2%, however none of this has any bearing on the value of additional damage reduction from say armor, that scales totally independently but in a similar fashion. Because these defensive stats scale independently from each other and have a multiplicative relationship they can each be examined and valued independently.

Section 1B - The Diminishing Returns of Resilience Rating

For Resilience the main factor that counters the exponential scaling of percentage based damage reduction is the diminishing returns on Resilience rating. The more Resilience rating you have the less damage reduction is awarded by each additional point of rating, as shown in the graph below. This is how Blizzard controls the overall scaling of Resilience as a whole, and it is what they change when they want to alter the way Resilience scales.

Posted Image

As you can see in the graph, the amount of additional damage reduction provided by additional Resilience gradually declines as Resilience rating increases.

Section 1C - Baseline PVP Damage Reduction

Mists of Pandaria added a new factor to the way Resilience scales, the baseline PVP damage reduction that all players have which was increased from 65% to 72% in patch 5.4. What this has effectively done is significantly shrink the damage reduction gap between under geared players and fully geared players. This combined with the limited availability of Resilience on gear and through gems has significantly lessened the impact of resilience as a stat at level 90. The difference in damage reduction between a fresh level 90 and someone in the best possible pvp gear is now relatively small.

Section 1D - Effective Health

Effective Health (or EH) is perhaps the most critical metric for measuring survivability. Effective health is essentially how much pre-mitigated damage it takes to kill you. If you have 100k health and 0% damage reduction your effective health is just that same 100k. However if you have 100k health and 50% damage reduction your effective health is 200k, because someone would need to do the equivalent of 200k pre-mitigaged damage to kill you.

It is also important to note that more than just increasing the size of your effective health pool, damage reduction also increases the relative effectiveness of heals on you. With 50% damage reduction a 1k heal actually restores 2k of effective health. This is one of the reasons why increasing your effective health through damage reduction is better than increasing your effective health an equivalent amount through raw stamina.

Effective health is really the stat that best indicates the value you are getting from Resilience and it is the stat you need to be paying attention to when evaluating the survivability of your character. Effective health is displayed on the graphs below as a percentage relative to displayed health, an effective health (EH) value of 150% for a player with a 100k displayed health pool would mean that player has an effective health from just Resilience of 150k. When you factor in other effects like armor and damage reduction from talents your EH is higher but we are just looking at Resilience by itself here.

Looking at effective health over the full scale of resilience as shown below illustrates how the exponential returns of damage reduction and the diminishing returns of Resilience rating combine to cancel each other out and generate perfectly linear returns.

Posted Image

As you can see the effective health returns of resilience are perfectly linear, adding 100 Resilience rating will always increase your effective health by 1.235% relative to your displayed health.

Section 1E - The Full Scale of Resilience

This next graph brings it all together displaying both the scaling of damage reduction and effective health based on Resilience rating at level 90 in patch 5.4.

Posted Image

This graph should drive home once again that although the damage reduction you get from additional Resilience diminishes the more you get your effective health continues in a linear fashion anyway thanks to the increasing relative value of that damage reduction.

Section 1F - Resilience on Items

At this point you may be thinking "I want to get as much resilience as possible and become totally unkillable!" which sounds great, but unfortunately it is a bit impractical. Although there is no Resilience cap you are extremely limited in the amount of resilience you can get in game on current season items. Most fully geared players will have around 3375, which is what you get from the PVP trinket set bonus and the PVP meta gem. The lack of resilience on gear is not a big issue because currently a player with nothing but the baseline 72% reduction already has 357% effective health, which is more than most fully geared players had at the end of Cataclysm.

For those of you contemplating using a PVE trinket or two here are some facts to help you make your decision. The 2600 Resilience offered by the PVP trinket set bonus provides roughly 2.31% additional damage reduction from baseline, which is about 8.25% effective damage reduction after factoring in the value scaling. Furthermore the set bonus will always increase your effective health by 32.11% relative to displayed health, so if you are currently at baseline you would go from 357% EH to 389% EH.

Section 2 - PVP Power

PVP Power is a relatively new stat introduced to the game in Mists of Pandaria which acts as the offensive compliment to Resilience. The idea behind this new stat is to encourage players to use PVP gear in PVP by putting major PVP-only offensive gains onto PVP gear (or in the case of healers, healing gains). PVP Power increases all damage done to players (under all circumstances), and healing done (while outside PVE-instances), by a percentage that increases based on how much PVP Power rating you have. The amount of each bonus you get is also dependent on your class and spec.

-Healing specs receive 100% of the healing bonus but 0% of the damage bonus
-All other specs receive 100% of the damage bonus and a partial healing bonus depending on class.
-Damage specs for Druids, Monks, Paladins, Priests, and Shamans receive a 70% healing bonus.
-All other specializations and classes (including tanking) receive a 40% bonus to healing from PvP Power.

PVP Power is currently the primary differentiator between PVP gear and PVE gear. It is a "free" stat on PVP gear, meaning it is not factored into the item's stat budget. As a result when comparing PVP items to PVE items of the same item level all the general-purpose stats should be equivalent but the PVP gear will have PVP Power on it as well making it a better choice for PVP. This fact is particularly important due to the presence of item level limits in all instanced PVP. The item level limits change with each season, but their purpose is to limit the item level of PVE gear to be equal to or lower than the item level of the current season's PVP gear. These two factors together generally ensure that the current season's PVP gear is always the best gear for instanced PVP. However the item level limits do not function in the open world, so in world PVP a player in the latest heroic raiding gear will likely (and unfortunately) have a large gear advantage.

Section 2A - The Absolute vs Relative Returns of PVP Power

PVP Power's scaling is very straight forward; it has linear returns when looking at it in an absolute sense, adding 400 PVP Power will always give you another +1% damage or healing in PVP depending on your spec. So every additional point of PVP Power will increase your damage/healing by the same amount. However I have seen some players around the forums describing PVP Power as having diminishing returns, and they are correct to an extent.

If you evaluate the returns of PVP Power in a relative sense it does have diminishing returns, going from 0%-1% will give you the same absolute damage increase as going from 30-31%, but in the latter case that damage increase is smaller relative to the damage you are already doing. Virtually everything in the game operates the same way. Think about primary stats for example; +3000 strength would give a warrior a pretty nice bump in damage right now. However if Blizzard said "Hey we like you random warrior!" and bumped their strength up to 100,000 then that +3000 strength would suddenly be worth a lot less to them even though it would still increase their damage by the same amount. The reality is that in order for a stat to offer you consistent relative gains as you gear up the stat would need to have increasing absolute returns, in other words it would need to give you more and more damage or healing the more of it you got. (For more info on absolute vs relative valuation see Appendix A)

Now it is time to look at the actual scaling of PVP power, which is shown in the graph below.

Posted Image

As shown in this graph PVP Power's returns are perfectly linear. You gain either +1% damage or healing for every 400 PVP Power you have, this is also the “baseline” upon which the reduced healing bonuses for non-healing spec are based. The red line shows the healing bonus for hybrids (70% of baseline), and the yellow line shows the healing bonus for everyone else (40% of baseline).

Section 2B - PVP Power's Interaction with Resilience

The most common misconception that most players seem to have about PVP Power is that it acts as a kind of "Resilience Penetration" which counteracts the target's damage reduction 1 for 1, that is simply not true. PVP Power increases your damage by the percentage shown in your stat panel, it always increases it by that same amount regardless of how much damage reduction the target has. Your outgoing damage is calculated first, then the target's damage reduction mitigates that damage according to their stats.

Take for example a warrior who's swing always does 100 damage in PVE. This warrior gets a +50% damage increase from PVP Power in PVP, so that 100 damage get's increased to 150 damage in PVP. This warrior has now decided to attack a paladin that looked at him the wrong way. Lets say the paladin has +50% damage reduction; so when the warrior's 150 damage hits him it is reduced to down to 75. That is how PVP Power and Resilience interact, the outgoing damage is boosted up by PVP Power, then the total incoming damage is mitigated down by Resilience.

Section 3 - Gemming for PVP

One of the most popular questions I get is "What should I gem for?" Unfortunately I cannot provide a clear cut answer to that question. The answer is dependent on your gear level, class, spec, play style and more. My recommendation is to check how the pros of your class/spec are gemming, and then experiment to see what works best for your personal play style. However I will say that it is typically ineffective to gem PVP Power at level 90, gemming primary stats is generally more effective for damage or healing.

Choosing the best gem for survivability is significantly more complicated. Stamina gems will almost always provide you with more effective health, but Resilience gems don't lag too far behind in the effective health they add. Additionally Resilience gems scale up in value proportional to the amount of healing you receive because they increase the amount of effective health restored by that healing. For now I recommend gemming for resilience if you are looking to increase your survivability, but I will look at this topic in more depth in a future guide.

Section 4 - Closing Thoughts and Remarks

I hope that this guide has been illuminating for everyone that has taken the time to read it, I have tried to provide as much accurate and detailed information as possible about the way these stats work. If you have questions about these stats or about anything written here feel free to ask and I will do my best to get you an answer.

I error checked this guide many times however I am still human, so if you believe you see an error please let know and I will look into it. I also want to make it clear that although I am a member of Blizzard's Forum MVP program I am NOT a Blizzard employee and nothing in this guide is based on any kind of inside information. All the data in this guide was all generated based on direct in-game observation and calculations based on that observed data.

General Notes
-This guide is written for level 90 players, the numbers are different at lower levels
-This guide just underwent a major update and revision, if you notice a typo please let me know!
-Some of the graph types from previous versions of the guide have been removed to reduce the complexity of the guide, but may return in the future.

Section 5 - Appendices

This is some additional information about some of the ideas, concepts, and data discussed in this guide. Additional appendices may be added over time as needed.

Appendix A - Valuation and Frame of Reference

Absolute valuation compares numbers based on addition and subtraction, IE going from 30% damage reduction to 60% damage reduction is an absolute gain of 30%. Relative valuation compares numbers using ratios, IE going from 30% damage reduction to 60% damage reduction is a relative gain of 100%. Why is this important? Because taking the example a step further, going from 60% to 90% is an absolute gain of 30%, exactly the same as before, but it is a relative gain of 50%, half as much as before. So if this trend were to continue, it would signify linear absolute returns but diminishing relative returns. This guide primarily uses absolute valuation because it is much easier to understand and work with when comparing a large number of data points to a common baseline.

Frame of reference is another critical concept for evaluating data, particularly when most of the data is in percentages. There are two basic methods that can be used to evaluate a string of data points; constant frame of reference, or progressive frame of reference. Constant frame of reference uses one common baseline value as a reference point, every data point is compared to that baseline. In contrast a progressive frame of reference compares each data point to the data point before it. Take for example this set of data points: 1, 2, 3, 4, 5. On a graph using a constant frame of reference they would be displayed as 1, 2, 3, 4, 5 (all data points were compared to a baseline of 0). However on a graph using a progressive frame of reference they would be displayed as 1, 1, 1, 1, 1 (each data point was 1 larger than the number before it).

Using a constant frame of reference makes it much easier to evaluate the actual changes in the progression of a data set, where as using a relative frame of reference allows you to better evaluate changes in the rate of change over the progression of a data set. Since we are more interested in the actual changes than the rate of change when viewing the scaling of stats I chose to use a constant frame of reference for the graphs in this guide.

Appendix B - Formulas

The formula I use to calculate damage reduction from resilience in patch 5.4 is:
Reduction %  = ((28900*0.72)+x)/(28900+x)
x represents resilience rating.

The formula I use to calculate damage increase from PVP is:
Damage increase %  =  x/400
x represents PVP Power rating

#3752127 PROOF @hydrahacked

Posted Kunuxx on 27 August 2012 - 10:17 PM

just need to get 2 more teams out of way and admiral kunuxx shall get #1 title

i already know how to get xandyns team out, ill use this photo o______o of him and do some scanning. season 11 style

#3652816 AMA. Samulex--Current r1 and top 5 Rogue on Nightfall

Posted Porkz on 28 February 2012 - 09:27 PM

Why did u go 1-6 at NAO?

#3622282 best ever since sliced bread

Posted cgarafola on 23 January 2012 - 05:35 AM

View PostEasiestsap, on 22 January 2012 - 11:34 PM, said:

No I did not. I am probably done posting and visiting this site it is getting retarded.  People who said they quit etc. still post just to troll, this site used to be good and helpful.

You barely scraped gladiator in S9 playing rogue dk shaman on shadowburn, i know exactly who you are and how bad you are at this game, every video you've ever posted on this website is awful. Period. Dont act like a victim to the trolls, you're a one time glad and you all of a sudden get 5 helmets and a r1 helmet? Bullshit, you cant even get glad now when rogues are the most op class in the game. You trolled the OP trying to help people in the first place, then YOU got trolled, and now you are crying like a little kid. Grow the fuck up bitch.

On another note, you couldnt get glad last season, and i did ALL the GLAD(you know the title you couldnt get) rmp sales for arenarating.com  last season when it was arguably the hardest season to get glad if you didnt have a high mmr to begin with. Plus Id like to know what carries you do since you cant even get glad yourself. Do people pay you for 1800? Who even trusts a 1 time s9 mmr bugged glad for sales in the first place, good thing this is the internet and you can make up anything you want about yourself, or everyone would know how bad you are. Take a seat kid, you're a trash can.

#3621944 best ever since sliced bread

Posted Evilchronos on 22 January 2012 - 11:40 PM

View PostEasiestsap, on 22 January 2012 - 11:34 PM, said:

No I did not. I am probably done posting and visiting this site it is getting retarded.  People who said they quit etc. still post just to troll, this site used to be good and helpful.

I'm sure an average rogue like yourself can still find useful tips on this site :)

#3621466 best ever since sliced bread

Posted cgarafola on 22 January 2012 - 06:36 PM

Its ok people have been getting glad since s1, joshj didnt know that existed for him until he bought someone else's account for his own glad titles.

#3628425 1.63 Second RMP Kill

Posted Thugonomics8 on 29 January 2012 - 11:56 PM

View Postcgarafola, on 29 January 2012 - 09:42 PM, said:

IF RMP is this overpowered, why isnt it R1-10 in every single bracket? I'll tell you why, because this is a once in every 10 games experience. Stop crying cuz you got globaled, its been happening since season 5. This is nothing new, just because it happened to you because you play with a trash can rogue who can't sap priest, SHS garrote mage, and blind rogue doesnt mean rmp is overpowered, it means you or team sucks because you let them get a perfect opener. Thats like playing against a cleave as rmp and not nova'ing the dk and the ret and wondering why your priest died 3 seconds into the match because he got crit by a 50k hammer of wrath through pain suppression. If you let a team get a perfect opener, THEY SHOULD KILL SOMEONE. AKA "PERFECT" opener. I will consent and say that 1.6 seconds is a bit hasty, but technically it wasnt 1.6 seconds, theres the cast time of frostbolt, + ice lance. + global. + icelance. which is around 4 seconds, which is about the time it takes to kill a shaman in a korean swap with deep freeze.
Because RMP can only be in the 3s bracket?

#3453709 Rogue Mage Priest

Posted Hawtstuffs on 25 August 2011 - 06:04 AM

This comp is extremely bad there are so many things that need to be fixed in order to make this comp viable. There are way to many flaws in this comp for it to even be a 2200 capable comp.

-Kidney shot and deep freeze should not share DR, doesnt really need an explanation
-Cheapshot should not share DR with anything, rogues can't even use this ability anymore because it's so bad.
-Blind and Psychic scream should not share DR, it messes all of the CC chains up
-Sap and Sheep shouldn't share DR, if you're bad enough to get sapped you deserve to be CC chained
-Smoke bomb is way to broken this ability needs to be looked at badly, you can simply just walk out of it, Smoke bomb should apply a root on the players that are afflicted with the smoke bomb debuff, kind of like elemental shamans earth bind totem

#3396273 BiS s10 gear

Posted clusky on 26 July 2011 - 02:33 PM

View PostSolleks, on 25 July 2011 - 05:25 AM, said:

I'm too lazy to do it, who wants to make me a profile?

21/21 exp,
5.00% hit (or as VERY close as you can get)
crit reforged

UHW+heroic Rag trinket

while maintaining 3500 + resil
probably won't be able to use tier legs if using 2x PvE trinkets to stay above 3500 resil



#3379894 Anyone tried this?

Posted ragnarok3X on 19 July 2011 - 02:58 PM

Was bored and looking thru some mods on curse and i came across this http://wow.curse.com...ils/bgglad.aspx and was wondering if anyone has tried this for rbgs?

#3369894 Shadowpriests are fair

Posted Mageic on 15 July 2011 - 09:47 AM

View PostZerud, on 15 July 2011 - 09:12 AM, said:

Care to explain why?
You didn't do lower damage then mages last patch and only slightly lower overall damage then locks. This patch you don't even do less damage and the damage was traded for burst which in most situation can cause way more pressure then overall spread damage. If you lose AS a spriest you fucked up and didn't heal enough.

You act as if this damage is completely random and your mindblasts just randomly range from 10k-70k lol; you have cooldowns and if you're good you will use them at the right time when all your procs are up which btw are extremely easy to line up all you really need to do it is 1 mindflay after you have 3 orbs, pop your trinket/wings macro and WOW THE STARS HAVE ALIGNED FOR YOU.

#3328566 Only 3v3 R1 Hunter US?

Posted Azaelz on 28 June 2011 - 11:39 AM

Oh, btw since we're on the topic... I was just curious if any 22 year old Caucasian Males, who are Canadian and live in Southern Ontario near the border to New York State got rank 1 as a Warlock while playing on their MSi laptop as Shadowplay?  Nope??  Well I guess that just leaves me...

#3328546 Only 3v3 R1 Hunter US?

Posted Azaelz on 28 June 2011 - 11:33 AM

No one cares.


#3321365 Rogue Cleave Tunnel Vision Syndrome

Posted Azaelz on 25 June 2011 - 05:56 AM

View PostSplash, on 25 June 2011 - 04:56 AM, said:

Whenever I try playing DK Rogue X or War Rogue X , I always tend to play a lot worse than I normally do and generally make more stupid mistakes. I can easily function at a 2600+ level while playing traditional comps such as RLS or RMP but whenever I am paired with another hung two-handed wielding lad my IQ dissipates by like 20 points. I am calling this RCTVS. Maybe it is just me and playing brainless zerg melee cleaves is not on my threshold of intelligence but I have no other idea what it can be.

Anyone else suffer from the same cause, and how I can find a cure for this?

Everyone feels that way because you actually have to play worse to do well as cleave because it's generally better to just do shit like spam all your cooldowns instantly and tunnel vision the shit out of w/e the ideal target is.  It's two totally different play styles so you have to really dumb down what you're doing and Rogues who're off kicking and and sapping off of other CCs etc etc can actually be a detriment to their team.

I know Reckful says the same thing and he actually is a lot worse at even interrupting when he's Cleaving just because he has to be such a dumbed down version of himself to play the setups well.


Posted Azaelz on 25 June 2011 - 01:41 AM

View PostDelirious, on 22 June 2011 - 10:04 AM, said:

Their mmr is still higher than Talbadar's

As for Coachvollex... LOL

Gratz on Gladiator once again bro


Not too worried :P