There is a lot of discussion about the balance between power and conditions. Here is some maths as well as some personal opinion on the subject. Since I believe the question of boon vs condition is also related, I added it to the discussion. The takeaway is that things are not as bad – as some people think, but they probably still need a rework.
How many stats for power and conditions
One of the most argued thing about the power vs condi balance is that condi only need 1 stat (or 2 including condi duration) while power needs 3. Therefore condi can more easily add defensive stats, making the gameplay « easier ». The truth is not that simple for the very simple reason that there is no such thing as « condi » weapons or « condi skills » (or very few at least) but usually « hybrid » weapons or skills. So to reach the max damage, a hybrid skill needs not only condition damage and condition duration but also power stats. And you can actually balance the condi/power ratio of skills to make them balanced to a corresponding full power skill.
One important difference is that condition ignores armor. This is a very important feature and actually a very nice one. We do not want power and condi to be equivalent, ideally, both should fill a slightly different niche and adds some depth to the playstyle and in particular the team composition choice. So thanks to this armor-ignoring characteristic, the niche for power and condi is obvious: condi should be better to take down « bunkers » and power to take down « squishies ».
In this section, I will compare a power build to a hybrid build and see if we can have a power/condi ratio on the hybrid build to that both builds are balanced for an equal number of offensive stats.
Damage output = « Coef » * Power * (1 + crit chance *(crit damage – 1))/armor
Condition output = (condi base + condi multiplier * condi damage) * (1 + condi duration)
Let’s take for example bleeding for our condition damage and assume a reference damage « coefficient » of 1000 for the power build (not to be confused with the weapon damage coefficient which needs to be multiplied by the weapon coefficient).
At level 80, in PvP, armor goes from 1920 for light armor without toughness to 3411 for heavy armor with 1200 toughness. The average value is about 2600, which corresponds approximately to a medium armor with 560 toughness. Therefore, I will now assume that condition and power should be balanced for 2600 armor, power dominating for lower armors and condition dominating for higher ones.
We now get
Damage output = 1000 * (1000+Power) * (1 + (precision+84)/2100 * (ferocity/1500 + 0.5 ))/2600
Bleeding output = number of stacks (or duration in seconds) * (22 + 0.06 * condition damage) * (1+expertise/1500)
With this values, a full power build (1000 damage coefficient, 0 bleeds) gets:
- Berserker amulet: 1282
- Paladin amulet: 1001
- (old) Soldier amulet: 863
It is then easy to find a value for a hybrid build which matches those approximately for corresponding amulets : I chose 600 damage coef, 6.5 bleed stacks
- Viper: 1304 (58% of it being condition damage)
- Mercenary : 1035 (53% being condition damage)
- Dire: 846 (72% being condition damage)
This value can be tweaked, but it is easy to see that the difference is overall minimal. So one can easily change a power weapon to a hybrid weapon by reducing the damage output by 40% and adding 6.5 bleeds per initial 1000 damage.
For the sake of clarity, I have to mention that using a power amulet on a hybrid build ends up giving lower output while the same is true with hybrid amulet on power build. So in the future, when I mention power amulet, it is implied using a power build, and hybrid amulet on hybrid builds.
The take-home message from this is that the way stats work, you can get balanced damage output despite the alleged « less stats ». Actually, as you can see, with only damage stat (the others being defensive stats), power is a (slightly) better investment.
One additional point which needs to be mentioned is damage multipliers vs condition duration increase. Power damage is usually increased by traits adding some damage multipliers. They are usually on the order of 10%. On the other hand, there are many traits/runes (and food in other game modes) that increase condition duration by 20% or more. If we balance the base power vs condi balance, those « multipliers » also need to be balanced. It does not need to be reduced to 10% (since unlike power builds, in hybrid builds only a portion of the damage comes from conditions or power) but it definitely needs some balancing considerations.
How boons/debuff affect the output
Another important question is how do damage-augmenting boons (might, fury), and damage reducing boons and debuff (protection, resistance, weakness) affect the overall power/condi balance.
Using the same values as above for power build and hybrid builds
25 Might:
- Berserker: 1719
- Paladin: 1368
- Soldier: 1157
- Viper: 1906
- Mercenary: 1504
- Dire: 1315
You see that might strongly favors conditions over power. This is easy to understand from the formulas:
Power damage is proportional to 1000 (base power)+ power
Bleeding damage is proportional to 22 + 0.06*Power.
1000 power doubles the power output while 1000 condition gives 82 which is much more than the double of 22. If bleedings were instead 60+0.06*power, might would have exactly the same effect on power build than hybrid builds. In general, the base condition damage should be 1000 times the condition multiplier for might to be balanced. Actually burning is much more balanced on this perspective.
Technically, we don’t need to have this. There is nothing wrong with might favoring hybrid builds (as we can see later, fury naturally favors power builds). It even makes sense since might needs build up, just like condition damage, so both playstyle favor slightly longer fights while power (and fury) are more immediate. But I feel that for bleeding, this is just too much and the base tick should be increased (to maybe 45-50).
Fury:
- Berserker: 1468
- Paladin: 1080
- Soldier: 948
- Viper: 1352
- Mercenary: 1082
- Dire: 869
As expected, fury favors power builds, but I feel that unlike might, this favoring is not excessive and I think it is good as it is.
Protection:
- Berserker: 859
- Paladin: 671
- Soldier: 578
- Viper: 1124
- Mercenary: 876
- Dire: 769
Resistance:
- Viper: 546
- Mercenary: 483
- Dire: 235
Protection is a very powerful tool against power builds. The effect on condition build is milder. On the other hand, resistance is just a slaughter for hybrid builds. About 50 to 70% of the hybrid build damage comes from conditions, so resistance corresponds to a 50% to 70% damage reduction. If instead we change resistance to 50% damage reduction from conditions (or 50% reduced duration for incoming conditions) we get
« half » resistance:
- Viper: 925
- Mercenary: 759
- Dire: 541
« half » resistance + protection
- Viper: 745
- Mercenary: 600
- Dire: 463
I think this is a much healthier value. In general, I think condition negation should be stopped and condition mitigation should become the main thing. I will discuss this a bit further later. Of course, with resistance effectivity halved, resistance should become a more prevalent boon.
Some people suggested 33% condition damage reduction instead of my suggested 50%. But this would imply that to get the same reduction as full power builds, we need BOTH resistance and protection. This thus favors hybrid builds. With my suggestion, having only resistance goes a long way to reduce the damage taken (but not as much as protection for power builds), and protection + resistance is a mild overshoot.
Weakness:
- Berserker: 853
- Paladin: 698
- Soldier: 643
- Viper: 1150
- Mercenary: 912
- Dire: 786
We can see that weakness is not too different from protection (but more heavy reduction for berserker builds than for low crit chance). On the other hand, for sake of balance, we can feel sad that there is no corresponding condition/debuff for reducing condition damage output.
So overall, for balance, the base tick of some conditions should be increased slightly so that might does not get overly stronger on hybrid builds compare to power builds. Burning is a good example of a « balanced » condition from that perspective. Also, resistance should have its efficiency reduced (for example by half) and a debuff should be added that reduces condition damage output.
Condition cleanse and the boon/debuff balance
While I didn’t mention it specifically, all of those maths above only make sense if you assume condition do not get cleansed. If they do, then that will of course reduce the condition output and favor power builds.
This is I believe the problems we have currently with conditions. In order to be balanced, the cleanse have to be taken into account, and conditions have to be made comparatively stronger.
Overall, negations make the game unhealthy. So condition cleanse/resistance are unhealthy, the plethora of blocks and reflects is unhealthy etc…
An additional problem is that there is really 2 categories of conditions:
- damaging conditions (bleeding, burning, confusion, torment)
- debuff (blind, chilled, crippled, fear, immobilize, poison, slow, taunt, vulnerability, weakness)
Poison is technically hybrid, but I consider it mostly a debuff.
Because the main way to handle condition damage is condition cleanse, we have now an lot of cleanses and AOE cleanses flying around and debuff are very fast removed. Because of this, they cannot play their role as a boon equivalent.
My suggestion (and I am not the only one to suggest it) is simply to split those 2 things. Then, a good way to proceed (according to me) is to make condition cleanse only affect debuffs (and reduce significantly the amount of cleanse available). The soft CC debuff (especially chill and immobilize) probably need additional specific cleanses (it is already the case on a few classes, and taunt and fear can already be stun-breaked) considering they can be more dangerous (more CC than debuff). You would then have boons and debuffs mirroring each other, with condi cleanse and boon rip also mirroring each other. We would get both power vs condition damage balance (from above arguments) and boons/debuff balance.
A small (but nice) benefit to this is that we could start to gray out condition damage in our HP bar, allowing to more easily see he damage we take for each enemy skill and learn to react to it, and also just get an idea of how low our actual life is.
But the most important point with this balancing is adding build diversity. You would now have builds supporting their team thanks to boons or thanks to debuffs (currently, debuff support is much weaker than boon support such that necromancer or thief is considered unviable in a support role). Some classes have natural preference for boons (elementalist, guardian, engineer), some have preference for debuffs (necromancer, thief). Some have mixed (mesmer for example often has this duality). On the other side, some classes can AOE cleanse conditions (guardian, necromancer) while some can AOE boon rip (necromancer, mesmer, thief should probably get that too). You also have classes which can share protection to everyone (elementalist/guardian), some which can share resistance (revenant, mesmer), some which can spread weakness (necromancer, thief) some which can spread whatever new debuff to reduce the condition damage output.
This creates a great rock-paper-scissor which puts more emphasis on strategy in team comp selection and is much healthier than the current system.
(edited by Silverkey.2078)