Thread: 4.11+ DRK Suggestions/Balance Changes from a "Hardcore" Player

Originally Posted by Lyth

Again, it's not relative to one number...

Then let's use every savage fight in creator and SB, assuming a 95 percentile skill level, since you don't want to use the average of all of them combined into one meaningless statistical number.

A9S
Drk: 97%
Pld: 82.8%

A10S
Drk: 97%
Pld: 88.2%

A11S
Drk: 100% (warrior is actually 99% here)
Pld: 93%

A12
Drk: 98%
Pld: 86.9%

See a trend? Now here's Delta:

O1S
Pld: 98%
Drk: 92.7%

O2S
Pld: 100% (tied with warrior)
Drk: 92.7%

O3S
Pld: 96%
Drk: 94.3%

Neo
Pld: 97%
Drk: 95.2%


These are the fights people are concerned with when discussing that drks are the weakest. All of these numbers are in comparison each tank's performance in each savage fight in comparison to who performs the best. These are the numbers that matter. It gets averaged into what I had listed in my earlier post, but since you wanted to see every individual fight... ¯\_(ツ)_/¯

Originally Posted by Zykor

These are the fights people are concerned with when discussing that drks are the weakest. All of these numbers are in comparison each tank's performance in each savage fight in comparison to who performs the best. These are the numbers that matter. It gets averaged into what I had listed in my earlier post, but since you wanted to see every individual fight... ¯\_(ツ)_/¯

Wouldn't it be more accurate to use Alexander 1-4?

Fact of the matter is that expansions are pretty hard resets, and even though Omega is 'level 70' content compared to 60, Omega 1-4 is still entry tier and thus should be compared to Alexander 1-4.

Originally Posted by Lyth

My concern isn't about who suffered the most historically. My concern is about the impact of balance decisions on present pick rates. That's why I take issue with the way that you chose to represent your data.

I'm not even going to bother to pick apart how irrelevant absolute numbers are for different expansions. Use percentages if you want an accurate comparison between different thresholds. Absolute numbers are a trash comparison for a realistic gauge. If you still disagree then let's agree to disagree because I am baffled that you still cling to this argument.

As for overall usage, what a terrible measurement for balance lol. Take SB machinist for example, they currently need zero buffs and are meta, yet they are the least played class. But I have heard many times that they are very unfun to play from people that once enjoyed them.

I've also heard how drk was very unfun to play because of how much more management a person had to do back in HW, which is beyond the capabilities of many players in this game.

My only discussion regarding previous woes for tanks was in regards to the specific comment that balance is far worse now than it was in creator, which is simply false :eyes: Drks are in a bad spot now, I've never argued that. Them getting picked significantly more now than paladins did in creator doesn't mean that a less than a quarter pick rate for drks in SB is okay. EDIT: In terms of speed kills I should add, which was always what I'm basing my stuff around.

Originally Posted by Zykor

My only discussion regarding previous woes for tanks was in regards to the specific comment that balance is far worse now than it was in creator, which is simply false

The size of the gap between the highest-performing and lowest-performing tank is only half the story when it comes to balance. Consider a few hypothetical situations:

1) Tank A is better than Tank B 50% of the time, and worse than Tank B the remaining 50%. The gap in performance is 15%.

2) Tank A is better than Tank B 75% of the time, and worse than Tank B the remaining 25%. The gap in performance between the two is 10%.

3) Tank A is better than Tank B 100% of the time; Tank B is always a worse choice. The gap in performance between the two is 5%.

Case 1 is definitely balanced. Case 3 is definitely not balanced. Case 2 is not perfectly balanced, but is certainly much closer to being balanced than Case 3.

It is much more important, from a balance standpoint, that each class have particular situations and niches in which they shine, than it is for the gap in performance to just be made smaller.

Originally Posted by Crater

[Them there balance words]

The main problem with that is what if one tank's niche is Savage raids where balance actually matters and one is The Diadem?

Originally Posted by Metsonm

The main problem with that is what if one tank's niche is Savage raids where balance actually matters and one is The Diadem?

They can be built to have niches that are distinct, unique, and relevant, all within Savage raiding. It just requires that SE be willing to promptly (more frequent updates than one balance pass every three months that touched 3-4 classes) and decisively (ie: piss off with these 10-20 potency buffs that don't do anything but reduce the magnitude of how overpowered/underpowered one class is compared to another).


Even if we're just balancing the tanks purely by their damage, it's easy to create a meaningful, balanced class dynamic that gives every tank a niche without falling back that retarded "Welp, guess one of the tanks is just always going to be out of the meta" talking point, by considering that there are different types of damage, and they should all be weighted differently.

For example:
- Burst damage should be weighted more heavily than sustained, consistent damage, because burst damage has distinctive utility: It can be used to push phases (see: All of the various skips that could happen back in A11S); it has more synergy with raid buffs; it can be used to have a tank take over a DPS role (see: For many groups back in A7S, the Warrior would take a melee jail); and it becomes stronger when a fight has long phase transitions or downtime (because you can use the downtime to hide the cooldown on your burst, improving your effective burst uptime). The tank that does the highest burst damage, should do the lowest sustained damage.
- Personal damage should be weighted more heavily than damage added through raid buffs: Personal damage functions at full effectiveness even if the rest of the party sucks or is underperforming; raid buffs are reliant on having a whole party putting out good DPS to function.


And so you could, for instance, do something like this:
- WAR does the highest burst damage, but the lowest total DPS (put Berserk back to 50%, buff Fell Cleave, but nerf the Storm's Eye damage bonus and WAR's standard combos)
- PLD has moderate burst, and the highest personal DPS.
- DRK has little to no burst, and personal damage in between WAR and PLD, but offers a raid damage buff that is greater than the difference between DRK and PLD personal damage (constant, not burst; think something like the passive 2% Crit bonus that all Bard songs give with 100% uptime, and not Trick Attack or Battle Litany which are burst skills)


In that case, WAR has the lowest overall DPS on a dummy, but can be instrumental to giving your group easier phase skips and phase pushes, which can result in both easier progression and faster clear times (depending on which mechanics are skippable), can potentially have higher DPS in fights with long stretches of downtime (since Berserk has more effective uptime as the fight's downtime increases), and WAR would have the unique ability to stand in as a melee DPS for situations like the A7S jails, or even possibly something like A8S Steam Regulators, because of how high they burst.

PLD would become a good all-around tank, damage-wise, and could contribute a fair bit to phase skips and pushes (but not nearly as much as WAR), while also adding the most overall DPS to groups where people underperform, or if the party splits up and raid buffs won't cover everybody.

DRK would be the go-to for total sustained raid damage, in fights with little downtime and where the party never needs to split targets - but would fall short of the other two tanks in fights where burn phases and phase skips are important to fast clear times.


If the balance were handled that way, then not only would every tank have their own niche and speciality, but every tank combination would also have a place, both in progression and for high-end clears.

WAR/X would be preferred for fights like A11S with lots of skips and lots of downtime, and would be a strong composition for fights like most of Midas; PLD versus DRK in the remaining slot would vary depending on whether PLD's help made a different for phase pushes, how much mileage you got out of DRK's raid buffs, etc. But for fights like v3s or A10S, or even going back to A3S, DRK/PLD would be preferred, because the fights are full-party slugfests with basically full uptime, with an emphasis on total DPS rather than burst.

Obviously that's not the only way you could do it, but the big takeaway here should be that it's really fucking stupid for one tank to have both the best burst damage, and the best total DPS, and that priority #1 for tank balance should be to give those advantages to two different tanks, and then balance the third tank as a happy medium.

Originally Posted by Aana

...

Amusingly enough, every normalisation approach, including the above decision to express dps as a fraction of WAR, is percentile specific, fight specific, and ilvl specific. The amount that you consider to be "significant" is a matter of opinion. The only reason why you'd want to normalise it is because 5% or 3 normalisation points looks smaller than 250 dps.

Some people love the present tank balance. Others dislike it. If you look at the jobs that people actually play, it's extremely unsurprising who fits into which group. Yourself included.