DPS discrepancy between classes (courtesy of ESO Logs)

Vildebill

MLGProPlayer wrote: »

Vildebill wrote: »

MLGProPlayer wrote: »

Vildebill wrote: »

This thread is the perfect demonstration of how people in the world perceive facts. Extremely biased and only in the direction of the person's own interest. Source criticism seems to be the hardest thing to learn.

What's the context of the data? Who are the players? What buffs were available? How good are they at their role (and how do you even measure that?)? Consistency? When I see this graph 100 questions pop up in my mind, questions that needs answers before I can draw an even somewhat precise conclusion. Without that information, the data is worthless.

But yeah, the world is burning, everything is a mess, nothing ever gets better, it was all better before, come and vote on my extremist conservative party and all the world's troubles will be solved. It has worked perfectly fine everywhere, I promise you

It's trials data from the top 1% of performers. It's extremely telling, looking at both raw performance numbers and usage rates, that there is a huge discrepancy between classes.

Yeah probably just as telling as the doomsday Altmer nerf you so happily spread across the forums for over a month, like a prophet in black robes. And look how that turned out.

I'm sorry if I'm sounding harsh, I don't know you in real life, but by the you you always sound here on the forums, talking negatively on a lot of stuff that turns out to be not even remotely as bad, gives me a hard time to take you seriously.

Racial balance was (and still is) an issue that needs attention.

Class balance has been an issue since day 1. We just have more tools than ever before to pinpoint the problems. Before ESO Logs, all we had to work with was in-game leader board data and personal anecdotes. Now we have raw evidence to support those personal anecdotes. The issues can't be ignored anymore.

I get that fanboys/girls will always say that issue X isn't an issue and that people should just shut up and play the game. But that's how games turn to ***. There needs to be informed discussion on the issues.

Racial passives has never been this close, you can say what you want about that. And all races will never perform equally good on all roles. What's the fun in that anyway?

Yeah, we need to have a discussion based on real facts, statistics and tests. You're not really contributing to that. If players like Liko, who standardize his tests with the same debuffs, and obviously are very consistent in his skill and rotation, can pull far better numbers than your graph can, what conclusions can be drawn by that? That your graph and data does not represent consistent player skill, buff uptime, gear, racials, and a million other variables.

You draw conclusions based on shallow facts. And you are spreading misinformation.

Runefang

MLGProPlayer wrote: »

Vildebill wrote: »

This thread is the perfect demonstration of how people in the world perceive facts. Extremely biased and only in the direction of the person's own interest. Source criticism seems to be the hardest thing to learn.

What's the context of the data? Who are the players? What buffs were available? How good are they at their role (and how do you even measure that?)? Consistency? When I see this graph 100 questions pop up in my mind, questions that needs answers before I can draw an even somewhat precise conclusion. Without that information, the data is worthless.

But yeah, the world is burning, everything is a mess, nothing ever gets better, it was all better before, come and vote on my extremist conservative party and all the world's troubles will be solved. It has worked perfectly fine everywhere, I promise you

It's trials data from the top 1% of performers. It's extremely telling, looking at both raw performance numbers and usage rates, that there is a huge discrepancy between classes.

Can't believe I have to explain why the top 1% is a terrible stat to look at, especially right now with a tiny sample size.:

1. Some classes have 300 parses (some less!) so top 1% equals 3 parses.
2. If those 3 parses were from groups that didn’t have the best war horn uptime or didn’t have many Necro or didn’t get that major berserk from a storm atro etc. The smaller the sample size the more likely that’s true.
3. Top players are in the best groups running the best dps classes it's even more likely that all the top stamcros and stamblades get the best buffs while the best magdens and best stamplars do not.
4. Top groups are more likely to upload their logs because they actually bother to log period, these top groups are then more likely to only use the best classes. It creates skew in the sample set.

Imagine a trial group decided they’d help one member achieve maximum dps. Have all storm atros drop on them so only they get major berserk. Time it with a flesh atro and a major slayer buff for a huge dps buff window. Surround them with war machine and master architect so they can run another set besides Lokke. Give them guard to save them casting a minor force buff. Add powerful assault as well. Find a way to give them all possible damage synergies. That parse would be enormous and skew the 1% because it introduces a significant outlier. That's an extreme example (that I 100% expect to happen one day, I look forward to the day Hodor or other do that) but it illustrates a point. Small sample sizes don't provide you with the right confidence level that the result is representative of the overall population.

The Uninvited

If you check the log now, Stamina Sorc is in the lead and Magicka Necro is coming in at third place.

All this says. is this needs more time to have a somewhat realistic view of the current situation.

killahsin

The Uninvited wrote: »

If you check the log now, Stamina Sorc is in the lead and Magicka Necro is coming in at third place.

All this says. is this needs more time to have a somewhat realistic view of the current situation.

Correct the data provided and the way it was provided did more to hurt the argument than it did to help it. This is not how you look at class balance.

Uryel

That testing is done in Sunspire, a 12-person group instance. Now try the same in 4-people area, or even better, solo-duo, and you'll see Nightblade drop in the depths of the chart.

The latest "balance" patch that spreads debuffs and buffs accross the board especially for large group content has little to no impact on Nightblade damage in large groups, as there will always be someone who brings some buff / debuff. For small groups, it's really bad.

As a Stamblade, my most damaging attack no longer debuffs enemy armor. What used to give me 8% more damage doesn't anymore, it buffs defense instead (and to add insult to injury, it's located in the "assassination" skill line !). Even now that they rely on the highest ressource pool, shades are still useless for damage.

As a person who mostly plays duo with my wife, who plays a magblade, we can see the difference in small group content : we've been nerfed again.

Runefang

MLGProPlayer wrote: »

MaxJrFTW wrote: »

Runefang wrote: »

MLGProPlayer wrote: »

Runefang wrote: »

MLGProPlayer wrote: »

Runefang wrote: »

That graph doesn’t show what you think it does. The dps difference from top Stam to bottom Stam is around 10k dps. Same for mag. Difference between top Stam and top mag is also 10k.

Here is the per second data. The differences are still astronomical. I filtered for the 99th percentile to ensure we're only getting the absolute top parses.





Overall DPS difference: 24k
Magicka: 15k
Stamina: 11k

Balance in this game is a joke.

First you picked your world view. Then you picked the data to match it. Good job.

Data doesn't lie.

No but why did you pick 99th percentile over the 95th then? Because it told the story you wanted to tell.

Because it shows what the top 1% players with their respective classes are capable of doing. Why would you want to lower the bar?

This.

You can't objectively look at balance without looking at the top tier of players. These players get the most out of their classes.

Less skilled players are not representative of the maximum damage output of a class.

I guarantee the person who has the best stamcro dps log so far could log onto a magden and outdo the current best magden parse. And given there is only 59 magden parses on there, they alone would be the top 1% thus improving the result radically. They won't do that because they are in a trial group that pushes scores and so they'll stick to the stamcro.

I don't disagree that class dps disparity is an issue, it's just not nearly as bad as you think. It's a bit like when climate changers post a photo of the artic with the ice melted in the middle of summer to illustrate the ice is disappearing. Like no kidding, that happens every year. Facts and data can be used to lie just as easily as tell the truth.

Runefang

The Uninvited wrote: »

If you check the log now, Stamina Sorc is in the lead and Magicka Necro is coming in at third place.

All this says. is this needs more time to have a somewhat realistic view of the current situation.

You're probably filtering on the data completely differently thats all. Magcro in its current state isn't coming close to the top of any dps leaderboard.

The Uninvited

Runefang wrote: »

The Uninvited wrote: »

If you check the log now, Stamina Sorc is in the lead and Magicka Necro is coming in at third place.

All this says. is this needs more time to have a somewhat realistic view of the current situation.

You're probably filtering on the data completely differently thats all. Magcro in its current state isn't coming close to the top of any dps leaderboard.

Exact same filters used.

Elusiin

SirAndy wrote: »

MLGProPlayer wrote: »

Runefang wrote: »

First you picked your world view. Then you picked the data to match it. Good job.

Data doesn't lie.

Whatever you do, please never ever get a research job ...

You only get a research job if you go through years of school. PhDs aren't handed out like political positions are to unqualified individuals, or like candy. But yes, unless we know how the tests were conducted and the sets used, this data is useless.

TriangularChicken

This..

joseayalac wrote: »

This is fine, why should all the classes have the same dps? not every class is as good when it comes to healing or tanking, why should it be the case for dd?

Almost every class is viable to complete content as a dd, so I think balance is good.

and ..

colossalvoids wrote: »

That is what you'll get when just looking at pretty colored thingys without any context, right. T3hg already said it all for those who unable to think by themselves.
Different specs, different builds, different player skills, group buffs yada yada.

!

People forget that the newer trials involve way more mechanics than the old ones do. The bosses are not target dummies (where the dps is very close for most classes), people have to play mechanics, a mag dk with whip as spammable will have a more difficult time to dps when he has to play the ice mechanic at Lokkestiiz than any other ranged class that can start with their rotation right after leaving the ice...just one example of many.
I love comparing numbers, but this chart isn't telling as much as you think.

Tommy_The_Gun

Wait a second... DK has better dps than NB... And DK is also superior at tank role ?! WUT ?!

david_m_18b16_ESO

J2JMC wrote: »

It's ridiculous that people look at the dps a class does in a 12 man trial and make definitive conclusions on the balance of the game as a whole. I don't know why there is an expectation for classes in an MMO to all be able to dps, tank, and heal the same way. In a game that features 12 man trials, 4 man instanced dungeons, open world dungeons, ZvZvZ pvp, 5v5 pvp, and whatever else I am forgetting, I would expect some classes and their variations to be better at certain things than others. In fact, I wish ZOS would double down more on giving classes and their resource variations uniquely defined strengths and weaknesses.

Vildebill wrote: »

This thread is the perfect demonstration of how people in the world perceive facts. Extremely biased and only in the direction of the person's own interest. Source criticism seems to be the hardest thing to learn.

What's the context of the data? Who are the players? What buffs were available? How good are they at their role (and how do you even measure that?)? Consistency? When I see this graph 100 questions pop up in my mind, questions that needs answers before I can draw an even somewhat precise conclusion. Without that information, the data is worthless.

Those questions all have been awnsered.

david_m_18b16_ESO

MLGProPlayer wrote: »

MaxJrFTW wrote: »

Runefang wrote: »

MLGProPlayer wrote: »

Runefang wrote: »

MLGProPlayer wrote: »

Runefang wrote: »

That graph doesn’t show what you think it does. The dps difference from top Stam to bottom Stam is around 10k dps. Same for mag. Difference between top Stam and top mag is also 10k.

Here is the per second data. The differences are still astronomical. I filtered for the 99th percentile to ensure we're only getting the absolute top parses.





Overall DPS difference: 24k
Magicka: 15k
Stamina: 11k

Balance in this game is a joke.

First you picked your world view. Then you picked the data to match it. Good job.

Data doesn't lie.

No but why did you pick 99th percentile over the 95th then? Because it told the story you wanted to tell.

Because it shows what the top 1% players with their respective classes are capable of doing. Why would you want to lower the bar?

This.

You can't objectively look at balance without looking at the top tier of players. These players get the most out of their classes.

Less skilled players are not representative of the maximum damage output of a class.

I suspect that most ESO arent used to group parse. So many post going against commun knowelge of how to read raid parses.

FakeFox

Werewolf is not a functional build and should not be included. Otherwise you should also include drunken Nord fistfighters and alike.

code65536

Tommy_The_Gun wrote: »

Wait a second... DK has better dps than NB... And DK is also superior at tank role ?! WUT ?!

1. Where were you in the days prior to Morrowind when DK was the better tank and better DD and nightblades were irrelevant outside of PvP?
2. Actually, nightblade tanking is pretty strong right now. Specifically for maintanking Sunspire, it actually outperforms DK.

On a more serious note, the magblade fall from grace is quite startling.

Massacre_Wurm

FakeFox wrote: »

Werewolf is not a functional build and should not be included. Otherwise you should also include drunken Nord fistfighters and alike.

Do we have drunken Nord skill lines ? Or drunken Nord sets ?

PaulD

This has some meaning re a particular trial, not a lot, insufficient separation of data to equalize buffs, buffing, serious groups vs tryouts, etc. For pity's sake, you've got me in there on a pet sorc that's set up to do solo VMA, which requires more health than you'd normally see in a trial. (Folks were trying to fill out a 12 man and didn't care.)

Moreover, it looks like this is tilted towards burst, not sustained damage. And Sunspire does seem to favor burst over sustain. Some trials do not.

Just because it's data doesn't make it meaningful data for general comparisons.

Tommy_The_Gun

Just one question OP.
Were those DPS results achived on a Vamp characters ?

Narvuntien

MagDK is a close range dps which means it's okay that it does more damage than NB.

I think it might be a bit low because you kind of had to relearn how to play the rotation with the whip changes last patch, I'd of expected its dps to be higher.

I think mag Warden, Necro and NB are a bit on the low side though and Stamina Templar is a huge outlier for stamina builds (the only one that is below its mag version)

sirpz

MLGProPlayer wrote: »

Difference between overall top and bottom DPS: ~24k

Difference between top and bottom stamina DPS: ~11k

Difference between top and bottom magicka DPS: ~15k

At ZOS HQ:

This data is literally useless. Show me 6mil or iron dummy parses for each class done at least 10 times. Every encounter is different, this is also taking into account an extremely wide spectrum of players. We need consistency. You're not gonna have many top end players playing as WW, or you have necros so high because the only people playing necros in vet trials already are mainly the better players who powerleveled and already had BiS gear waiting for them. People freaking out over this are rediculous, if there was a 35k difference between top and bottom dps on dummy's I'd have a much bigger concern.

FearlessOne_2014

I henceforth refuse to support ESO until ESO and ZOS start showing some support to Mage-based playstyles. Simple as that.
Until then as ESO reserve the right to discriminate against Magic based players. I reserve the right to take my wallet elsewhere that support it.

Your loss ZOS, Koreans gain.. GG WP ZOS not!




That was going to be money going to you. But since looking at the current patch notes. I decided to cancel my ESO Plus for the near future and redirect that money to Korean who respec their Mage playerbase.

What happened to play your way ZOS? Why lie about that? It what was making ESO a gem in the desert of MMOs.

REKT

Miraslova

First, the results can be easily fabricated, with that out of the way, rotation is a big factor in DPS so there is no accurate way to establish if some are doing a rotation effectively.

FakeFox

Massacre_Wurm wrote: »

FakeFox wrote: »

Werewolf is not a functional build and should not be included. Otherwise you should also include drunken Nord fistfighters and alike.

Do we have drunken Nord skill lines ? Or drunken Nord sets ?

Why do elitists force me to wear armor and use skills? I want to play the way I want!

Point is there are many theoretical playstyles that simply don't work for obvious reasons and there is no point in including them. Werewolf has a skillline and fitting gear, however that does not make it a functioning build that should be a part of endgame DPS. Werewolf does not have proper DPS Skills and a rotation that is literally one button. So in terms of rotation Werewolf and drunken Nord are the same. But thanks for questioning obvious jokes.

PaulD

This whole thing is reminding me very much of a character that popped up on a political forum a few weeks ago who was claiming that since average life expectancy at the time Social Security was started was 61 (IIRC, I don't make any effort to remember silly stuff) that this proved Social Security was a scam from the start.

OFC, that is absolutely ridiculous, I gave him the link to the official standard arcturial tables for the 1920 - 1929 era, and calmly explained that the AVERAGE is grossly skewed because about 10% died by age 11 or so, as was shown by the tables. (If you get a five on a test, and do well on all the others, that one test is going to skew the crap out of your grades.)

Taking those who actually survived reaching age 11 (and they still had a pretty high death rate due to childhood diseases and polio and smallpox and scarlet fever to age 18 or so) and calculating the percentage that reached age 65, I got over 2/3's that made age 11 were going to live to be old enough to get full Social Security. Presuming, OFC, they repeated the 1920's over and over and didn't move to the 70's when they got to 65 so they could take advantage of antibiotics and actual surgery, ya know.

Even back then, under ten's weren't part of the work force. In fact, that was when the child labor laws were passed.

This data is much the same case. It's grossly skewed for one trial, and moreover there's no compensation for far too many things that make huge differences. And there are far too few parses for many classes, so few that the data isn't actually comprable.

daemonios

FearlessOne_2014 wrote: »

I henceforth refuse to support ESO until ESO and ZOS start showing some support to Mage-based playstyles. Simple as that.
Until then as ESO reserve the right to discriminate against Magic based players. I reserve the right to take my wallet elsewhere that support it.

Your loss ZOS, Koreans gain.. GG WP ZOS not!




That was going to be money going to you. But since looking at the current patch notes. I decided to cancel my ESO Plus for the near future and redirect that money to Korean who respec their Mage playerbase.

What happened to play your way ZOS? Why lie about that? It what was making ESO a gem in the desert of MMOs.

REKT

You know why ZOS and other companies keep doing the bare minimum? Because people like you keep spending hundreds of dollars at a time on fake currencies to buy shiny pixels. If it were all about subscribers, they would have to get their act together and provide actual balance and performance improvements. I'm pretty sure in a few weeks or months you'll be saying the same about BDO and coming back to buy more Clown Crates in the Clown Store.

The_Lex

People really should consider taking a statistics course before trying to compile and interpret statistical data.

caperon

So, how about grabing single target dmg data? You will see a lot of stamblades and some odd stamdk on the top places.

The only issue with balance is the necro ultimate. That will be nerfed since with time the only dd with place in a trial will be stamcros and 1 nb (for the passive if the tank isnt nb), the only question is when. I supose in 6 months.

SoLooney

Runefang wrote: »

Jhalin wrote: »

I’m just sad NBs isn’t even top 3 in their only viable role

Stamblades still top the damage to boss numbers, they just don’t have the aoe damage of a Stamcro. Magblades well.... no good news there.

Stamblades do not out dps stam necros because they dont have access to major vulnerability, and group dps will suffer if you dont have a few stam necro dps

BlueRaven

Question: Should ranged and melee do the same damage? (I don’t have an answer for this.)

Chadak

Runefang wrote: »

No but why did you pick 99th percentile over the 95th then? Because it told the story you wanted to tell.

Oh man are you gonna die of fifty aneurysms a day if you ever take college-level statistics.