DPS Comparison of RACES on MagPlar on PTS4.3.0 (Tests, Graphs, and Analysis)

HatchetHaro · January 2019

Sup, I've spent so much time on the PTS just parsing and doing the exact same rotation that it has made me numb. Let's get on with it.
To make things simpler, I've also copied the posting style @susmitds used for their own tests, because I can't be arsed.
I do have another thread in the PTS forums that documents my testing procedure in more detail, but here I'd like to analyze and provide commentary on what these PTS changes mean and perhaps what can be done to improve upon them.

Aim

To provide an accurate set of data concerning damage difference across races, using Magicka Templars as an example. These numbers will also be compared against data collected from the Murkmire PTS (equivalent to live).

Procedure

Consistency is key. Since these are parses that are focused on testing for the difference in DPS across different races, it is important to minimize as many variables as possible. This means that the goal of my parses isn't trying for the highest number each time; the goal is to replicate each parse perfectly across races, including identical rotations, identical builds, and identical builds. That being said, there are a few caveats involving certain builds that are just not representative of the effectiveness of a certain race, so certain variables are unavaidable; those variables are kept at a minimum.

I have selected six races to be compared: Khajiit, Altmer, Dunmer, Breton, Argonian, and Nord. These are typical Magicka races aside from the Nord, which is representative of a race with no racial passives in the context of Magicka DPS.

These are all Magicka Templar parses on the 6mil target skeleton, with Worm, Elemental Drain, and Orbs provided by an assistant. Each parse is replicated with the exact same rotations, and any parse with a single missed DoT ability, two missed spammable abilities, two missed light attacks, or with exceptionally lucky or unlucky crits (IE over 5% average difference from base crit for the top 5 damage sources in each CMX parse) are discarded. Trust me, I didn't just do this for Wrathstone; I did the exact same for Murkmire, three times.

Builds

https://imgur.com/a/USbuj28

Mundus:
The Thief (Critical chance)

CP:

Thaumaturge - 66
Master-at-Arms - 66
Staff Expert - 9
Elemental Expert - 49
Elfborn - 56
Spell Erosion - 24

Arcanist - 100

Gear:

Head - Zaan - Heavy - Infused - Magicka
Shoulder - Zaan - Medium - Divines - Magicka
Chest - Perfected Siroria - Light - Infused - Magicka
Hands - Perfected Siroria - Light - Divines - Magicka
Waist - Perfected Siroria - Light - Divines - Magicka
Legs - Perfected Siroria - Light - Infused - Magicka
Feet - Perfected Siroria - Light - Divines - Magicka
Necklace - Spell Strategist - Infused - Spell Damage
Ring - Spell Strategist - Bloodthirsty - Spell Damage
Ring - Spell Strategist - Bloodthirsty - Spell Damage
Staff - Spell Strategist - Inferno - Infused - Fiery Weapon
Staff - Maelstrom - Inferno - Infused - Absorb Magicka

Food:
Khajiit, Altmer, Dunmer, Argonian, Nord - Clockwork Citrus Filet (Magicka + Health + Magicka Recovery)
Breton - Mistral Banana-Bunny Hash (Magicka + Health)

Vampirism:
None or Stage 1
Because I completely forgot and by the time I realized I was already 2/3 done with the parsing.
No difference in sustain or damage between Stage 1 Vampirism and no Vampirism.
Does not affect data or damage difference in any way.

Rotation:

1. Purifying Light > Blockade of Fire > Vampire's Bane > Blazing Spear > 4x Elemental Weapon
2. Purifying Light > Blockade of Fire > Elemental Weapon > Blazing Spear > Vampire's Bane > 3x Elemental Weapon
3. Purifying Light > Blockade of Fire > Elemental Weapon > Blazing Spear > 3x Elemental Weapon > Vampire's Bane
4. Purifying Light > Blockade of Fire > Elemental Weapon > Blazing Spear > 2x Elemental Weapon > Channeled Acceleration

Refresh Channeled Focus on cooldown, replacing one Elemental Weapon cast.
No heavy attacks are cast, and no Spell Symmetry is used.

Radiant Oppression is slotted, but is not used. Executes are unpredictable and can make results vary a lot between parses. Once more, consistency is key.

Results

I like myself some charts, so here are some charts.

Khajiit - 49315.6 - By far the best race for Magicka DPS. Their raw crit by itself already gives them 4.52% more damage (assuming Minor Force and 56 points into Elfborn), and with Aggressive Horn it goes up to 5.27%. Add to that 750 extra Magicka and you have the recipe for quite the powerhouse in Magicka damage-dealing. Compared to no racials, they deal 6.56% more damage.

Altmer - 48700.2 - Altmer has really respectable DPS, and is very powerful with their combination of +2000 max Magicka and 258 Spell Damage. Their sustain, however, really isn't much, with only about 65-75 Magicka restored per second, which is on the same level of Argonians at this point. Compared to no racials, they deal 5.23% more damage.

Dunmer - 48367.4 - They're only 750 Magicka behind Altmer, but their 258 Spell Damage is still extremely powerful. Couple that with the exact same bonuses in Stamina form makes them the ultimate hybrid race, viable and powerful in both Magicka and Stamina damage. Compared to no racials, they deal 4.51% more damage.

Breton - 48309 - Right here, in this test, compared to no racials, they deal 4.38% more damage. However, let's address something here: this number is not representative of the Breton's power. This test was designed to be done without any heavy attacks or Spell Symmetry casts, and while other races mostly barely able to fully sustain a 6mil parse in gold food, Bretons are able to sustain the 6mil parse easily in bi-stat food, so close to being able to drop the Absorb Magicka glyph for a Berserker glyph as well. If the test was redesigned with sustain as a bigger focus, such as allowing heavy attacks / Spell Symmetry, Bretons will shine so much more. Right now, they have the most utility out of all the races simply due to their insane Magicka sustain, and from that, they have much more flexibility with their builds and rotations. In a more realistic scenario, I have no doubts they would match or even surpass Altmer.

Argonian - 47011.4 - Argonians have always been bad at damage-dealing; they drew power from their utility and dominated through their ability in tanking, healing, and PvP. However, with the huge incoming nerfs to their utility, what is left of it will not give Argonians any visible DPS or even sustain advantage over other races. The +1000 max Magicka only gives them a small boost in Magicka damage, only better than Stamina races, but at this point they are at the bottom for Magicka DPS as a Magicka race, and Stamina DPS in general. Compared to no racials, they deal 1.58% more damage.

Nords - 46277.8 - Same as Redguards, Bosmer, Orcs, and Imperials in terms of Magicka damage, they represent the bottom of the Magicka DPS list. At least most of them have Stamina bonuses that make them great for Stamina DPS.

More data

Right here, I have the full spreadsheet I've created for all the necessary calculations involving this particular parsing session.

And here is the same test done during PTS 4.2.4 Murkmire (well, mostly the same; it was in all-Divines gear and with no Minor Force)

Conclusion

An issue with these new racial changes in its current form is that they are actually increasing the DPS gap between races. The main issue here is that the Khajiiti has been over-buffed in terms of Magicka damage. By taking their already-powerful Weapon Crit (surpassed only by Redguards and their sustain) and applying it to Spell Crit, what little balance this game has just gets tipped far too much.

Let's take Khajiiti out of the equation for now, and we find that the damage difference is much more reasonable with Altmer at the top, with 5.23% more damage than the worst race. However, compared to Murkmire's 5.68, the gap really isn't lowered by that much. This is still a significant chunk of damage, big enough for Altmer (or whatever is at the top) to be the absolutely dominant choice for players looking to play Magicka DPS. I believe that the sweet spot for that damage difference between the best and worst race at is somewhere around 3%, where races keep their identity but people aren't as forced to play something they don't like.

Let's also take a look at Bretons in their current form right now. While I am perfectly fine with Bretons having superior Magicka sustain as a whole, being able to sustain on bi-stat food better than Argonians can on gold food is honestly insane. To put it into perspective, it's even more than the Stamina sustain on Redguards.

Lastly, let's address the issue that I have a personal interest in: Argonians. They suck right now. They used to be dominant in terms of utility but now that that's nerfed, they have nothing going for them at all. I'm personally fine with the nerfs, honestly; their utility was overpowered and made them far too dominant for tanking, healing, and PvP. However, with the nerfs to their utility that brings that dominance in those roles down, there aren't any buffs to their damage-dealing potential at all, and it leaves them in a really bad place right now. They don't even have their trademark sustain to help with DPS like Bretons and Redguards do. All in all, they really need something to bring their DPS effectiveness up on par with other races.

Seraphayel · January 2019

Very nice, thanks for putting so much time into this!

My question is if we even should put Nord / Khajiit in the now / then calculations because they are / were not meant to be Magicka DPS.

If we remove Nord from your first example (Wrathstone) and Khajiit from your second (Murkmire) the difference between the "Magicka DPS" races is 1k DPS (Khajiit -> Breton) which amounts to acceptable ~2% difference for Magicka Templars (for other classes it's even less and the best to worst racial distribution is different). I don't know why Breton is missing in your Murkmire example but I'd say the then Magicka DPS races were almost the same and there didn't change much.

One point that's obvious is that Argonians are not meant to be a pure Magicka DPS race at all. To make them compete with Altmer, Breton, Dunmer or Khajiit in Wrathstone for the Magicka DPS spot is in my opinion not ZOS goal. They're better tailored towards the tank/healer role overall.

zaria · January 2019

Seraphayel wrote: »

Very nice, thanks for putting so much time into this!

My question is if we even should put Nord / Khajiit in the now / then calculations because they are / were not meant to be Magicka DPS.

If we remove Nord from your first example (Wrathstone) and Khajiit from your second (Murkmire) the difference between the "Magicka DPS" races is 1k DPS (Khajiit -> Breton) which amounts to acceptable ~2% difference for Magicka Templars (for other classes it's even less and the best to worst racial distribution is different). I don't know why Breton is missing in your Murkmire example but I'd say the then Magicka DPS races were almost the same and there didn't change much.

One point that's obvious is that Argonians are not meant to be a pure Magicka DPS race at all. To make them compete with Altmer, Breton, Dunmer or Khajiit in Wrathstone for the Magicka DPS spot is in my opinion not ZOS goal. They're better tailored towards the tank/healer role overall.

Having an race with no racial skills help giving an baseline, its also nice for pointing out impact of race.

Difference between magic races is 2%
The huge 25 parse test got 1.3-1.6% difference but did not test templar but DK and NB in both magic and stamina
Also Templar is a bit special since it give an bonus on crit making Khajiit very strong in that class.

shaielzafine · January 2019

Interesting finds. You confirmed my suspicions about sustain on magplar with diff races. I do agree with you about argonians. They really have nothing going for them after the nerfs, since their sustain is not as good as breton for magicka healing and not as good for tanking either compared to Nords or even Imperials (imperial is bugged on pts with the red diamond passive). It leaves them underperforming on pts not just for dps but for other roles, even though on live they do have too many advantages to use especially for PvP.

HatchetHaro · January 2019

Seraphayel wrote: »

Very nice, thanks for putting so much time into this!

My question is if we even should put Nord / Khajiit in the now / then calculations because they are / were not meant to be Magicka DPS.

If we remove Nord from your first example (Wrathstone) and Khajiit from your second (Murkmire) the difference between the "Magicka DPS" races is 1k DPS (Khajiit -> Breton) which amounts to acceptable ~2% difference for Magicka Templars (for other classes it's even less and the best to worst racial distribution is different). I don't know why Breton is missing in your Murkmire example but I'd say the then Magicka DPS races were almost the same and there didn't change much.

One point that's obvious is that Argonians are not meant to be a pure Magicka DPS race at all. To make them compete with Altmer, Breton, Dunmer or Khajiit in Wrathstone for the Magicka DPS spot is in my opinion not ZOS goal. They're better tailored towards the tank/healer role overall.

More numbers is always better. In this case, I am comparing the effectiveness of their racial passives, so it's good to have a baseline race for comparison.

Breton is missing from my Murkmire example because I never got to it. Some data is missing because testing all of these take a ton of time.

The issue with tanking and healing right now is that they don't have an effectiveness ceiling compared to DPS. What this means is that all races can tank to the same effect given enough skill in the player doing the tanking; some of the best tanks I've seen are Imperial and Bosmer, and Woeler runs a Khajiit tank and is widely regarded on the forums as one of the best tanks around. One of my raid healers runs an Orc MagSorc healer and she is really good.

DPS is a completely different story, because they have an actual measurable ceiling to their effectiveness. Their stats affect their numbers directly, and given even unlimited skill, an Argonian will always pale in comparison to other races. It is this disparity in effectiveness ceilings in utility roles compared to DPS that makes DPS testing so much more important. Basically, you can be a Redguard tank and do really well in end-game scorepushing, but bring an Argonian DPS and people will complain about it, no matter your actual skill.

susmitds · January 2019

Nice comparison. Pretty much matches with my predictions.

susmitds wrote: »

Thorstienn wrote: »

Great job.
Was wondering what made Breton not excellent for Templar? I mean sustain is always good, and would think they would have the largest Mag pool for execute phase, this a larger jesus beam.

Simply, the fact that Khajiit got better parses on average. I didn't explicitly test this as I am not used to Magicka Templar DPS, not having one on live, but theoretical analysis, as well as tests from other players, some guild mates and friends tended to go this way.

Also I personally tested Stamina Templar for Khajiit, Dunmer and Orc, though not as extensively as these ones, and Khajiit averaged almost 700 DPS above the 2nd Dunmer, with their highest parse being as high as 1.4k higher than Dunmer. Obviously I had a smaller sample of 5 and the error rate is high as such. Still, the overall results are matching with theoretical analysis.

The results will even out somewhat with higher sample sizes of around thirty. The results are fine but there is one small flaw with the testing method.

HatchetHaro wrote: »

Each parse is replicated with the exact same rotations, and any parse with a single missed DoT ability, two missed spammable abilities, two missed light attacks, or with exceptionally lucky or unlucky crits (IE over 5% average difference from base crit for the top 5 damage sources in each CMX parse) are discarded.

Consistency varies wildly between the races. By discarding edge cases, you are forcefully decreasing the parse variance of the race. This would have been a non-factor, if all races had the same variance. However, It is wildly different, with Khajiit having almost three times the variance (square of standard deviation) of the race with the second highest variance (dunmer for stam, Altmer for magicka). This introduces nonuniform bias in the data.

An easier way to understand this. Say, if an Altmer gets unlucky with crit in the execution phases, it has 2000 magicka and 200 spell damage to statically increase raw damage to makeup for it. However, say a Khajiit get unlucky, it has just 750 to statically increase raw damage, which is just barely above no racial bonus.

casparian · January 2019

How I suspect ZOS sees it:

- Khajiit is the clear winner for highly skilled players (in well coordinated groups) who can manage its low sustain.
- Altmer and Breton are very close, even for less experienced players.
- Everyone else play stam or support.

I’m not sure that’s so bad. Not everyone will be able to pull ahead of Altmer and Breton with a cat, and for those that can, top tier performance is the reward.

zaria · January 2019

casparian wrote: »

How I suspect ZOS sees it:

- Khajiit is the clear winner for highly skilled players (in well coordinated groups) who can manage its low sustain.
- Altmer and Breton are very close, even for less experienced players.
- Everyone else play stam or support.

I’m not sure that’s so bad. Not everyone will be able to pull ahead of Altmer and Breton with a cat, and for those that can, top tier performance is the reward.

Khajiit is very strong on templars because Piercing Spear give an +10% damage on critical strikes.

They are in the middle of the row for other magical or stamia classes tested.
https://forums.elderscrollsonline.com/en/discussion/455092/raid-buffed-dps-comparison-of-each-race-by-class-tests-graphs-interpretation-and-final-score/p1

This effect should already be visible on stamplar today.
If op change Piercing Spear to something else or nerf it.

casparian · January 2019

zaria wrote: »

casparian wrote: »

How I suspect ZOS sees it:

- Khajiit is the clear winner for highly skilled players (in well coordinated groups) who can manage its low sustain.
- Altmer and Breton are very close, even for less experienced players.
- Everyone else play stam or support.

I’m not sure that’s so bad. Not everyone will be able to pull ahead of Altmer and Breton with a cat, and for those that can, top tier performance is the reward.

Khajiit is very strong on templars because Piercing Spear give an +10% damage on critical strikes.

They are in the middle of the row for other magical or stamia classes tested.
https://forums.elderscrollsonline.com/en/discussion/455092/raid-buffed-dps-comparison-of-each-race-by-class-tests-graphs-interpretation-and-final-score/p1

This effect should already be visible on stamplar today.
If op change Piercing Spear to something else or nerf it.

Not just Templars. Nightblades have exactly the same passive under a different name.

susmitds · January 2019

casparian wrote: »

zaria wrote: »

casparian wrote: »

How I suspect ZOS sees it:

- Khajiit is the clear winner for highly skilled players (in well coordinated groups) who can manage its low sustain.
- Altmer and Breton are very close, even for less experienced players.
- Everyone else play stam or support.

I’m not sure that’s so bad. Not everyone will be able to pull ahead of Altmer and Breton with a cat, and for those that can, top tier performance is the reward.

Khajiit is very strong on templars because Piercing Spear give an +10% damage on critical strikes.

They are in the middle of the row for other magical or stamia classes tested.
https://forums.elderscrollsonline.com/en/discussion/455092/raid-buffed-dps-comparison-of-each-race-by-class-tests-graphs-interpretation-and-final-score/p1

This effect should already be visible on stamplar today.
If op change Piercing Spear to something else or nerf it.

Not just Templars. Nightblades have exactly the same passive under a different name.

In case of Nightblades, NBs have ways to increase crit chance with passive, so it tends to even out. Most NBs tend to rock high crit chances anyways. Dunmer and Breton actually ended up with better averages for me for stam and mag.

For Templars , Templars have the critical damage buff but no way to boost crit chance from within their class. This is where, the Khajiit passive comes handy.

Tabbycat · January 2019

From a lore perspective, Khajiit should never be the best magicka users, ever. Altmer should. Every TES fan knows this.

As a TES fan, I'm disappointed and sad that ZOS is ignoring racial lore so much. I can only hope they redeem themselves with better lore friendly balancing before this goes live.

mairwen85 · January 2019

Assumption based on sustain metrics: For dedicated healer post U21, Breton = best choice?

BTW. thank you for this effort. I'm loving how the community is pulling this together. What we need now is a single thread that collates all of these into one place for quick reference.

John_Falstaff · January 2019

Tabbycat wrote: »

From a lore perspective, Khajiit should never be the best magicka users, ever. Altmer should. Every TES fan knows this.

As a TES fan, I'm disappointed and sad that ZOS is ignoring racial lore so much. I can only hope they redeem themselves with better lore friendly balancing before this goes live.

Not every TES fan knows that Alfiq, Dagi and Dagi-raht are spellcasters, apparently. And Alfiq will be introduced in the upcoming Elsweyr chapter as NPCs. So yes, khajiits are pretty much magic users too.

nsmurfer · January 2019

Tabbycat wrote: »

From a lore perspective, Khajiit should never be the best magicka users, ever. Altmer should. Every TES fan knows this.

As a TES fan, I'm disappointed and sad that ZOS is ignoring racial lore so much. I can only hope they redeem themselves with better lore friendly balancing before this goes live.

We don't have as much Khajiit lore as other races. How many Khajiit named heroes do we even know of? We mostly knwo about their birthcycle, religious system and general way of life. As for info on specific, we don't have them. I bet that is going to change. ALso BTW, J'zargo, a Khajiit was the strongest mage follower in the game(literally stat and level wise) he went above Lv 200 which is more than any other follower.

HatchetHaro · January 2019

mairwen85 wrote: »

Assumption based on sustain metrics: For dedicated healer post U21, Breton = best choice?

BTW. thank you for this effort. I'm loving how the community is pulling this together. What we need now is a single thread that collates all of these into one place for quick reference.

It's actually kind-of funny; Bretons are actually on par with Argonians right now on live when it comes to racial choice for healers, precisely because Breton Magicka sustain still outweighs Argonian healing. With the incoming buff to Breton sustain (and nerf to Argonian sustain and healing), however, Bretons will become by far the best choice for healers with the race changes; their sheer amount of sustain puts them far beyond other races.

susmitds wrote: »

Nice comparison. Pretty much matches with my predictions.

susmitds wrote: »

The results will even out somewhat with higher sample sizes of around thirty. The results are fine but there is one small flaw with the testing method.

HatchetHaro wrote: »

Each parse is replicated with the exact same rotations, and any parse with a single missed DoT ability, two missed spammable abilities, two missed light attacks, or with exceptionally lucky or unlucky crits (IE over 5% average difference from base crit for the top 5 damage sources in each CMX parse) are discarded.

Consistency varies wildly between the races. By discarding edge cases, you are forcefully decreasing the parse variance of the race. This would have been a non-factor, if all races had the same variance. However, It is wildly different, with Khajiit having almost three times the variance (square of standard deviation) of the race with the second highest variance (dunmer for stam, Altmer for magicka). This introduces nonuniform bias in the data.

An easier way to understand this. Say, if an Altmer gets unlucky with crit in the execution phases, it has 2000 magicka and 200 spell damage to statically increase raw damage to makeup for it. However, say a Khajiit get unlucky, it has just 750 to statically increase raw damage, which is just barely above no racial bonus.

The thing here is that I am testing for average dps, not the variance. Besides, even if you factor in variance, with enough tests, they all average down to, well, the average dps. You raised a scenario where if an Altmer and a Khajiit get unlucky with their crit, the Khajiit would have their damage dropped more than the Altmer does, and that is true; however, you didn't mention the opposite of that scenario, where an Altmer and a Khajiit get lucky with their crit.

Just to prove the point here, I went back onto the PTS and got the damage numbers for Light Attacks on both Khajiit and Altmer, taking into account the relevant buffs to make this more accurate, such as fully stacking Siroria's Boon, proccing Spell Strategist, proccing Templar's Minor Sorcery by casting Vampire's Bane, Channeled Acceleration, Elemental Drain, and Spell Power potions.

Since Light Attacks scale the exact same way as most abilities, we can use this as an approximation of DPS. Keep in mind that this is just an approximation; I did not factor in frontbar:backbar times, abilities that don't scale with Spell Power (EG Purifying Light), and other raid buffs.

Khajiit non-crit Light Attacks are 5152 damage, and Altmer are 5460. Factoring in 51.8% base crit and 190% crit damage (56 Elfborn, Templar 10% Crit Damage, Minor Force), the average damage for Light Attacks would be 7925 for Khajiits and 8006 for Altmer.

Lets say, then, you are parsing and you have a permanent, idk, Minor Misfortune buff on you that lowers your Critical Chance by 10% (that is, 51.8% x 90%, not (51.8 - 10)%, although it doesn't really matter).

Based on that, Khajiiti will be left with 53.82% Crit Chance, and Altmer would be left with 46.62% Crit Chance. From that, the Khajiit will lose about 3.5% of their damage, and the Altmer would lose about 3.18% of their damage, so yes, you are correct in that Khajiiti would have a higher variance in their damage scenarios.

However, if you have a permanent Minor Fortune buff that increases your Critical Chance by 10% instead (51.8% x 110%) instead, Khajiiti and Altmer would have 65.78% and 56.98% Critical Chance respectively, and that calculates to, well, 3.5% more damage for Khajiit and 3.18% more damage for Altmer.

Let's pretend, instead, that the Minor Misfortune and Minor Fortune buffs are additive instead of multiplicative, where the 10% would just directly be subtracted or added to your base crit. An unlucky Khajiit and Altmer would lose about 5.85% and 6.13% damage respectively, and a lucky Khajiit and Altmer would gain about 5.85% and 6.13% damage respectively.

Basically, variance is just that, variance. It doesn't matter if the variance is high or low, the average will still remain the same.

zaria wrote: »

casparian wrote: »

How I suspect ZOS sees it:

- Khajiit is the clear winner for highly skilled players (in well coordinated groups) who can manage its low sustain.
- Altmer and Breton are very close, even for less experienced players.
- Everyone else play stam or support.

I’m not sure that’s so bad. Not everyone will be able to pull ahead of Altmer and Breton with a cat, and for those that can, top tier performance is the reward.

Khajiit is very strong on templars because Piercing Spear give an +10% damage on critical strikes.

They are in the middle of the row for other magical or stamia classes tested.
https://forums.elderscrollsonline.com/en/discussion/455092/raid-buffed-dps-comparison-of-each-race-by-class-tests-graphs-interpretation-and-final-score/p1

This effect should already be visible on stamplar today.
If op change Piercing Spear to something else or nerf it.

The change in racial difference between 180% Crit Damage and 190% Crit Damage is not that high.
I'll just use the +8% crit chance the only variable, so I'm not counting the +750 Max Magicka and Stamina Khajiiti have.
Once more, I'm just going to assume 51.8% base crit (and therefore 59.8% for Khajiit).

Just from this crit difference, on 180% Crit Damage (56 Elfborn + Minor Force), Khajiiti will deal 4.52% more damage compared to no crit passives.
On 190% Crit Damage (56 Elfborn + Minor Force + Templar 10%), Khajiiti will deal 4.91% more damage compared to no crit passives.
Lets put Major Force into the picture as well. At 195% Crit Damage, Khajiiti will deal 5.09% more damage; at 205% Crit Damage, Khajiiti will deal 5.44% more damage.

So, not a huge difference. If anything, if you want approximate numbers at 180% rather than 190%, it is easy to just take 10% crit damage out of all of my parsing numbers with some simple algebra. Given, of course, this does not take the different scaling of different classes into account, but at least you can see what the approximate difference is like.

Lets take a theoretical 50000 dps parse on Altmer that I want to convert from 190% Crit Damage to 180% Crit Damage.

DPS = Base DPS * Crit Chance * Crit Damage + Base DPS * (1 - Crit Chance)
50000 = B * 0.518 * 1.9 + B * 0.482
50000 = B * 1.4662
B = 34101.76

Then we just take that number and calculate the 180% Crit Damage from that.
34101.76 * 0.518 * 1.8 + 34101.76 * 0.482
= 48233.53 DPS

Rinse and repeat for other races.

susmitds · January 2019

Your results and interpretations are on point.

I was just noting that from a statistical point of view, it is better to use random sampling to convert a larger sample size to a smaller one than manually removing the edge data.

Tabbycat · January 2019

John_Falstaff wrote: »

Tabbycat wrote: »

From a lore perspective, Khajiit should never be the best magicka users, ever. Altmer should. Every TES fan knows this.

As a TES fan, I'm disappointed and sad that ZOS is ignoring racial lore so much. I can only hope they redeem themselves with better lore friendly balancing before this goes live.

Not every TES fan knows that Alfiq, Dagi and Dagi-raht are spellcasters, apparently. And Alfiq will be introduced in the upcoming Elsweyr chapter as NPCs. So yes, khajiits are pretty much magic users too.

But those breeds are not featured prominantly as playable in the TES series.

For example, the Morrowind description of Khajiit, as follows from the UESP wiki:

Khajiit are cat-like people who come from Elsweyr, known for high intelligence and agility. These traits make them very good thieves and acrobats, but Khajiit are also fearsome warriors. However, they are rarely known to be mages. Khajiit mostly stay on land, but piracy and skooma trade does draw some to work as sailors.

https://en.uesp.net/wiki/Morrowind:Khajiit

HatchetHaro · January 2019

susmitds wrote: »

Your results and interpretations are on point.

I was just noting that from a statistical point of view, it is better to use random sampling to convert a larger sample size to a smaller one than manually removing the edge data.

That much is true. I'm not going to exclude the possibility that my parses, no matter how consistent, may also have been affected enough by crit randomness to show data that is skewed in comparison to real numbers.

There is still a reason why I opted not to use my execute and also opted to stick with a static rotation: it is to make sure I pull off the exact same rotations every time to reduce the variance I get from my parses. I can at least trust that to keep my data consistent and as accurate as possible without going insane from staring at the screen and doing the exact same thing over and over again.

To be honest, I really should just make a script to do that for me.

I'll see what 4.3.1 holds and will make another thread with new results.

John_Falstaff · January 2019

Tabbycat wrote: »

John_Falstaff wrote: »

Tabbycat wrote: »

From a lore perspective, Khajiit should never be the best magicka users, ever. Altmer should. Every TES fan knows this.

As a TES fan, I'm disappointed and sad that ZOS is ignoring racial lore so much. I can only hope they redeem themselves with better lore friendly balancing before this goes live.

Not every TES fan knows that Alfiq, Dagi and Dagi-raht are spellcasters, apparently. And Alfiq will be introduced in the upcoming Elsweyr chapter as NPCs. So yes, khajiits are pretty much magic users too.

But those breeds are not featured prominantly as playable in the TES series.

For example, the Morrowind description of Khajiit, as follows from the UESP wiki:

Khajiit are cat-like people who come from Elsweyr, known for high intelligence and agility. These traits make them very good thieves and acrobats, but Khajiit are also fearsome warriors. However, they are rarely known to be mages. Khajiit mostly stay on land, but piracy and skooma trade does draw some to work as sailors.

https://en.uesp.net/wiki/Morrowind:Khajiit

ZOS, in their justification - right from the horse's mouth for you - specifically stated that new racials try to encompass the whole racial diversity of khajiits. I don't know what else is to be said. That's pretty much self-explanatory.

HatchetHaro · January 2019

John_Falstaff wrote: »

Tabbycat wrote: »

John_Falstaff wrote: »

Tabbycat wrote: »

From a lore perspective, Khajiit should never be the best magicka users, ever. Altmer should. Every TES fan knows this.

As a TES fan, I'm disappointed and sad that ZOS is ignoring racial lore so much. I can only hope they redeem themselves with better lore friendly balancing before this goes live.

Not every TES fan knows that Alfiq, Dagi and Dagi-raht are spellcasters, apparently. And Alfiq will be introduced in the upcoming Elsweyr chapter as NPCs. So yes, khajiits are pretty much magic users too.

But those breeds are not featured prominantly as playable in the TES series.

For example, the Morrowind description of Khajiit, as follows from the UESP wiki:

Khajiit are cat-like people who come from Elsweyr, known for high intelligence and agility. These traits make them very good thieves and acrobats, but Khajiit are also fearsome warriors. However, they are rarely known to be mages. Khajiit mostly stay on land, but piracy and skooma trade does draw some to work as sailors.

https://en.uesp.net/wiki/Morrowind:Khajiit

ZOS, in their justification - right from the horse's mouth for you - specifically stated that new racials try to encompass the whole racial diversity of khajiits. I don't know what else is to be said. That's pretty much self-explanatory.

Let's just put it this way. ZOS can really do anything they want with the racial passives, and they can still make it fit the lore just by changing the name of the Passive itself. For example, they added Weapon Damage to Dunmer and changed the name of that passive from Destructive Ancestry to Ruination, and they changed... everything for Khajiit.

I can just as easily come up with something new.

Nord:
Rugged → Atmoran Strength: Increases your Max Stamina by 4321

Argonian:
Life Mender → Wrath of the Hist: Increases your Weapon and Spell Damage by 50,000

Bosmer:
Y’ffre’s Endurance → Spinner's Gift: Increases your Max Magicka by 2

Redguard:
Adrenaline Rush → Yokudan Boon: Increases your Stamina and Magicka Recovery by 4.68

Okay the last one is a bit of a stretch, but you understand what I mean about lore and game balance in ESO.

Kiralyn2000 · January 2019

Perhaps I just don't have the proper viewpoint for the competitive/trials/etc scene, but that chart looks to me like proof that race is a pretty minor thing, and there's no reason for people to get so worked up about it. They're all very capable if even Nord can pull 46k+, and the difference between top & bottom is only 6.5%.

(and it also occurs to me that anyone who was into meta/min-max was already likely using one of the good-for-Mag races, and the top four on your chart are much closer to each other than 6.5%. So even less reason to get worked up.)

HatchetHaro · January 2019

Kiralyn2000 wrote: »

Perhaps I just don't have the proper viewpoint for the competitive/trials/etc scene, but that chart looks to me like proof that race is a pretty minor thing, and there's no reason for people to get so worked up about it. They're all very capable if even Nord can pull 46k+, and the difference between top & bottom is only 6.5%.

(and it also occurs to me that anyone who was into meta/min-max was already likely using one of the good-for-Mag races, and the top four on your chart are much closer to each other than 6.5%. So even less reason to get worked up.)

Yeah you don't have the view from the competitive raiding scene.

Let me just put it this way: people at this level care a lot about any damage difference, no matter how small, enough to exclude someone from raid teams simply because they are running a Khajiit StamPlar rather than a Redguard. It's why you see people devastated about Dunmer no longer being BiS for Magicka DPS.

Fact of the matter is that a difference of 5% is already a lot, and in the case of a full trial fight can mean the difference between being able to burn and having to do mechanics properly. It all adds up in the entirety of a trial.

Even if, say, your group is really good, it can be a difference in getting the top score on the leaderboards and getting 5th place, and trust me, when it gets up to that point, it gets even more competitive and way more toxic. If you are playing a race you like that is not BiS, the blame would be shifted onto you and you'd be booted from those teams, no matter your skill level.

Basically, the end-game raiding scene can be very elitist, all because of differences way less than 5%.

Hallothiel · January 2019

nsmurfer wrote: »

Tabbycat wrote: »

From a lore perspective, Khajiit should never be the best magicka users, ever. Altmer should. Every TES fan knows this.

As a TES fan, I'm disappointed and sad that ZOS is ignoring racial lore so much. I can only hope they redeem themselves with better lore friendly balancing before this goes live.

We don't have as much Khajiit lore as other races. How many Khajiit named heroes do we even know of? We mostly knwo about their birthcycle, religious system and general way of life. As for info on specific, we don't have them. I bet that is going to change. ALso BTW, J'zargo, a Khajiit was the strongest mage follower in the game(literally stat and level wise) he went above Lv 200 which is more than any other follower.

Ah you mention J’zargo, my favourite companion & kick ass warrior mage. Who says Khajiit can’t be good at magic?!

Godspeed · January 2019

Tabbycat wrote: »

From a lore perspective, Khajiit should never be the best magicka users, ever. Altmer should. Every TES fan knows this.

As a TES fan, I'm disappointed and sad that ZOS is ignoring racial lore so much. I can only hope they redeem themselves with better lore friendly balancing before this goes live.

There is still time for changes to be made. I can’t see the cats being the best at magic making it to live.

zaria · January 2019

Tabbycat wrote: »

From a lore perspective, Khajiit should never be the best magicka users, ever. Altmer should. Every TES fan knows this.

As a TES fan, I'm disappointed and sad that ZOS is ignoring racial lore so much. I can only hope they redeem themselves with better lore friendly balancing before this goes live.

Its one build, nerf it, would not play other than healer on templar anyway.
Read other parses who put more work into it.

Or rather we keep Dunmer racials, then nerf DK down to warden level its an tanking class, then nerf elemental damage not reason to complain.
Or go full Stalin, Skyrim level racials is an option, that would balance things

Yes I expected that.
And its still on the table, do not rock the boat like an Leeroy Jenkins.

Yes we could all go MAD here demanding nerfs on each other races Mexican standoff style . Overland might become hard again

nsmurfer · January 2019

Godspeed wrote: »

Tabbycat wrote: »

From a lore perspective, Khajiit should never be the best magicka users, ever. Altmer should. Every TES fan knows this.

As a TES fan, I'm disappointed and sad that ZOS is ignoring racial lore so much. I can only hope they redeem themselves with better lore friendly balancing before this goes live.

There is still time for changes to be made. I can’t see the cats being the best at magic making it to live.

Khajiit gets higher parses in perfect conditions in one magicka class by 1% max difference, everyone loses their head. Dunmer gets highest parses in every stamina classes in the same condition by same difference, no one bats an eye.

zaria · January 2019

nsmurfer wrote: »

Godspeed wrote: »

Tabbycat wrote: »

From a lore perspective, Khajiit should never be the best magicka users, ever. Altmer should. Every TES fan knows this.

As a TES fan, I'm disappointed and sad that ZOS is ignoring racial lore so much. I can only hope they redeem themselves with better lore friendly balancing before this goes live.

There is still time for changes to be made. I can’t see the cats being the best at magic making it to live.

Khajiit gets higher parses in perfect conditions in one magicka class by 1% max difference, everyone loses their head. Dunmer gets highest parses in every stamina classes in the same condition by same difference, no one bats an eye.

Recommend 3 changes here, first change Piercing Spear to sustain or something, don't want to be bis, the sea is full of battleships, the sky of bombers and winter is coming, dive below bis level and cruise
Second keep flame damage and remove spell damage on Dunmer. Make DK more of an tanking class, nerf, elemental effects.
Give us an evil ending on Morrowind main quest there we don't stop the asteroid.

Compare Dunmer with other responses, all other are asking for small buffs.
As that is that they need and know its something they can get.
Dunmer, winter is coming lets invade Soviet demanding the moon.
Seriously start to dislike the xenophobic slavers now, before it was just RP.

Note that this does not relate to all the thousands of Dunmer stamina players who created this. They deserve just as much honor as the Khajiit magic players or other non meta build who did not made it.

Seraphayel · January 2019

HatchetHaro wrote: »

Kiralyn2000 wrote: »

Perhaps I just don't have the proper viewpoint for the competitive/trials/etc scene, but that chart looks to me like proof that race is a pretty minor thing, and there's no reason for people to get so worked up about it. They're all very capable if even Nord can pull 46k+, and the difference between top & bottom is only 6.5%.

(and it also occurs to me that anyone who was into meta/min-max was already likely using one of the good-for-Mag races, and the top four on your chart are much closer to each other than 6.5%. So even less reason to get worked up.)

Yeah you don't have the view from the competitive raiding scene.

Let me just put it this way: people at this level care a lot about any damage difference, no matter how small, enough to exclude someone from raid teams simply because they are running a Khajiit StamPlar rather than a Redguard. It's why you see people devastated about Dunmer no longer being BiS for Magicka DPS.

Fact of the matter is that a difference of 5% is already a lot, and in the case of a full trial fight can mean the difference between being able to burn and having to do mechanics properly. It all adds up in the entirety of a trial.

Even if, say, your group is really good, it can be a difference in getting the top score on the leaderboards and getting 5th place, and trust me, when it gets up to that point, it gets even more competitive and way more toxic. If you are playing a race you like that is not BiS, the blame would be shifted onto you and you'd be booted from those teams, no matter your skill level.

Basically, the end-game raiding scene can be very elitist, all because of differences way less than 5%.

Where is the difference 5%? People are already playing the "best" race for their class/build. After the upcoming changes the "best" races might alter a little bit but that's down to 1%, if you're unlucky 2% of a difference. It doesn't matter at all. We're not talking about 5% differences here, a Khajiit is not 5% better than an Altmer for Magicka builds and a Dunmer is not 5% better than a Redguard for Stamina builds. Trial groups that are so anxious to lose X times <1% because of racials do have other problems than racial changes. Most balancing efforts due to classes and weapons cause higher DPS losses than those for the new racials and people adapt. They should adapt here as well, where the differences range from minimal to nonexistent.

zaria · January 2019

Godspeed wrote: »

Tabbycat wrote: »

From a lore perspective, Khajiit should never be the best magicka users, ever. Altmer should. Every TES fan knows this.

As a TES fan, I'm disappointed and sad that ZOS is ignoring racial lore so much. I can only hope they redeem themselves with better lore friendly balancing before this goes live.

There is still time for changes to be made. I can’t see the cats being the best at magic making it to live.

No they could just drop all racial passives and give us Skyrim ones as in skill stating bonuses.

HatchetHaro wrote: »

Kiralyn2000 wrote: »

Perhaps I just don't have the proper viewpoint for the competitive/trials/etc scene, but that chart looks to me like proof that race is a pretty minor thing, and there's no reason for people to get so worked up about it. They're all very capable if even Nord can pull 46k+, and the difference between top & bottom is only 6.5%.

(and it also occurs to me that anyone who was into meta/min-max was already likely using one of the good-for-Mag races, and the top four on your chart are much closer to each other than 6.5%. So even less reason to get worked up.)

Yeah you don't have the view from the competitive raiding scene.

Let me just put it this way: people at this level care a lot about any damage difference, no matter how small, enough to exclude someone from raid teams simply because they are running a Khajiit StamPlar rather than a Redguard. It's why you see people devastated about Dunmer no longer being BiS for Magicka DPS.

Fact of the matter is that a difference of 5% is already a lot, and in the case of a full trial fight can mean the difference between being able to burn and having to do mechanics properly. It all adds up in the entirety of a trial.

Even if, say, your group is really good, it can be a difference in getting the top score on the leaderboards and getting 5th place, and trust me, when it gets up to that point, it gets even more competitive and way more toxic. If you are playing a race you like that is not BiS, the blame would be shifted onto you and you'd be booted from those teams, no matter your skill level.

Basically, the end-game raiding scene can be very elitist, all because of differences way less than 5%.

This parse is an outlier, think Dunmer magDK in current meta so nerf templar.
On parses on the big test was all relevant races was within 800 damage on an 60K parse.

However and Bosmer magblade would outdps an bis magDK. So obviously no magDK== no min/max Dunmer.
I assume this is the real problem?
Well an one tricks pony live dangerous. And yes this is why I say templar should be changed.
Bis hurt you in the long run.

Next update stamina builds get an serious nerf if going life so much that an Dunmer stamblade might feel an nerf.
In short class and system changes is far more important, also player skills for some reasons the most prestigious tank is Khajiit, he like the race and is good.

ck37090 · January 2019

My question.. who the heck decided to make the Khajit a premiere magica race? They mostly kept existing themes with all other races (albiet they completely removed the dark elf from what WAS their main niche) ... But seriously... Why did they decide to make the Khajit the biggest damage dealer in Magicka and stamina? I just don't get where their heads are at

HatchetHaro · January 2019

Seraphayel wrote: »

HatchetHaro wrote: »

Kiralyn2000 wrote: »

Perhaps I just don't have the proper viewpoint for the competitive/trials/etc scene, but that chart looks to me like proof that race is a pretty minor thing, and there's no reason for people to get so worked up about it. They're all very capable if even Nord can pull 46k+, and the difference between top & bottom is only 6.5%.

(and it also occurs to me that anyone who was into meta/min-max was already likely using one of the good-for-Mag races, and the top four on your chart are much closer to each other than 6.5%. So even less reason to get worked up.)

Yeah you don't have the view from the competitive raiding scene.

Let me just put it this way: people at this level care a lot about any damage difference, no matter how small, enough to exclude someone from raid teams simply because they are running a Khajiit StamPlar rather than a Redguard. It's why you see people devastated about Dunmer no longer being BiS for Magicka DPS.

Fact of the matter is that a difference of 5% is already a lot, and in the case of a full trial fight can mean the difference between being able to burn and having to do mechanics properly. It all adds up in the entirety of a trial.

Even if, say, your group is really good, it can be a difference in getting the top score on the leaderboards and getting 5th place, and trust me, when it gets up to that point, it gets even more competitive and way more toxic. If you are playing a race you like that is not BiS, the blame would be shifted onto you and you'd be booted from those teams, no matter your skill level.

Basically, the end-game raiding scene can be very elitist, all because of differences way less than 5%.

Where is the difference 5%? People are already playing the "best" race for their class/build. After the upcoming changes the "best" races might alter a little bit but that's down to 1%, if you're unlucky 2% of a difference. It doesn't matter at all. We're not talking about 5% differences here, a Khajiit is not 5% better than an Altmer for Magicka builds and a Dunmer is not 5% better than a Redguard for Stamina builds. Trial groups that are so anxious to lose X times <1% because of racials do have other problems than racial changes. Most balancing efforts due to classes and weapons cause higher DPS losses than those for the new racials and people adapt. They should adapt here as well, where the differences range from minimal to nonexistent.

Yeah, no, the story is just different here, especially where I'm involved.

I run Argonian DDs, have written and updated two detailed DPS class guides up till Summerset, got several top scores over the year before Summerset that were only beaten by the top two guilds NA, got third IR NA during Summerset as well as a 48k vDSA score, and was still kicked out of a vCR+3 progression group simply for being an Argonian even after having previously achieved a 121k score in there. Right now I lead my own raid group for raid progression.

This isn't about balancing out the top races for DPS; this is about the bottom two right now that are just bad at both Stamina and Magicka and therefore stuck without an identity in the damage-dealing role.

The whole "just play a BiS race" thing doesn't work; this is an Elder Scrolls game, and some people will naturally forge an identity with their favorite race. And this isn't just me; plenty of raiders have been forced to race-change away from their favorite races just to stay relevant in the raiding scene.

I'm not denying that there will always be a BiS; I am saying that the gap still needs to be lower in terms of DPS. Racial identity can still be achieved by giving them unique flavor bonuses, something like the swim-speed bonus on Argonians.