Change Critical Meta

Syrpynt

I know I'm going to get a LOT of resitance to this, but here it goes:

I believe the game would be healthier and more diverse if critical chance was capped at 40%, BUT critical damage bonuses would replace some set slots that improve chance.

This is because critical chance is supposed to be an old school "skill" called "luck." It was mainly a high risk, yet high reward, feature.

My goal in this proposal is to make spell/weapon damage the relatively "constant"/reliable damage metric.

While Stamina and Magicka translates to your proficinecy in "healing done" as well as your total durability in a fight.

And lastly critical attacks/heals will do beyond ~2.5x damage when an attack/heal crits. 50% chance would be too much for a bonus multiplier of 2.5x damage. What I'm proposing basically makes "luck" (RNG) not a trait you can exploit and follow as the "only way to get the most damage" in ESO. The game is boring without build variety in end-game to have fun.

Anyway, expecting a lot of disagreement here, but I'm not really interested in any person's opinion that disagrees just because ALL of their dps builds are currently set up for the crit meta and don't want to lose their "kingly status" as a dps lord. That's not an argument, that's a whine.

WrathOfInnos

As predicted, I’m going to disagree. As you mentioned, doing good damage with low crit chance relies heavily on luck. I don’t think that success or failure should be determined by RNG, things need to be repeatable. Now it could be said that the current system also has plenty of luck built in (especially for Sorcs with frag procs), and if there were a way to smooth that out I would be all for it. IMO damage was much more consistent when everyone was running 80%+ crit chance and lower crit damage (Elsweyr through Greymoor IIRC), now with the ~60% crit meta and high crit damage the variance is much higher. I’m opposed to continuing that trend with another crit chance nerf or crit damage buff.

Interestingly, you can think about the absolute extremes, 0% crit chance or 100% crit chance. Both result in entirely repeatable DPS performance. I’d be in favor of either of these, as long as other stats were tuned down to ensure 100% crit builds are not overpowered (which has been a reality before, maybe even now), or tuned up so that 0% crit builds did approx the same damage they do right now.

Edit: I’ll also add that it’s not uncommon to reach 2.5X multiplier on crit this patch. Nightblades with Shadow can even achieve 2.78X, but IMO that’s only worth trying with Acuity, and typically they’d be better of using Thief for more crit chance.

Syrpynt

WrathOfInnos wrote: »

As predicted, I’m going to disagree. As you mentioned, doing good damage with low crit chance relies heavily on luck. I don’t think that success or failure should be determined by RNG, things need to be repeatable. Now it could be said that the current system also has plenty of luck built in (especially for Sorcs with frag procs), and if there were a way to smooth that out I would be all for it. IMO damage was much more consistent when everyone was running 80%+ crit chance and lower crit damage (Elsweyr through Greymoor IIRC), now with the ~60% crit meta and high crit damage the variance is much higher. I’m opposed to continuing that trend with another crit chance nerf or crit damage buff.

Interestingly, you can think about the absolute extremes, 0% crit chance or 100% crit chance. Both result in entirely repeatable DPS performance. I’d be in favor of either of these, as long as other stats were tuned down to ensure 100% crit builds are not overpowered (which has been a reality before, maybe even now), or tuned up so that 0% crit builds did approx the same damage they do right now.

Edit: I’ll also add that it’s not uncommon to reach 2.5X multiplier on crit this patch. Nightblades with Shadow can even achieve 2.78X, but IMO that’s only worth trying with Acuity, and typically they’d be better of using Thief for more crit chance.

Critical chance isn't about repeatability, at a fundamental level, it is literally RNG unless you are at 100% chance--people are thinking about raw damage when they talk about "consitency." So I don't see how your disagreement makes any sense.

What I'm looking for is what "amount of critical multiplier" would make it on par with "raw power" with a 40% cap so that other builds are viable. Is it 2.5x? Is it 2.0x? Is it 3.0x? And when I say critical multiplier, I mean unbuffed from other sources. This is base critical damage done.

Critical chance is literally: Your character got lucky and hit the JACKPOT!

If you are allowed to always hit jackpot, then of course it's going to be the lost powerful--but my point is there's no real "gamble" after a certain chance (currently). That's what isn't healthy about it.

RNG = High risk, high reward.



If people are opposed to a cap, I propose another method:

Every 5% over a 40% critical chance, has a critical failure chance of 5%, where you do 1/2x base damage. But I still would say that critical damage is above 2.0x as the base multiplier the risk and reward is still high!

There's just so much irony is saying people want consitency in RNG... I can't.

YandereGirlfriend

I would rather not touch Critical Chance again but I would be all for increasing the competitiveness of other damage stats.

This would require:

• Reducing the Critical Damage stacking ceiling (or introducing Critical Resistance to PvE enemies)
• Increasing the "power budget" of item set bonuses (and, specifically, 5-piece bonuses...) such as Weapon/Spell Damage and Max Stats
• Increasing class-specific affinities for stacking alternative bonuses (e.g. Sorcerer Spell/Weapon Damage passive, Warden Northern Storm Max Magicka scaling, Templar unique Weapon Damage passive, etc.) so that different classes are incentivized to stack different types of damage stats

That would lead to a much healthier meta where different classes would reach peak damage output by stacking different damage stats rather than homogenizing everyone into stacking Critical Chance and Damage.

CP5

As it is, not scoring a crit just feels real bad. When you're tasked with killing a particular threat very quickly, and you see the damage number for your big attack not crit, at best it leaves a bad taste in your mouth and at worst causes a wipe (thinking scalecaller hard mode). Reducing crit chance only leads to more situations like this, because either scoring consistent critical hits would be needed to run content or the content would be balanced around not needing crits, and then having many would greatly change how the fight plays out. Consistency is needed for consistent content, and banking on getting lucky to do well isn't enjoyable. Just imagine whipping at the end of a 20min boss fight because rng isn't in your favor, and you just low roll in execute.

Syrpynt

YandereGirlfriend wrote: »

I would rather not touch Critical Chance again but I would be all for increasing the competitiveness of other damage stats.

This would require:

• Reducing the Critical Damage stacking ceiling (or introducing Critical Resistance to PvE enemies)
• Increasing the "power budget" of item set bonuses (and, specifically, 5-piece bonuses...) such as Weapon/Spell Damage and Max Stats
• Increasing class-specific affinities for stacking alternative bonuses (e.g. Sorcerer Spell/Weapon Damage passive, Warden Northern Storm Max Magicka scaling, Templar unique Weapon Damage passive, etc.) so that different classes are incentivized to stack different types of damage stats

That would lead to a much healthier meta where different classes would reach peak damage output by stacking different damage stats rather than homogenizing everyone into stacking Critical Chance and Damage.

I would take it even further and say that classes and races don't have stat passives or bonuses. Let people look how they want and play how they want. But that's a bit extreme, even for TES fans since SP titles always had racial passives and bonuses.

I'm just seeing that tieing stats to each class makes more people angry that they can't be a Breton Stamplar if weapon damage is the passive.

CP5 wrote: »

As it is, not scoring a crit just feels real bad. When you're tasked with killing a particular threat very quickly, and you see the damage number for your big attack not crit, at best it leaves a bad taste in your mouth and at worst causes a wipe (thinking scalecaller hard mode). Reducing crit chance only leads to more situations like this, because either scoring consistent critical hits would be needed to run content or the content would be balanced around not needing crits, and then having many would greatly change how the fight plays out. Consistency is needed for consistent content, and banking on getting lucky to do well isn't enjoyable. Just imagine whipping at the end of a 20min boss fight because rng isn't in your favor, and you just low roll in execute.

No no, I totally understand. But that is in essence what CHANCE is! If you don't like RNG, then play weapon/spell damage! That's my point entirely.

The critical chance mechanic is this game is basically no longer a chance. It's planned. That's not how chance is supposed to work.

Chance should be high risk, high reward. If you want raw damage and no chance, the game should have another alternative: Raw power.

When I say to cap crit chance, I'm not saying the rest of the game stays as is. Why wouldn't they rebalance the game around peak damage, healing, and survivability?

lolo_01b16_ESO

While I don't mind meta changes and love thinking about new build ideas, I doubt that your suggestion would have the desired result. It can work well, especially in slower games. But in eso you don't really have those abilities with high damage and long cooldown where a crit becomes meaningful.
In a different game I played, hunters had a special shot that first needed a combo to unlock, then had a 5s cast time and a 40s cooldown so getting a crit with that one or maybe even two in a row really felt powerful. But in Eso I think the highest single hit you get from any ability is using Guardians Wrath on a target in execute range. And there the question whether you get a crit or not will generally change the fight duration by just a single second if at all, so not really something that makes you feel like you had a lucky day.

Amerises

So I would agree that a proposed nerf wouldn't be bad. I'm certainly not opposed to making the game more difficult, and with the most recent changes this year lowering dps ceiling around 5%, that's okay. I am however against nerfing for nerfings sake. This last crit nerf really made people need to building MORE specifically into it, rather than looking at other builds. It really had the opposite effect, and now with the kilt being released, it made it even MORE critical to build around critical strikes.

What's interesting to me is, though I'm not opposed, you are certainly for. No one is making you run crit builds. And honestly, most end game guilds just want to see a parse of 80k+, which for most classes isn't too difficult with a non-meta build.

CP5

Player's wouldn't just sit back and let crit, if made even more valuable than it is now, go unused. Like last patch, making people crit less often for more damage only raised the value of the crit bonus, rendering sets that didn't add any to the bin. Adding a soft-cap on crit making you deal less damage for having more crit is also counterintuitive and would leave players more frustrated than anything.

wills43b14_ESO

Short answer:

Critical chance as close to 50% as possible while crit dmg is as high as possible maximizes dps variance. There is still an expectation, but the probability for extremes increases (i.e. if you think about the distribution dps is all over the place). Theoretically it will all converge... eventually... as n increases towards infinity.

It's already a nightmare trying to test things within existing RNG variance before you factor in crit potential. When I produce parses to show the upward extent, I like consistency. Yeah it means getting that top parse is rare, but I also know when to "expect" if you will when variance is low. When I don't know what to expect, because the variance is so high and I can't get a great estimate of the mean (need MANY repetitions), it's hard to know where I stand.

All I can do to alleviate that problem is compare critical rates from CMXs, but it's depressing. IMHO reducing variance is better for consistency and for helping players understand how good they are as a player. When variance is outrageously high (assuming equal skill level), it takes many repetitions to being to understand how you fall. Basic statistics there.




Long answer:
I will break out the simulations in R if anyone really wants . But so far no one has wanted that the other times I've talked about this.

wills43b14_ESO

For skills like unstable wall and orb, which tick a lot, this isn't a big deal. But, our biggest hitting skills (atro, frags, incap, nightblade bow, destro ult, etc.) are generally much lower n, so their sampling variance is pretty extreme. I did some simulations with batches of 10-30 frag casts and simulated the total damage done... pretty wild. It's normal at 50% crit, but as crit approaches 0% or 100% the density is heavily skewed to one side, which is favorable imo. This means you will almost always be about the mean with very few parses either very low or very high (depending on whether you're close to 0% or 100%).

wills43b14_ESO

Syrpynt wrote: »

WrathOfInnos wrote: »

As predicted, I’m going to disagree. As you mentioned, doing good damage with low crit chance relies heavily on luck. I don’t think that success or failure should be determined by RNG, things need to be repeatable. Now it could be said that the current system also has plenty of luck built in (especially for Sorcs with frag procs), and if there were a way to smooth that out I would be all for it. IMO damage was much more consistent when everyone was running 80%+ crit chance and lower crit damage (Elsweyr through Greymoor IIRC), now with the ~60% crit meta and high crit damage the variance is much higher. I’m opposed to continuing that trend with another crit chance nerf or crit damage buff.

Interestingly, you can think about the absolute extremes, 0% crit chance or 100% crit chance. Both result in entirely repeatable DPS performance. I’d be in favor of either of these, as long as other stats were tuned down to ensure 100% crit builds are not overpowered (which has been a reality before, maybe even now), or tuned up so that 0% crit builds did approx the same damage they do right now.

Edit: I’ll also add that it’s not uncommon to reach 2.5X multiplier on crit this patch. Nightblades with Shadow can even achieve 2.78X, but IMO that’s only worth trying with Acuity, and typically they’d be better of using Thief for more crit chance.

Critical chance isn't about repeatability, at a fundamental level, it is literally RNG unless you are at 100% chance--people are thinking about raw damage when they talk about "consitency." So I don't see how your disagreement makes any sense.

What I'm looking for is what "amount of critical multiplier" would make it on par with "raw power" with a 40% cap so that other builds are viable. Is it 2.5x? Is it 2.0x? Is it 3.0x? And when I say critical multiplier, I mean unbuffed from other sources. This is base critical damage done.

Critical chance is literally: Your character got lucky and hit the JACKPOT!

If you are allowed to always hit jackpot, then of course it's going to be the lost powerful--but my point is there's no real "gamble" after a certain chance (currently). That's what isn't healthy about it.

RNG = High risk, high reward.



If people are opposed to a cap, I propose another method:

Every 5% over a 40% critical chance, has a critical failure chance of 5%, where you do 1/2x base damage. But I still would say that critical damage is above 2.0x as the base multiplier the risk and reward is still high!

There's just so much irony is saying people want consitency in RNG... I can't.

You missed the boat entirely friend. You're relying on the value of a point estimate, regardless of the uncertainty of that estimate. Rely on distributions and probability, not point estimates. When you compare "raw power" it's misleading. You're comparing a point estimate, but if the uncertainty is high, then inevitably people will opt to bank on chance. I can break out the simulation densities if you'd like so you can see that even 75-80% crit chance is far more desirable than 50-60% chance. There's still variation, but it's far more predictable. If dps is unpredictable, it quite literally means skill is less influential. But if skill is not influential, then people won't care to play the game at a competitive level, because people who are less skillful are almost equally likely to perform as well (this is at the extremes, ofc).

wills43b14_ESO

Actually, I'm on my soapbox, so here's some histograms I generated for this awhile ago:

I believe this is a simulation of 10 frag casts total damage.





You can see here that even if you WANT some rng, it's better to keep crit dmg low if you want crit chance rng.

If you want to keep crit damage REALLY HIGH, then crit chance needs to be very low or very high.

wills43b14_ESO

YandereGirlfriend wrote: »

I would rather not touch Critical Chance again but I would be all for increasing the competitiveness of other damage stats.

This would require:

• Reducing the Critical Damage stacking ceiling (or introducing Critical Resistance to PvE enemies)
• Increasing the "power budget" of item set bonuses (and, specifically, 5-piece bonuses...) such as Weapon/Spell Damage and Max Stats
• Increasing class-specific affinities for stacking alternative bonuses (e.g. Sorcerer Spell/Weapon Damage passive, Warden Northern Storm Max Magicka scaling, Templar unique Weapon Damage passive, etc.) so that different classes are incentivized to stack different types of damage stats

That would lead to a much healthier meta where different classes would reach peak damage output by stacking different damage stats rather than homogenizing everyone into stacking Critical Chance and Damage.

Good luck adjusting the power budget around the variance. That will be a nightmare for ZOS. They'll probably use the expectations (means), which isn't a great way to go around it whenever the uncertainty (variance) is very large.

Merforum

Syrpynt wrote: »

I know I'm going to get a LOT of resitance to this, but here it goes:

I believe the game would be healthier and more diverse if critical chance was capped at 40%, BUT critical damage bonuses would replace some set slots that improve chance.

This is because critical chance is supposed to be an old school "skill" called "luck." It was mainly a high risk, yet high reward, feature.

My goal in this proposal is to make spell/weapon damage the relatively "constant"/reliable damage metric.

While Stamina and Magicka translates to your proficinecy in "healing done" as well as your total durability in a fight.

And lastly critical attacks/heals will do beyond ~2.5x damage when an attack/heal crits. 50% chance would be too much for a bonus multiplier of 2.5x damage. What I'm proposing basically makes "luck" (RNG) not a trait you can exploit and follow as the "only way to get the most damage" in ESO. The game is boring without build variety in end-game to have fun.

Anyway, expecting a lot of disagreement here, but I'm not really interested in any person's opinion that disagrees just because ALL of their dps builds are currently set up for the crit meta and don't want to lose their "kingly status" as a dps lord. That's not an argument, that's a whine.

I have suggested this for years. The easiest best thing ZOS could do is implement hard CAPs on stuff that is being abused/meta such as crit chance/damage. Most other games never let crit chance be above 50% and have other requirement like being in stealth or behind the target or only allowing crit on INITIAL 'surprise' attack but NOT after battle begins. Crit is never supposed to be just FREE additional damage, like it has always been in ESO.

Lapin_Logic

No, Clearing content and getting a desired reward already has FAR too much RNG, as for PVP crit shouldn't be a dead stat, no one would be able to time a "Burst" from a stat you have no control over, seriously try playing a magblade in anything with 30% crit, the whole class is built around crits not power.

Instead we should get to the bottom of "why you hate crit chance" because if you build for it then you are sacrificing other aspects of your build.

Syrpynt

wills43b14_ESO wrote: »

Actually, I'm on my soapbox, so here's some histograms I generated for this awhile ago:

I believe this is a simulation of 10 frag casts total damage.





You can see here that even if you WANT some rng, it's better to keep crit dmg low if you want crit chance rng.

If you want to keep crit damage REALLY HIGH, then crit chance needs to be very low or very high.

None of your stats show 40% chance, which is where I wanted to see that as the cap.

And great, you can show the charts and defend that the dependability drops, but that's my argument about RNG--ESO is FAKING critical chance to be actual chance when you can build 60-80%. my visuals for what chance should look like with an ideal/perfect cosine wave to normalize the attacks.

Y axis is "damage", X axis is time in seconds.



Yes I know that actual RNG data would not look this ideal. Bare with me. Less critical chance would look closer to the shallow cosine wave for average dps, and more critical chance would "rock the boat" but the average would still be the same.

This means you can have raw power and critical chance builds and they'll on average be the same power. But if people are trying to measure their reliability for crit chance, but don't want to do the statistics of it--then is that excuse to allow for 60-80% chance to make it more "reliable" and overpowered? No. No at all.

Crit chance is not MEANT to be reliable. If you want reliability, you go for raw power. If you want wild chance to do crazy damage beyond the "average", critical chance should strike HARD and it will feel good when it does. But asking for it to hit hard ALL THE TIME is just taking place of raw power's purpose! So although it sucks to not crit when you want it to, it will crit sometimes in a GREAT way that will be worth the risk to some. THOSE are the players that know what RNG is truly about.

I don't even want to get into the fact that RNG on computers aren't actually "randomly" generated, but the main point is that ESO's crit meta is being exploited and replacing the dependability of what RAW POWER is supposed to do.

Whether you do or don't want to hear it, that's the fact. Critical chance in this game can bypass the "chance" aspect of it and make it more dependable, and the strongest builds primarily because it isn't left up to chance anymore. There's no space for players to come up short, thus it's not really chance--but it's planned. I cannot repeat/emphasize this enough.

Edit: Typos, clarification on some topics.

FrancisCrawford

wills43b14_ESO wrote: »

YandereGirlfriend wrote: »

I would rather not touch Critical Chance again but I would be all for increasing the competitiveness of other damage stats.

This would require:

• Reducing the Critical Damage stacking ceiling (or introducing Critical Resistance to PvE enemies)
• Increasing the "power budget" of item set bonuses (and, specifically, 5-piece bonuses...) such as Weapon/Spell Damage and Max Stats
• Increasing class-specific affinities for stacking alternative bonuses (e.g. Sorcerer Spell/Weapon Damage passive, Warden Northern Storm Max Magicka scaling, Templar unique Weapon Damage passive, etc.) so that different classes are incentivized to stack different types of damage stats

That would lead to a much healthier meta where different classes would reach peak damage output by stacking different damage stats rather than homogenizing everyone into stacking Critical Chance and Damage.

Good luck adjusting the power budget around the variance. That will be a nightmare for ZOS. They'll probably use the expectations (means), which isn't a great way to go around it whenever the uncertainty (variance) is very large.

It's still not clear what genuine problem you're trying to solve:

• Players are too stupid to understand combat logs with high variance (so we should dumb down the game for them)?
  [snip]
• Fights with identically good or bad player rotations might last a different number of seconds when replayed (which is bad because -- why?)?

[edited for bashing]

CP5

FrancisCrawford wrote: »

wills43b14_ESO wrote: »

YandereGirlfriend wrote: »

I would rather not touch Critical Chance again but I would be all for increasing the competitiveness of other damage stats.

This would require:

• Reducing the Critical Damage stacking ceiling (or introducing Critical Resistance to PvE enemies)
• Increasing the "power budget" of item set bonuses (and, specifically, 5-piece bonuses...) such as Weapon/Spell Damage and Max Stats
• Increasing class-specific affinities for stacking alternative bonuses (e.g. Sorcerer Spell/Weapon Damage passive, Warden Northern Storm Max Magicka scaling, Templar unique Weapon Damage passive, etc.) so that different classes are incentivized to stack different types of damage stats

That would lead to a much healthier meta where different classes would reach peak damage output by stacking different damage stats rather than homogenizing everyone into stacking Critical Chance and Damage.

Good luck adjusting the power budget around the variance. That will be a nightmare for ZOS. They'll probably use the expectations (means), which isn't a great way to go around it whenever the uncertainty (variance) is very large.

It's still not clear what genuine problem you're trying to solve:

• Players are too stupid to understand combat logs with high variance (so we should dumb down the game for them)?
  [snip]
• Fights with identically good or bad player rotations might last a different number of seconds when replayed (which is bad because -- why?)?

Let's say a fight has a particular target that needs to die in x seconds or else you wipe. Let's say on average with or without this change the player is able to kill this target and not wipe. With a high crit rate they are able to do this consistently and can manage a few hits not critting. Now if there is a huge difference between crits and normal hits, and the crit rate is low, the player is likely to either get lucky, score some extra crits and clear the target faster, or not score enough crits and die. This would require base damage to be high enough to clear without gambling on your success in the fight, forgoing crit chance, or leave crits so sought after you would build for them far more than people already do (since zos lowered crit rate and raised crit damage we already see how players react). This change would only make crit rate even more highly required.

[edited to remove quote]

Syrpynt

CP5 wrote: »

Let's say a fight has a particular target that needs to die in x seconds or else you wipe. Let's say on average with or without this change the player is able to kill this target and not wipe. With a high crit rate they are able to do this consistently and can manage a few hits not critting. Now if there is a huge difference between crits and normal hits, and the crit rate is low, the player is likely to either get lucky, score some extra crits and clear the target faster, or not score enough crits and die.

Base damage needs to be fixed if crit damage gets adjusted.

Theoretically:

Total average power =
( BP*(100% - CC%) + BP*(1.5 + CDB)*(CC%) ) / (2 * 100%)

BD is base power
CC% is critical chance
CDB is summed Critical Damage bonuses in decimal form.
It's all divided by 2 to average the critical power and base power, and 100% to bring it back to decimal value form.

There should be a cap for critical chance, and sets should more base-power options.

CP5 wrote: »

This would require base damage to be high enough to clear without gambling on your success in the fight, forgoing crit chance, or leave crits so sought after you would build for them far more than people already do (since zos lowered crit rate and raised crit damage we already see how players react).

That's exactly the point. The game shouldn't be built with a meta hard-on for critical builds, and base damage should be higher, but if you want to "take the chance" and make a critical build (pun intended), then you are resting your skills on RNG to do 2.0x or 3.0x your base damage but less frequently (and I mean, with a chance cap of 40% +/- 5%).

This would also make critical strikes more like a true RNG-type of finisher/execution!

CP5 wrote: »

This change would only make crit rate even more highly required.

Not if base power became competitive enough, and critical chance was capped. But to compensate for critical chance cap--critical damage would be increased from 1.5x to 2.0 or 2.5. This would make combat less reliable, and more on luck. Which is what chance/luck is about.

zvavi

6 reasons why no:
1. Consistency in mechanics. Some places require damage tests, and leaving it to rng if you can or can't pass it is pretty stupid way to decide your group's wipe. Therefore making those changes is bad.
2. Consistency in PvP. Dying to rng is the least fun thing. It literally feels like ***. Therefore making those changes is bad.
3. Amount of outgoing damage ticks. Let's take the trial dummy as an example. Parses have over 1000 hits on them. At the high (yes, high) crit chance of 40%, it will still be consistent enough to crit 40%~ of the time. Therefore in non short dps tests, it would still be not very "luck based". Therefore making those changes is pointless.
4. Let's address your "weaker attacks penalty" which is silly as heck, it punishes people that build for crit. Adds server calculations. Punishes people for having sorc or NB in group if they don't have a "crit build". Over all this part of the idea is worse.
5. Your whole premise of game being more diverse of crit is capped is not true, because the current meta is wearing no 5 piece crit sets (people are wearing kilt, but I wouldn't call it a crit set, it is more of a p2w flavor of the year item, just like stranglers were). More than that, in dungeons where you don't get high warhorn, brittle and catalyst building for crit is not that important. There are a lot of sets that are used in all kinds of situations right now, and most sets are within 5% of damage of one another, it is pretty diverse so no change is needed.
6. Even if you make crit useless (ye, I said it, useless) there will be another flavor of the year best stat, and everyone will be building for it. So there will not even be more build diversity. Because it all comes down to the math the top groups are doing. Within long fights and lots of damage ticks the average crit hits would always be around the crit chance. So there will not be more build diversity.

FrancisCrawford

CP5 wrote: »

FrancisCrawford wrote: »

wills43b14_ESO wrote: »

YandereGirlfriend wrote: »

I would rather not touch Critical Chance again but I would be all for increasing the competitiveness of other damage stats.

This would require:

• Reducing the Critical Damage stacking ceiling (or introducing Critical Resistance to PvE enemies)
• Increasing the "power budget" of item set bonuses (and, specifically, 5-piece bonuses...) such as Weapon/Spell Damage and Max Stats
• Increasing class-specific affinities for stacking alternative bonuses (e.g. Sorcerer Spell/Weapon Damage passive, Warden Northern Storm Max Magicka scaling, Templar unique Weapon Damage passive, etc.) so that different classes are incentivized to stack different types of damage stats

That would lead to a much healthier meta where different classes would reach peak damage output by stacking different damage stats rather than homogenizing everyone into stacking Critical Chance and Damage.

Good luck adjusting the power budget around the variance. That will be a nightmare for ZOS. They'll probably use the expectations (means), which isn't a great way to go around it whenever the uncertainty (variance) is very large.

It's still not clear what genuine problem you're trying to solve:

• Players are too stupid to understand combat logs with high variance (so we should dumb down the game for them)?
• Devs are too stupid to understand combat logs with high variance (so they should dumb down the game for themselves)?
• Fights with identically good or bad player rotations might last a different number of seconds when replayed (which is bad because -- why?)?

Let's say a fight has a particular target that needs to die in x seconds or else you wipe. Let's say on average with or without this change the player is able to kill this target and not wipe. With a high crit rate they are able to do this consistently and can manage a few hits not critting. Now if there is a huge difference between crits and normal hits, and the crit rate is low, the player is likely to either get lucky, score some extra crits and clear the target faster, or not score enough crits and die. This would require base damage to be high enough to clear without gambling on your success in the fight, forgoing crit chance, or leave crits so sought after you would build for them far more than people already do (since zos lowered crit rate and raised crit damage we already see how players react). This change would only make crit rate even more highly required.

In which particular fights have you experienced this problem?

The only close-call DPS check I've had for a while is when I took a garbage build into Spindleclutch 2 to see if that was a good place to farm vampires for the endeavor, and had to solo Bloodspawn (and another boss) before I had enough vampires racked up. That was not a situation likely to repeat, and one I could have averted with a better build anyway.

Actually, there also were a few times recently when @DarcyMardin and I struggled to duo the final boss of a vet dungeon we can breeze through on normal. But any time I've been in such a situation, my group (of 1 or 2 or 4 or whatever) is probably doing some hundreds of parcels of crittable damage across the whole fight. So how far off from the mean is total damage likely to be?

zvavi

FrancisCrawford wrote: »

CP5 wrote: »

FrancisCrawford wrote: »

wills43b14_ESO wrote: »

YandereGirlfriend wrote: »

I would rather not touch Critical Chance again but I would be all for increasing the competitiveness of other damage stats.

This would require:

• Reducing the Critical Damage stacking ceiling (or introducing Critical Resistance to PvE enemies)
• Increasing the "power budget" of item set bonuses (and, specifically, 5-piece bonuses...) such as Weapon/Spell Damage and Max Stats
• Increasing class-specific affinities for stacking alternative bonuses (e.g. Sorcerer Spell/Weapon Damage passive, Warden Northern Storm Max Magicka scaling, Templar unique Weapon Damage passive, etc.) so that different classes are incentivized to stack different types of damage stats

That would lead to a much healthier meta where different classes would reach peak damage output by stacking different damage stats rather than homogenizing everyone into stacking Critical Chance and Damage.

Good luck adjusting the power budget around the variance. That will be a nightmare for ZOS. They'll probably use the expectations (means), which isn't a great way to go around it whenever the uncertainty (variance) is very large.

It's still not clear what genuine problem you're trying to solve:

• Players are too stupid to understand combat logs with high variance (so we should dumb down the game for them)?
• Devs are too stupid to understand combat logs with high variance (so they should dumb down the game for themselves)?
• Fights with identically good or bad player rotations might last a different number of seconds when replayed (which is bad because -- why?)?

Let's say a fight has a particular target that needs to die in x seconds or else you wipe. Let's say on average with or without this change the player is able to kill this target and not wipe. With a high crit rate they are able to do this consistently and can manage a few hits not critting. Now if there is a huge difference between crits and normal hits, and the crit rate is low, the player is likely to either get lucky, score some extra crits and clear the target faster, or not score enough crits and die. This would require base damage to be high enough to clear without gambling on your success in the fight, forgoing crit chance, or leave crits so sought after you would build for them far more than people already do (since zos lowered crit rate and raised crit damage we already see how players react). This change would only make crit rate even more highly required.

In which particular fights have you experienced this problem?

The only close-call DPS check I've had for a while is when I took a garbage build into Spindleclutch 2 to see if that was a good place to farm vampires for the endeavor, and had to solo Bloodspawn (and another boss) before I had enough vampires racked up. That was not a situation likely to repeat, and one I could have averted with a better build anyway.

vSS HM.

hundergrn

IMO... Crit chance doesn't necessarily equate to Luck or RNG.

When testing builds, taking parses, against target dummies that are equivalent to end game content health and resistance you quickly find that the logs read fairly close to the prescribed percentages of Crit Chance. Whether it be 20,40,60,70%, if you hit the slab of meat (stone?) long enough the highs and lows even out to the build you present and give you an idea of how it should flow during longer fights.

The recent changes to Crit accumulation that brought the high end of pure crit from 70+ to ~60% while going full in on Crit chance only took Crit builds to push Crit chance and bonus and limited build variety for the play style. Further hindering Crit chance will only push those builds to cap quicker to focus on Bonus and Damage limiting the limited pool even further and possibly even finding larger dmg numbers to hit.

Min/Max players exist. They will either push full damage with a 25-35% Crit chance and as much damage they can muster between set bonus, buffs, wep dmg, and skills.

For my 2 NB, I have a few different builds for them. Some new, Some old.

The Magblade PvE builds sit at 2894 Spell DMG base with 13k Pen, and 50--60 Crit Chance depending on the set.
The Stamblade PvP build sits at 5640 Wep DMG base with 11k Pen, and 33.2% Crit Chance.
The Stamblade PvE build sits at 4670 Wep DMG base with 3.8k Pen, and 60.2% Crit Chance.

When I take them through the same PvE content, they do about the same DPS. When I parse them with a trail dummy the DPS between the 3 variations come within a 1-3k DPS in variation of each other. Why? Because with the same Crit Bonus Damage, the variation of Wep/Spell DMG, Penatration, and Crit Chance evens out over time.

TLDR if crit chance is capped at 40%, the limitation will push for capping Crit DMG Bonus and Weapon/Spell DMG. Dilute the build variety for players that want to maximize their toons, and chance next to nothing for the people who just follow meta (outside of gathering different gear)

FrancisCrawford

zvavi wrote: »

FrancisCrawford wrote: »

CP5 wrote: »

FrancisCrawford wrote: »

wills43b14_ESO wrote: »

YandereGirlfriend wrote: »

I would rather not touch Critical Chance again but I would be all for increasing the competitiveness of other damage stats.

This would require:

• Reducing the Critical Damage stacking ceiling (or introducing Critical Resistance to PvE enemies)
• Increasing the "power budget" of item set bonuses (and, specifically, 5-piece bonuses...) such as Weapon/Spell Damage and Max Stats
• Increasing class-specific affinities for stacking alternative bonuses (e.g. Sorcerer Spell/Weapon Damage passive, Warden Northern Storm Max Magicka scaling, Templar unique Weapon Damage passive, etc.) so that different classes are incentivized to stack different types of damage stats

That would lead to a much healthier meta where different classes would reach peak damage output by stacking different damage stats rather than homogenizing everyone into stacking Critical Chance and Damage.

Good luck adjusting the power budget around the variance. That will be a nightmare for ZOS. They'll probably use the expectations (means), which isn't a great way to go around it whenever the uncertainty (variance) is very large.

It's still not clear what genuine problem you're trying to solve:

• Players are too stupid to understand combat logs with high variance (so we should dumb down the game for them)?
• Devs are too stupid to understand combat logs with high variance (so they should dumb down the game for themselves)?
• Fights with identically good or bad player rotations might last a different number of seconds when replayed (which is bad because -- why?)?

Let's say a fight has a particular target that needs to die in x seconds or else you wipe. Let's say on average with or without this change the player is able to kill this target and not wipe. With a high crit rate they are able to do this consistently and can manage a few hits not critting. Now if there is a huge difference between crits and normal hits, and the crit rate is low, the player is likely to either get lucky, score some extra crits and clear the target faster, or not score enough crits and die. This would require base damage to be high enough to clear without gambling on your success in the fight, forgoing crit chance, or leave crits so sought after you would build for them far more than people already do (since zos lowered crit rate and raised crit damage we already see how players react). This change would only make crit rate even more highly required.

In which particular fights have you experienced this problem?

The only close-call DPS check I've had for a while is when I took a garbage build into Spindleclutch 2 to see if that was a good place to farm vampires for the endeavor, and had to solo Bloodspawn (and another boss) before I had enough vampires racked up. That was not a situation likely to repeat, and one I could have averted with a better build anyway.

vSS HM.

How many player-seconds do the DPS checks take? If it's in the low hundreds (e.g. 3 players, 90 seconds) then I'd doubt the RNG aspect is all that big of a deal.

zvavi

FrancisCrawford wrote: »

zvavi wrote: »

FrancisCrawford wrote: »

CP5 wrote: »

FrancisCrawford wrote: »

wills43b14_ESO wrote: »

YandereGirlfriend wrote: »

I would rather not touch Critical Chance again but I would be all for increasing the competitiveness of other damage stats.

This would require:

• Reducing the Critical Damage stacking ceiling (or introducing Critical Resistance to PvE enemies)
• Increasing the "power budget" of item set bonuses (and, specifically, 5-piece bonuses...) such as Weapon/Spell Damage and Max Stats
• Increasing class-specific affinities for stacking alternative bonuses (e.g. Sorcerer Spell/Weapon Damage passive, Warden Northern Storm Max Magicka scaling, Templar unique Weapon Damage passive, etc.) so that different classes are incentivized to stack different types of damage stats

That would lead to a much healthier meta where different classes would reach peak damage output by stacking different damage stats rather than homogenizing everyone into stacking Critical Chance and Damage.

Good luck adjusting the power budget around the variance. That will be a nightmare for ZOS. They'll probably use the expectations (means), which isn't a great way to go around it whenever the uncertainty (variance) is very large.

It's still not clear what genuine problem you're trying to solve:

• Players are too stupid to understand combat logs with high variance (so we should dumb down the game for them)?
• Devs are too stupid to understand combat logs with high variance (so they should dumb down the game for themselves)?
• Fights with identically good or bad player rotations might last a different number of seconds when replayed (which is bad because -- why?)?

Let's say a fight has a particular target that needs to die in x seconds or else you wipe. Let's say on average with or without this change the player is able to kill this target and not wipe. With a high crit rate they are able to do this consistently and can manage a few hits not critting. Now if there is a huge difference between crits and normal hits, and the crit rate is low, the player is likely to either get lucky, score some extra crits and clear the target faster, or not score enough crits and die. This would require base damage to be high enough to clear without gambling on your success in the fight, forgoing crit chance, or leave crits so sought after you would build for them far more than people already do (since zos lowered crit rate and raised crit damage we already see how players react). This change would only make crit rate even more highly required.

In which particular fights have you experienced this problem?

The only close-call DPS check I've had for a while is when I took a garbage build into Spindleclutch 2 to see if that was a good place to farm vampires for the endeavor, and had to solo Bloodspawn (and another boss) before I had enough vampires racked up. That was not a situation likely to repeat, and one I could have averted with a better build anyway.

vSS HM.

How many player-seconds do the DPS checks take? If it's in the low hundreds (e.g. 3 players, 90 seconds) then I'd doubt the RNG aspect is all that big of a deal.

It isn't big deal. But op wants to create a system that will do so. he wants to have rng based builds. Putting aside the fact that his ideas don't support his vision, the argument of dps checks exist is about his vision.

CP5

Every 10% you have to kill 3 targets and if by chance with the suggested changes you low roll on one or two of the many targets that are needed people will die.

And to continue to elaborate on what people have added. The more value you add to critical hits the more people would build for them. If you make base attacks hit harder and crits hit harder than people will want more critical hit damage, the punishing soft cap would only create frustration, or crit would become worthless. ESO doesn't have a way to do things like 'hitting weak spots' and pegging crits on things like flanking would make some fights laughably easy while denying crits from happening in others. If anything, the trend to a lower crit % higher crit damage has done the opposite of what the OP is looking for.

AcadianPaladin

Right now high dps includes lots of rotation practice, proficiency with animation canceling, and stacking crit. Sure there are some exceptions but that is to swim against the flow of the river.

Amerises

zvavi wrote: »

FrancisCrawford wrote: »

zvavi wrote: »

FrancisCrawford wrote: »

CP5 wrote: »

FrancisCrawford wrote: »

wills43b14_ESO wrote: »

YandereGirlfriend wrote: »

I would rather not touch Critical Chance again but I would be all for increasing the competitiveness of other damage stats.

This would require:

• Reducing the Critical Damage stacking ceiling (or introducing Critical Resistance to PvE enemies)
• Increasing the "power budget" of item set bonuses (and, specifically, 5-piece bonuses...) such as Weapon/Spell Damage and Max Stats
• Increasing class-specific affinities for stacking alternative bonuses (e.g. Sorcerer Spell/Weapon Damage passive, Warden Northern Storm Max Magicka scaling, Templar unique Weapon Damage passive, etc.) so that different classes are incentivized to stack different types of damage stats

That would lead to a much healthier meta where different classes would reach peak damage output by stacking different damage stats rather than homogenizing everyone into stacking Critical Chance and Damage.

Good luck adjusting the power budget around the variance. That will be a nightmare for ZOS. They'll probably use the expectations (means), which isn't a great way to go around it whenever the uncertainty (variance) is very large.

It's still not clear what genuine problem you're trying to solve:

• Players are too stupid to understand combat logs with high variance (so we should dumb down the game for them)?
• Devs are too stupid to understand combat logs with high variance (so they should dumb down the game for themselves)?
• Fights with identically good or bad player rotations might last a different number of seconds when replayed (which is bad because -- why?)?

Let's say a fight has a particular target that needs to die in x seconds or else you wipe. Let's say on average with or without this change the player is able to kill this target and not wipe. With a high crit rate they are able to do this consistently and can manage a few hits not critting. Now if there is a huge difference between crits and normal hits, and the crit rate is low, the player is likely to either get lucky, score some extra crits and clear the target faster, or not score enough crits and die. This would require base damage to be high enough to clear without gambling on your success in the fight, forgoing crit chance, or leave crits so sought after you would build for them far more than people already do (since zos lowered crit rate and raised crit damage we already see how players react). This change would only make crit rate even more highly required.

In which particular fights have you experienced this problem?

The only close-call DPS check I've had for a while is when I took a garbage build into Spindleclutch 2 to see if that was a good place to farm vampires for the endeavor, and had to solo Bloodspawn (and another boss) before I had enough vampires racked up. That was not a situation likely to repeat, and one I could have averted with a better build anyway.

vSS HM.

How many player-seconds do the DPS checks take? If it's in the low hundreds (e.g. 3 players, 90 seconds) then I'd doubt the RNG aspect is all that big of a deal.

It isn't big deal. But op wants to create a system that will do so. he wants to have rng based builds. Putting aside the fact that his ideas don't support his vision, the argument of dps checks exist is about his vision.

VCR is like this as well. Most people pushing really and HM DLC trials and some dungeons are beating it by seconds not minutes. The reason people require high dps for Vet DLC trials is because more time = higher chance of mistakes = DoA. Rolling a dice for end game content doesn't sound fun.

wills43b14_ESO

Syrpynt wrote: »

wills43b14_ESO wrote: »

Actually, I'm on my soapbox, so here's some histograms I generated for this awhile ago:

I believe this is a simulation of 10 frag casts total damage.





You can see here that even if you WANT some rng, it's better to keep crit dmg low if you want crit chance rng.

If you want to keep crit damage REALLY HIGH, then crit chance needs to be very low or very high.

None of your stats show 40% chance, which is where I wanted to see that as the cap.

And great, you can show the charts and defend that the dependability drops, but that's my argument about RNG--ESO is FAKING critical chance to be actual chance when you can build 60-80%. my visuals for what chance should look like with an ideal/perfect cosine wave to normalize the attacks.

Y axis is "damage", X axis is time in seconds.



Yes I know that actual RNG data would not look this ideal. Bare with me. Less critical chance would look closer to the shallow cosine wave for average dps, and more critical chance would "rock the boat" but the average would still be the same.

This means you can have raw power and critical chance builds and they'll on average be the same power. But if people are trying to measure their reliability for crit chance, but don't want to do the statistics of it--then is that excuse to allow for 60-80% chance to make it more "reliable" and overpowered? No. No at all.

Crit chance is not MEANT to be reliable. If you want reliability, you go for raw power. If you want wild chance to do crazy damage beyond the "average", critical chance should strike HARD and it will feel good when it does. But asking for it to hit hard ALL THE TIME is just taking place of raw power's purpose! So although it sucks to not crit when you want it to, it will crit sometimes in a GREAT way that will be worth the risk to some. THOSE are the players that know what RNG is truly about.

I don't even want to get into the fact that RNG on computers aren't actually "randomly" generated, but the main point is that ESO's crit meta is being exploited and replacing the dependability of what RAW POWER is supposed to do.

Whether you do or don't want to hear it, that's the fact. Critical chance in this game can bypass the "chance" aspect of it and make it more dependable, and the strongest builds primarily because it isn't left up to chance anymore. There's no space for players to come up short, thus it's not really chance--but it's planned. I cannot repeat/emphasize this enough.

Edit: Typos, clarification on some topics.

You're missing A LOT of points that I made... although tbf I'm mostly speaking towards PvE if that makes a difference. I'm not an expert at PvP, so I won't discuss the tradeoffs of building crit there.

First: Okay, sure, I didn't show 40%. I don't have time to simulate it right now, but it's going to look close to the 50% examples, but slightly shifted towards the skewed example (but on the left). The points I make still stand. If you want crit at 40-50%, then crit damage shouldn't be skyrocketed (this is one way to reduce variance) and then they can just increase base dmg. If they have moderate crit and high crit dmg this makes for extremely unstable hits. Maybe in PvP this is okay, but this is horrible design for PvE. The other option I suggested was high or low crit, but high crit dmg. At best, ZOS would need to make it to where even most *bad* instances would be more than enough, but then that makes content too easy. Content should be skill based, not RNG. You'll have everyone rage quit instantly if it's only RNG.

Second: the stats is actually very hard - they'd probably have to simulate it... which is hard even then. I presented several options to fix the general problem with reliability though. Regarding your graph, That's fine for very large n. With a lot of ticks (like wall/orb/dots/etc.) this isn't too big of a problem, because the sampling variance will be shrunk ( sqrt(variance / n) ). The problem, in PvE at least, is that things like frag are already rng, lining up the rng at the right moment when buffs/debuffs are maxed, no adding in crit, that it's a *** show tbh. In PvP, I think this is more desireable, but there are generally tradeoffs to building a lot of crit as you put it. In PvE, you can't really build high crit anyways because crit dmg gets shrunk and then it's easily suboptimal (even if reliable) to a good chunk of the high variance builds. The problem is if ZOS keeps going in this direction, which has been the trend for 2-3 patches.

Third: raw power vs crit.... ZOS doesn't allow for enough raw power to make this viable. If they did, the crits would be insane (they would have to MASSIVELY dial down crit dmg for pvp, which is my entire argument of what they SHOULD do in the first place)! We're talking 4x damage. Even at expectation or slightly lower, that's an INSANE amount of raw power to get those numbers. PvP would be so busted. Again, you're focusing too much on point estimates or very low probabilities and not overall probability/distributions. Biggest fallacy in interpreting stats... that's the casino syndrome.

Fourth: Yes, there is no such thing as coded chaos blah blah blah, but it works *well enough*, so I won't continue into that argument for now.

Fifth: I'm not suggesting to bypass chance at all. Crit by its very nature has some kind of distribution. I'm saying they've made the distribution too wide or are rapidly heading that direction. I'm not sure what you mean by strongest builds. People go with what is "expected" to give the greatest power, regardless of the variance. Although, some score pushers might try to farm rng, but in raids that's fairly impractical (raid rng + that rng = very very very very small chance). Also, to finish this thought: building large crit % means sacrificing base power and crit dmg already, so that point is moot imo.

wills43b14_ESO

I guess what I'm trying to say if your points make more sense to PvP for me (without knowing some of the intricacies of build trade off), but make no sense in PvE, especially where a fight is very long with a lot of ticks on most stuff besides a few heavy hitters... although those heavy hitters really make it frustrating when variance is too large and it's no longer about skill.

PvE should be almost entirely skill based with a small amount of rng. For large skill gaps the variance doesn't matter, but when it's close it does.