Stat Bonuses to Health Recovery are Inconsistent

twing1_

TL;DR

Problem:
Item set bonuses to health recovery are being valued at several different values from a number of different sources.
2-4 pc item set bonuses value a stat bonus to health recovery at 129 health recovery
5 pc item set bonuses value a stat bonus to health recovery anywhere between 340 and 387 health recovery (derived from green pact and beekeeper's)
3 pc item set bonuses value a stat bonus to health recovery at 387 health recovery (derived from endurance)
Mundus stones value a stat bonus to health recovery anywhere between 188 and 338 health recovery (derived from steed mundus)

Solution:
A single value for a stat bonus to health recovery should be agreed upon, and then all stat bonuses to health recovery should be made consistent with this value, across all sources of obtainable health recovery.

Proposal 1 (recommended):
Stat bonuses to health recovery should be centered around 387 health recovery.
2-4 pc item set bonuses should be adjusted to 387 health recovery
5 pc bonus from green pact should be adjusted to 2200 maximum health and 501 health recovery
5 pc bonus from beekeeper's is unchanged
3 pc bonus from endurance is unchanged
Steed mundus should be adjusted to 271 health recovery

Proposal 2:
Stat bonuses to health recovery should be centered around 129 health recovery.
2-4 pc item set bonuses are unchanged.
5 pc bonus from green pact should be adjusted to 2200 maximum health and 167 health recovery
5 pc bonus from beekeeper's should be adjusted to 300 health recovery
3 pc bonus from endurance should be adjusted to 206 health recovery
Steed mundus should be adjusted to 90 health recovery

Proposal 3:
A brand new value for stat bonuses to health recovery should be assigned, and all stat bonuses to health recovery should be balanced around this new value.

With the recent removal of all proc sets from cyrodil, some interesting light has been shed onto the value of stat bonuses. For those of you who don't know, all proc sets were recently temporarily removed from cyrodil for performance tests, leaving the only sets available to use to be those that add unconditional stat bonuses to your character sheet. This provides us with a perfect opportunity to directly compare these sets and their stat bonuses to one another, without the question mark of exactly how proc conditions affect the value of stat bonuses. For the most part, all of these set's 5 pc bonuses are consistent in their stat density, relative to their 2-4 pc item set bonus counterparts. With some simple math, we are able to determine that a 5 pc set bonus to raw stats is valued at around ~2.32x a single 2-4 pc item set bonus to that respective stat.
Math:

Spriggans/spinners:
3460 penetration = 2.32x 1487 penetration, where 1487 penetration is the single 2-4 pc item set bonus to penetration

Draugr hulk/crafty alfiq:
2550 max stam or mag = 2.32x 1096 max stam or mag, where 1096 is the single 2-4 pc item set bonus to max stam or mag

Hundings/julianos:
300 weapon or spell damage = 2.32x 129 weapon or spell damage, where 129 is the single 2-4 pc item set bonus to weapon or spell damage

and so on.

This ratio of 5 pc bonuses being ~2.32x the value of a single 2-4 pc item set bonus to the corresponding stat is consistent among all sets that grant raw statistics, except for when it comes to health recovery. This is where things start to get wonky. Beekeeper's gear gives us the most direct comparison. Let's take a look at it.

Beekeeper's gear:
900 health recovery, or 6.97x(!) 129, where 129 is the single 2-4 pc bonus to health recovery

Where other raw stat sets give only 2.32x the value of a single 2-4 pc item set bonus, beekeeper's gear gives a whopping 6.97x the value of a single 2-4 pc item set bonus to health recovery!

This could mean one of two things:
1) Beekeeper's gear is a broken set and is giving a whole lot more stats than it is intended to
or
2) single 2-4 pc item set bonuses to health recovery are giving far less stats than they are intended to

In order to check to see if the problem is with the individual beekeeper's gear 5 pc bonus, we need to compare additional item sets that give bonuses to health recovery. The next cleanest comparison is the Endurance set, even though it is only a set of 3. In order to compare this set, we must first examine similar sets in Willpower and Agility.

Willpower/agility:
2 pc: 1752 max resources, or 1.59x 1096, where 1096 is the single 2-4 pc bonus to max resources
3 pc: 206 damage, or 1.59x 129, where 129 is the single 2-4 pc bonus to damage

Endurance:
2 pc: 1928 max health, or 1.59x 1206, where 1206 is the single 2-4 pc bonus to max health
3 pc: 618 health recovery, or 4.79x(!) 129, where 129 is the single 2-4 pc bonus to health recovery

Health recovery again is being granted a whole lot more resources than bonuses to other similar stats. The problem doesn't appear to be solely the beekeeper's gear set, as the endurance set is exhibiting a similar issue.

This leads me to believe that option number 2 is the more likely:

1) Beekeeper's gear is a broken set and is giving a whole lot more stats than it is intended to
or
2) single 2-4 pc item set bonuses to health recovery are giving far less stats than they are intended to

But if 129 health recovery isn't the true, intended value for a single 2-4 pc item set bonus to health recovery, what is?

Lets see if we can reverse engineer a value from beekeeper's gear, by using the established ratio of a 5 pc bonus = 2.32x single 2-4 pc bonus found in all other stat sets of the game.

900 health recovery = 2.32x single 2-4 pc bonus to health recovery

900/2.32 = single 2-4 pc bonus to health recovery

387 = single 2-4 pc bonus to health recovery

If we were to take beekeeper's gear's stat density as a guide, it would seem that the true value for a single 2-4 pc item set bonus to health recovery should be valued at 387 health recovery. Lets cross reference this number by doing the same thing to the health recovery bonus granted by the endurance set.

618 health recovery = 1.59x single 2-4 pc bonus to health recovery

618/1.59 = single 2-4 pc bonus to health recovery

387 = single 2-4 pc bonus to health recovery

This seems to confirm that the true value for 2-4 pc item set bonuses to health recovery is exactly 387 health recovery.

Even though it is in my opinion too perfect to be mere coincidence that the exact same value for a single 2-4 pc bonus to health recovery was reverse engineered from both of these sets, we should still cross reference this with other sources of health recovery in the game to make sure that it is consistent.

In order to do this, lets take a look at another set: Green Pact

While this isn't as clean of a comparison as beekeeper's or endurance because it grants both max health and health recovery instead of just raw health recovery, the existence of its stamina and magicka counterparts, bone pirate's tatters and bright-throat's boast, help us out immensely. Each of these sets have identical proc conditions, and thus their stat densities should be identical as well, as evidenced by the values for bone pirate's stamina and stam recovery being identical to bright throat's magicka and mag recovery.

Each of these sets give:
2000 max resources, or 1.82x 1096 max resources, where 1096 is the single 2-4 pc bonus to max resources
and
167 resource recovery, or 1.29x 129 resource recovery, where 129 is the single 2-4 pc bonus to max resources.

Now, lets take a look at green pact:
2500 max health, or 2.07x 1206 max health, where 1206 is the single 2-4 pc bonus to max health
and
356 health recovery, or 2.75x 129, where 129 is the current single 2-4 pc bonus to health recovery

This set, too, seems to grant more stats to health recovery in comparison to its magicka/stamina counterparts. It is important to note, however, that the max stats granted from this set, too, is inconsistent with it's magicka/stamina counterparts. It is for this reason that in order to reverse engineer a single 2-4 pc item set bonus to health recovery from this set, I will be comparing the total stat density of this set against the total stat densities of its magicka/stamina counterparts.

bone pirate/bright throat:
Both of these 5 pc bonuses each give a total of 3.11x 2-4 pc item set bonuses. This is the number that should be allocated to green pact's 5 pc bonus, based off of the identical proc condition.

Green pact:
2500 max health equates to 2.07x a single 2-4 pc item set bonus to max health, which leaves its 356 health recovery to be valued at 1.04x (3.11-2.07=1.04) a single 2-4 pc bonus to health recovery.

356 = 1.04x single 2-4 pc bonus to health recovery

356/1.04 = single 2-4 pc bonus to health recovery

340 = single 2-4 pc bonus to health recovery

This is consistent with neither the current 2-4 pc bonus to health recovery of 129 health recovery nor the value we derived from other 5 pc bonus health recovery sets of 387 health recovery for a single 2-4 pc item set bonus. Because of these inconsistencies, and also the inconsistency of green pact offering a higher value of 2-4 pc bonuses to max stats as its magicka/stamina counterparts, I believe that this individual set is simply out of whack in itself and needs adjusting. The clear and obvious way of doing this would be adjusting its stat density to better mirror its magicka/stamina counterparts. Since bone pirate and bright throat each give 1.82x 2-4 pc bonus to max resources and 1.29x 2-4 pc bonus to resource recovery, it would make sense that green pact (the health counterpart) should be balanced around these values as well. That would look something like this:

1.82x 1206 max health = 2200 max health
1.29x 129 health recovery (current value) = 167 health recovery (or 501 health recovery, if our previously derived 2-4 pc set bonus to health recovery is to be adopted)

Because green pact proved itself to be inconsistent in too many ways, I hesitate to trust any values reverse engineered from this set, and will look into other sources of health recovery in the game. One obvious point of comparison would be Mundus stones, because they have a consistent ratio with 2-4 pc item set bonuses for other stats in the game. This value is ~1.85x 2-4 pc bonuses for most other stats, and ~2.40x 2-4 pc bonuses to resource recovery.

Math:

Tower/mage:
2023 max resources = ~1.84x 1096 max resources, where 1096 is the single 2-4 pc bonus to max resources

Warrior/apprentice:
238 damage = ~1.84x 129 damage, where 129 is the single 2-4 pc bonus to damage

Serpent/atronach:
310 resource recovery = ~2.40x 129 resource recovery, where 129 is the single 2-4 pc bonus to resource recovery

So lets take a look at the health recovery mundus stone, the steed:

10% movement speed
238 health recovery

The addition of movement speed onto this mundus stone makes the comparison a little tricky, but with the swift jewelry trait we should be able to compare the data.

It can be found that jewelry traits equate to ~.80x 2-4 pc bonuses to stats, as derived by the infused trait granting 104 (~.80x 129, the single 2-4 pc item bonus to damage) additional damage in conjunction with a weapon or spell damage glyph, and the protective trait granting 1190 (~.80x 1487, the single 2-4 pc bonus to resistance) additional resistances. It can then be assumed that the value of 7% movement speed that the swift trait grants is equal to ~.80x of what a 2-4 pc item bonus to movement speed would be. From this, we can reverse engineer the 2-4 pc bonus to movement speed:

7 = .80x 2-4 pc bonus

7/.80 = 2-4 pc bonus

8.74 = 2-4 pc bonus for movement speed

From this value, we can calculate the stat density given to the movement speed portion of the steed mundus.

10 movement speed/8.74 = 1.14x 2-4 pc bonus

Because all other mundus stones that grant bonuses to stats that are not resource recovery are given the value of ~1.84x 2-4 pc bonuses to stats and because the steed grants movement speed on top of its health recovery, I am inclined to use the 1.84x coefficient in stat density calculations, versus the 2.40x coefficient of the strictly resource recovery mundus stones. However, I will be checking the math in both scenarios to see if we can make any sense of it.

In the 1.84x coefficient scenario, the steed mundus is using 1.14x of its 2-4 pc bonuses on movement speed, leaving its 238 health recovery to be equal to .70x 2-4 pc bonus to health recovery. This can be used to find out what the steed mundus stone values a single 2-4 pc bonus to health recovery at.

238 = .70x 2-4 pc bonus to health recovery

238/.70 = 2-4 pc bonus to health recovery

338 = 2-4 pc bonus to health recovery

This is not consistent with any of our previous findings. I will now try deriving the 2-4 pc bonus to health recovery from the steed mundus using the coefficient for resource recovery mundus stones, because the steed does grant resource recovery after all.

2.40x 2-4 pc bonus = 1.14x 2-4 pc bonus (to speed) + 238 health recovery

2.40x 2-4 pc bonus - 1.14x 2-4 pc bonus = 238 health recovery

1.25x 2-4 pc bonus = 238 health recovery

2-4 pc bonus to health recovery = 238/1.25

2-4 pc bonus to health recovery = 188

This, too, is neither consistent with any of our previous findings, nor the current 2-4 pc bonus to health recovery found in the game. This leads me to believe that the steed mundus, just like the green pact 5 pc bonus, cannot be trusted, and must be adjusted. This can be done by using our derived value for the 2-4 pc bonus to movement speed and the current stat densities of other mundus stones. Like I mentioned earlier, because the steed grants more than just resource recovery, I feel like it is fair to use the 1.84x 2-4 pc bonus value found in all the other mundus stones that grant stats that are not resource recovery. This is what it would look like:

1.84x 2-4 pc bonus total, consisting of 1.14x 2-4 pc bonus to movement speed (10% speed) and .70x 2-4 pc bonus to health recovery.

This would put its values at:
10% movement speed
90 health recovery (at its current 2-4 pc bonus value of 129 health recovery), or 271 health recovery with our proposed 2-4 pc bonus value of 387 health recovery

Conclusion
The health recovery stat is a mess right now, with inconsistencies running rampant between the 2-4 pc bonuses, 5 pc bonuses, 3 pc bonuses, and mundus stone values for health recovery vs. those of other comparable character stats. Though the analysis is far from definitive in terms of what value the stat bonus to health recovery should land on, it is very clear that the health recovery stat must be standardized. The way I see it, there are 3 possible solutions to this.

Solution 1 (preferred):
2-4 pc bonuses to health recovery should be valued at 387 health recovery, and all other bonuses to health recovery should be centered around this value.

To make this a reality, the following must happen:

2-4 pc bonuses to health recovery must be adjusted to 387 health recovery
5 pc bonus of beekeeper remains unchanged at 900 health recovery
3 pc bonus of endurance remains unchanged at 618 health recovery
5 pc bonus of green pact must be adjusted to grant 2200 max health, and 501 health recovery to bring it in line with its magicka/stamina counterparts.
The steed mundus must be adjusted to 10% movement speed and 271 health recovery to bring its overall stat density in line with the other non-resource recovery mundus stones, or adjusted anywhere up to 487 health recovery to bring it in line with the other resource recovery mundus stones

This to me makes the most sense. This is in part because the most recently updated item sets, beekeeper's gear and endurance, both reverse engineered this exact value for the 2-4 pc bonus to health recovery, and it seems that this is the intended value. The connection between these two sets is actually the only consistency in terms of bonuses to health recovery that I have been able to find in the entire game. It also makes sense to bump up the value of health recovery in comparison to recovery of stamina or magicka, because of the accessibility to heals in the game. The current 2-4 pc bonus value of 129 health recovery is pitifully small in comparison to the vast amounts of healing players receive in most content, and even in PvP where health recovery has the biggest impact the value of 129 health recovery is almost negligible.

Solution 2:
2-4 pc bonuses to health recovery should remain at 129, and all other bonuses to health recovery should be centered around this value.

To make this a reality, the following must happen:

2-4 pc bonuses to health recovery remain unchanged at 129 health recovery
5 pc bonus of beekeeper must be adjusted to 300 health recovery to bring it in line with similar sets that grant other character stats
3 pc bonus of endurance must be adjusted to 206 health recovery to bring it in line with similar sets that grant other character stats
5 pc bonus of green pact must be adjusted to grant 2200 max health, and 167 health recovery to bring it in line with its magicka/stamina counterparts
The steed mundus must be adjusted to 10% movement speed and 90 health recovery to bring it in line with the other non-resource recovery mundus stones, or adjusted anywhere up to 162 health recovery to bring it in line with the other resource recovery mundus stones

This solution, in my opinion, is not the way to go about it. The changes above would solve the inconsistencies across the health recovery statistic, but it would also introduce un-needed nerfs to the health recovery sets in the game. These sets are already very niche, with their primary uses being restricted primarily to PvP, and even in their most favorable environments they are not overperforming. The nerf would not be warranted, and would obliterate any use these sets would see.

Solution 3:
2-4 pc bonuses to health recovery should be adjusted to a new value, and all other bonuses to health recovery should be centered around this value.

This solution would by far require the most amount of work, and also the most amount of balancing. In my opinion, its best to build off of the balance already found in the game and make minor adjustments, rather than to scrap it all and start over.

kijima

Even your TL;DR was TL;DR

Sergykid

kijima wrote: »

Even your TL;DR was TL;DR

that doesn't make sense. I think you missed a negation there

WrathOfInnos

Good post. I feel like there are several examples of inconsistencies like this. Seems like someone started a big standardization effort 1-2 years ago, got about halfway done, then gave up. And the last few updates have actually made negative progress, taking things that were standard and changing the numbers to nonsense.

Why are drinks universally 7% weaker than foods? The justification that Bright Throats and Bone Pirate sets require this is complete BS.

Why is Seducer set still 10% cost reduction, after Magnus Gift was buffed to 15%? How are Alteration Mastery and Battlefield Acrobat supposed to fit with these (and why were they untouched when the cost reduction formula was changed)?

Why was Grundwulf nerfed from a balanced 200 mag/s to a useless 140, in the same patch when Bloodthorn was buffed from a useless 132 mag/s to a balanced 192?

How are Max Magicka/Stamina/Health bonuses getting no adjustments despite losing the 20% CP bonus in Flames of Ambition? This just puts them even further behind every other set bonus/trait/mundus etc.

Why are recovery and Spell/Weapon Damage weighted 1:1 on all set bonuses and jewelry enchants, but mundus stones give 310 recovery compared to 238 Weapon/Spell Damage? And on a related note why are Apprentice/Warrior stones so weak? Given the latest (pretty good) balance pass on PTS 3% crit = 129 Weapon or Spell Dmg, so using Thief as a baseline these mundus stones should be at least 262 (probably still wouldn’t be enough to justify using them).

twing1_

WrathOfInnos wrote: »

Good post. I feel like there are several examples of inconsistencies like this. Seems like someone started a big standardization effort 1-2 years ago, got about halfway done, then gave up. And the last few updates have actually made negative progress, taking things that were standard and changing the numbers to nonsense.

Why are drinks universally 7% weaker than foods? The justification that Bright Throats and Bone Pirate sets require this is complete BS.

Why is Seducer set still 10% cost reduction, after Magnus Gift was buffed to 15%? How are Alteration Mastery and Battlefield Acrobat supposed to fit with these (and why were they untouched when the cost reduction formula was changed)?

Why was Grundwulf nerfed from a balanced 200 mag/s to a useless 140, in the same patch when Bloodthorn was buffed from a useless 132 mag/s to a balanced 192?

How are Max Magicka/Stamina/Health bonuses getting no adjustments despite losing the 20% CP bonus in Flames of Ambition? This just puts them even further behind every other set bonus/trait/mundus etc.

Why are recovery and Spell/Weapon Damage weighted 1:1 on all set bonuses and jewelry enchants, but mundus stones give 310 recovery compared to 238 Weapon/Spell Damage? And on a related note why are Apprentice/Warrior stones so weak? Given the latest (pretty good) balance pass on PTS 3% crit = 129 Weapon or Spell Dmg, so using Thief as a baseline these mundus stones should be at least 262 (probably still wouldn’t be enough to justify using them).

Yeah, I've noticed those too. I eventually want to tackle them all, but I fear that compiling them all into one mega post will be too big a fest for the forums. People are already saying this single post is too long, lol. Better off posting about them individually. I've got ideas for a food overhaul that I think would make a lot of sense.

You are right on all accounts, though. A consistency overhaul must be performed on so many aspects of the game. There is no reason all bonuses to stats from similar sources should not be valued equally.

Food of the same quality should grant an equal number of stat bonuses, regardless of whether they give max stats or recovery. Mundus stones should all grant the same number of Stat bonuses as other mundus stones, regardless of the types of stats they give. The same should apply to weapon/armor traits as well. Racial passives should be the same way too. 5 pc item sets are harder to balance because of their conditionals, but there should be greater standardization of these as well. Especially on the 5 pc bonuses that grant unconditional bonuses to character stats.

Once the coefficients for sources of stats are set, all that would need to be done is balancing the different types of stats against each other, like bumping up bonuses to max stats as you've suggested to make up for the loss of the 20% cp multiplier. When adjustments are made to the 2-4 pc bonuses, under this new system, it would be crazy easy to calculate the adjustments needed to maintain balance for other sources of that specific stat, just by using the coefficients that have already been set in place. It would allow for ZOS to tweak all the things they need to, without upsetting the balance of the game.

milllaurie

Another big thing about health recovery - sometimes it is considered a form of healing and sometimes it is not.
Minor/major defiles work on them (I think historically it was done because Troll king was too strong in pvp). But now that troll king is nerfed and we have a lot more sources of hp recovery (primarily the golden foods everyone is running now), new alessian, 7th legion, eternal vigor, you name it:
I think it is way past the time it should be affected by battle spirit as a source of healing.
Hp recovery is way too strong in pvp, especially when major Fortitude was buffed. For comparison - to get 40% more healing you would need to have the following buffs: major mending, major vitality, minor vitality. I think everyone here knows that this is impossible without actually speccing into healing.
For 40% hp regen you need: drink any potion with hp line on it.
Another reason hp recovery is overperforming in pvp: most of the times it is passive and does not require you to switch bars or actually lose a gcd.
Please, balance team, take it to consideration. Even with proc sets gone, I can guess who is running beekeeper + endurance backbar without any questioning.

twing1_

Just came across some more interesting findings, for anyone interested.

Regarding the set "eternal vigor", it seems that the 5 pc bonus to health recovery is also valued at the magic number of 387 health recovery for a single 2-4 pc bonus, if the stam/mag recovery is to be weighted as only a single bonus to recovery either mag or Stam, as other sets value other hybrid stats as only one or the other (1 pc domihaus granting both 1096 Stam and mag, for example).

337/129=2.61

1011/387=2.61

Because of the similar conditionals on each part of this sets 5 pc bonus, we can assume that the health recovery below 50% is valued at an identical level as the Stat recovery while above 50% health.

This seems to both support that the typical 2-4 pc bonus to health recovery should be valued 387 health recovery, and also justifies the seemingly overwhelming stat density of this set: the 337 stam and mag recovery is only being valued as 337 stam recovery or mag recovery, and not both. This is likely because the set is designed so it can be used by both stam and mag users, instead of making 2 separate sets for a stam and a mag variant. Interesting.

Skullstachio

You know something @twing1_, I’m surprised your not on the ZoS Combat team, because with the whole post you wrote regarding all of the recovery statistics and the inconsistencies you pointed out, you should join the combat team and help make elder scrolls online even better. (I would nominate you if I could.) because with each new update as pointed out by the community over the years, they seem to break something every time and I fear they may be losing a bit of their edge.

Fair case and point, the combat team in their entirety needs to take a break, come back fresh and reevaluate things in a Repair Month of sorts.

twing1_

Skullstachio wrote: »

You know something @twing1_, I’m surprised your not on the ZoS Combat team, because with the whole post you wrote regarding all of the recovery statistics and the inconsistencies you pointed out, you should join the combat team and help make elder scrolls online even better. (I would nominate you if I could.) because with each new update as pointed out by the community over the years, they seem to break something every time and I fear they may be losing a bit of their edge.

Fair case and point, the combat team in their entirety needs to take a break, come back fresh and reevaluate things in a Repair Month of sorts.

Yo, I'd be down lmao. I would take a sick amount of satisfaction from leveling out all the tiny little sources of inconsistency in this game, and I would be meticulous af about it.

I applied to be a class rep a year or so ago, but didn't get the gig.

Where my job offer at? lol.

vesselwiththepestle

twing1_ wrote: »

Yo, I'd be down lmao. I would take a sick amount of satisfaction from leveling out all the tiny little sources of inconsistency in this game, and I would be meticulous af about it.

You might think they just need to employ an Excel Sheet to do that for them.