Its april 2016 yet no healer stat weights???

[Ghishlain]

I do not have a source, I just know from experience with 700ish crit that I often hit for 1.6x modifier rather than 1.5x. Though there is a fluctuation in healing (a standard deviation of 55hp/heal on my last tests of physick), so that may have skewed my experience and it's possible I am wrong.

I just looked up one of my logs and looked at Embrace (made up the most of my healing in the fight). There were a total of 43 casts and a 14% crit rate. My average crit hit = 2850 and my average regular hit = 1792 which equals 1.6. Granted there are a lot of factors which may also skew this data, so this test should be taken with a grain of salt. For example, I have no idea how much each are affected by fey illumination, convalescence and defiance.

I don't think it'd be too hard to test if push comes to shove - more time consuming than difficult. I might try to test it this week if I get time to do so.

Until then, I'd probably refrain from including that Crit potency bonus in any sort of healing theorycraft until we know for certain because the quote I gave you in my post does indicate that it would only affect damage dealt. I'm more inclined to believe S-E's patch notes than a random "It feels like this" (no offense).

[Kaurie]

Right but adjusting for a critical Medica isn't overly useful IMO. It literally means once in a blue moon I have a free GCD that just poofs. Its not efficient to randomly flip to cleric stance there for one GCD drop a dot and then flip back to AE healing mode. While you do save a slight bit of mana (assuming you catch that you crit and can skip the next AE... ASSUMING that you crit enough to properly skip it). Again rather than going through all of that for a few spastic extra criticals I would MUCH MUCH rather have a a tenth or two tenths of a second faster cast speed allowing me to better side cast, hit heals during tricky mechanics, and otherwise top people off early rather than latter.

All of those advantages FAR outweigh a few spastic critical. This thought process assumes one is considering spell speed vs critical.

To obtain a tenth of a second faster cast time, you need 2.5k Spell Speed. Again, it's less that CRIT is great wah wah, it's that Spell Speed and Determination suck - leaving Crit as the only viable choice.

Now if one were debating Det vs Crit... I would go back to the overheal issue... most crits = overheals which aren't useful. Now... one could ague how often does the X small percent matter for additional Det mechanically speaking. And that is a very healthy argument. None the less it would be brutally difficult to prove out how useful (or unuseful) extra det is, in which fights, at what HP thresholds under what conditions. Reguardless none of the fights are designed assuming that you get critical heals on any specific mechanic. They assume you hit some minimal healing threshhold to correct the damage taken, and everything past that is gravy. Personally since you don't control if and when people get randomly nipped, nibbled on or take other tid bits of damage they shouldn't, I prefer extra raw Det (even if its weighed less) over random criticals which sometimes are there when you need them, and sometimes aren't.

But any which way I readily admit stat weighting, preference, and personal playstyle can vary. Likewise ones exact raid composition, mechanics and expected damage can vary as well. So mileage for folks is most certainly going to vary some.

Statistically speaking, most crits would not be an overheal. It becomes even less if you are a decent healer. I always gear for CRIT and my overheal isn't too bad (though, it's elevated as I am always shielding people at full HP for upcoming hits).

I don't think it'd be too hard to test if push comes to shove - more time consuming than difficult. I might try to test it this week if I get time to do so.

Until then, I'd probably refrain from including that Crit potency bonus in any sort of healing theorycraft until we know for certain because the quote I gave you in my post does indicate that it would only affect damage dealt. I'm more inclined to believe S-E's patch notes than a random "It feels like this" (no offense).

None taken and makes perfect sense. In my previous calculations, i left out the severity increase for that reason specifically.

[DarkmoonVael]

I don't think it'd be too hard to test if push comes to shove - more time consuming than difficult. I might try to test it this week if I get time to do so.

Until then, I'd probably refrain from including that Crit potency bonus in any sort of healing theorycraft until we know for certain because the quote I gave you in my post does indicate that it would only affect damage dealt. I'm more inclined to believe S-E's patch notes than a random "It feels like this" (no offense).

To be honest, i would be more inclined to trust the feeling rather than the patch notes. Historically, they have had things lost in translation, full details omitted, and thing not clarified.

Also, when increasing crit on my SCH without changing ilvl, it did look like the crit adlos did increase in shield power. But, i didnt do any extensive testing.

I have just taken it that increase in crit increases frequency and magnitude of damage and healing spells.

.

Seems like the community has something to test and clarify for healers now

[Katlyna]

Spell Speed It takes a lot of this to become useful to reall be noticeable. I am at I think 600 SS with 2.47 CD or so. Like every 100 SS you get a -0.01 seconds off your GCD- Say you had 1050 SS, that's ~ 700 from the default which should give you 2.38 seconds GCD. That means every 36 casts you will get a free heal. In terms of saving people, no one should be that close to the brink of death that 0.07 seconds means clearing and not. Spell Speed does however affect your potency on your HoT. I think the effect of this is similar to DET, but as it also lowers you recast time this is more desirable. For my calculation, 100 SS = 0.01 shaved off a 2.5 GCD which means your added healing power is 0.01/2.5 + HoT potency bonus.

Your spell speed calculations look wrong. This is what I have stripping my Astros clothing off while under Diurnal:
864 - 2.19
821 - 2.20
802 - 2.21
765 - 2.22
703 - 2.25
605 - 2.28
569 - 2.29
542 - 2.30
515 - 2.31
506 - 2.31
473 - 2.33
464 - 2.33
418 - 2.35
372 - 2.37
354 - 2.37

[Kaurie]

Your spell speed calculations look wrong. This is what I have stripping my Astros clothing off while under Diurnal:
765 - 2.22
703 - 2.25
605 - 2.28
569 - 2.29
542 - 2.30
515 - 2.31
506 - 2.31
473 - 2.33
464 - 2.33
418 - 2.35
372 - 2.37
354 - 2.37

Is that your cast time or recast time? Healers have a shorter cast than recast time, which is important to take into account - Thanks for throwing the numbers though, I am at work and was operating from memory.

If GCD recast follows the same trend, it looks like 100 SS = 0.04 not 0.01 decrease. That would bump up the healing power to 4x what I said, so instead of 0.4% it's a 1.6% increase or 2.4% with HoTs. That's still 1/3 of CRIT though.

That also means it would take +625 (979 total) ss, not +2.5k ss to reach the 1/10th faster healing. Apologies for the error.

EDIT: Just realized you said while under Diurnal, with the 5% buff so that's with that I am guessing. Can you show the difference without the buff as well?

[Katlyna]

That is my recast time. As you mention its still less raw metric power than CRIT (of which I completely concur) but functionally its noticeable. Its small but when positioning and doing movement mechanics I notice the difference. Mind you I am slide casting my heals so I'm fairly sensitive to the time and positioning.

I will concur the mana savings of crit hadn't concerned to me before. But one has to properly react and cancel out uneeded followup heals after crits... which is relatively difficult depending on the content. Again shield crits = juicy as this becomes a non issue.

[StouterTaru]

Yes stats scale exponentially.
Which is why simulations and all subsequent calculations are done for specific gear points.
Basic math. It is called "Linearisation".

Assuming your "specific gear point" only has a static weapon damage and main attribute, the curves still exist as I described them. If your specific gear point includes fixed amounts of Det/SS/Crit and you only are going to vary one by a few points, then you have a pretty flat curve.

Two people of same job, with gear at the same ilvl, can have dramatically different secondary stats, and as a result have different marginal stat weights off by 20% or more from each other.

[Ghishlain]

Here's Dervy's blog testing SkS. If you assume SpS has a similar affect on spells that SkS has on Weaponskills, you're looking at around 26.5 SpS to reduce cast time by 0.01s. He also has the DoT calculation there too, which I would imagine has similar properties to SpS (don't quote me on that).

Also, take into account that Diurnal provides a multiplier to your speed, making SpS a less desirable stat as you need more SpS to get the same 0.01s reduction.


[EDIT]

To be honest, i would be more inclined to trust the feeling rather than the patch notes. Historically, they have had things lost in translation, full details omitted, and thing not clarified.

Also, when increasing crit on my SCH without changing ilvl, it did look like the crit adlos did increase in shield power. But, i didnt do any extensive testing.

I have just taken it that increase in crit increases frequency and magnitude of damage and healing spells.

.

Seems like the community has something to test and clarify for healers now

Well, two people confirming "the feel" is probably good enough reason to at least take a deeper look at the formula.

Anyone interested in doing the test, here would be my methodology.

1. Record your CRIT amount.
2. Cast Cure / Physick / Benefic on yourself A LOT (record all normal and Crit results)
3. Do the same test with a different crit level (preferably a lot higher or lower, in the magnitude of at least 150)

I would also recommend using a level 1 weapon if possible to reduce variance in the heal amount. As long as know min/max heal, we can get the average amount from the two datapoints. Also, as long as we know the "base heal", the "crit heal", and the crit value we can calculate the bonus multiplier and see if it lines up with Dervy's Crit formulas.

I'll see if I can pull this out sometime this week just to be sure as my curiosity has been raised as well.

[Granyala]

Spell Speed It takes a lot of this to become useful to reall be noticeable. I am at I think 600 SS with 2.47 CD or so. Like every 100 SS you get a -0.01 seconds off your GCD- Say you had 1050 SS, that's ~ 700 from the default which should give you 2.38 seconds GCD. That means every 36 casts you will get a free heal. In terms of saving people, no one should be that close to the brink of death that 0.07 seconds means clearing and not. Spell Speed does however affect your potency on your HoT. I think the effect of this is similar to DET, but as it also lowers you recast time this is more desirable. For my calculation, 100 SS = 0.01 shaved off a 2.5 GCD which means your added healing power is 0.01/2.5 + HoT potency bonus.

Added Healing Power (100 SS): 0.4% without HoTs, ~0.8% with HoTs

Critical Rate and Severity For every 200 Crit, you have a 5% chance to crit more and a 0.1x modifier. As the modifier is in question, I'll leave this out of the calculations.. The added healing power is then (Chance x Added Value) = (5% x 1.5) = 0.05 x 1.5 = 7.5% added to your healing power.

Added Healing Power (100 CRIT): 7.5%

Now, yes, crit is unpredictable, but the added healing power is 18.75x greater for CRIT than DET.

Your data is flawed:
Spell speed: reduces cast time by 0.01s for every 26.5SS (according to the data earlier in this thread).
So 100SS/26.5 ~ .0377 reduction of cast time.
(2.5-0.0377)/(2.5/100) = 98.5% of cast time. Roughly 1.5% faster cast times. A.k.a. 1.5% more throughput.
Add to that the HoT bonus which I cannot calculate right now but I think you'll easily end up in the 2%+ range.

Crit: 200 crit = 5% chance -> 5% * 0.5!! = 2.5%.
If it doesn't crit you would end up doing 1.0 instead of 1.5. Calculating each crit as an ADDITIONAL 1.5 is wrong.
A modifier of 1.6 would make that a 3% gain.

Sadly I do not have any data for DET.

Something seems off though. According to this napkin math, SS would have double the value compared to crit per stat point. So I do think source data is questionable.
DEVs aren't stupid, they will balance the secondary stats somehow.

Another thought: someone mentioned that 200(?) ACC would add 20% hit chance. Depending on our base hit chance this stat has the chance to outclass every other secondary by a whole order of magnitude, making it the best stat even if it does nothing for the healing. (not sure how important healer DPS is, as I do not raid high end)

I have to admit though: getting sub 5% performance increases for putting in Grade Vs in all of my slots is woefully ... underwhelming. :/

[Kaurie]

Snip/

Thanks for pointing out the issues Not sure if it was apparent, but I posted my calculations, so that people could point out errors as such. I'll adjust my original post.

TLDR the added healing power is 6.25x greater for CRIT than DET, and is 1.69x greater for CRIT than SS

CRIT 2.5% increase
SS 1.69% increase
DET 0.4% increase

On this note, the above is really for WHM/AST only. SCH the increase of CRIT is more substantial, as Spell Speed doesn't reduce the cast time of Eos/Selene AND critical hits from Eos/Selene proc a spell speed buff on you. This is coupled with the double shield applied with Adlo