Get the Desktop App for Battle.net Now
- All your games in 1 place
- Log in once
- Automatic game updates
It takes 1680 haste to get a clean HS+2HL or DL+1 GCD. That's a lot of haste and may not be for everyone. The haste plateaus are technically jumps in benefit, but if you're a healer and sticking to strict rotations, you're doing it wrong.
If you're doing funny things like canceling midway just for the sake of hitting HS on CD, you're doing it wrong. If someone needs a heal, they get it. If someone doesn't need a heal. They don't. Whether your HS is on or off CD is secondary.
I only recommend a 2 second cast haste because
1) People like clean, cut and dry numbers. It may not be best for you, but it does work for many.
2) Most bosses seem like they have a 2 second swing timer. It's hard to say for sure because of latency in logging. Letting the boss swing 2x between your heals may screw you over. It probably won't in Cataclysm though.
3) It adds another tick to HR. Nice, but not that big of a deal.
4) It's a realistic amount of haste to get.
The general guideline is and always will be:
Are people dieing because your casts are too slow? Yes - get more haste. No - you're good on haste.
Merry Christmas Holy Paladins!
Anyways for Christmas, I bring to you the my first look at holy mastery in Cataclysm.
This is data taken my PuG Heroic Vortex Pinnacle. I don't have raid data yet.
Here's what won't proc mastery:
Protector of the Innocent
Healing from Seal of Insight
For my spreadsheet, please see: https://spreadsheets.google.com/ccc?key=0ArLmeVrNbnLxdExWd3hHb1d4ZmcwSUFiVTJXN25NMWc&hl=en#gid=0
I make one very important assumption. Shields are either completely consumed, or completely wasted. I do NOT account for partial consumption of shields. This is significant because we start with 10% base shielding. For example, if we have 16% shields, but only the first 10% (10% relative to the total heal, not 10% of the shield) are used, then the last 6% due to rating are completely useless. However, because damage is very high relative to shield sizes, I'm going to assume that shields are always completely consumed when a shielded target takes damage. Without this assumption, mastery is devalued.
So including overheal, 48.47% of my TOTAL (not effective) healing allows for mastery to proc. This ratio is likely to be lower in raids because Holy Radiance is likely to be a bigger chunk of your healing. Your playstyle is likely to change this ratio as well.
Next up, if all of my shields were used, I would expect to see 555534 shields used. In reality, I saw 337760 shielding used. So that means, of the shields that occur, I expect to see 61% of them used.
So while the tooltips suggest 179.2 Mastery Rating will produce 1.25% extra shielding, this data instead predicts it will do ~0.375% extra shielding (0.5 healing that allows shield * 0.6 ratio of used shields).
This suggests that crit is better than mastery for average boost (due crit being RNG) to HPS and HPM. This however, does not account for additional benefits that crit provides in the form of Conviction and IoL procs.
DO NOT READ THIS AS MASTERY IS USELESS. Because Mastery and Crit stack multiplicatively, there is an ideal ratio to keep the 2 at. I'll get into that later after I confirm these findings in a raid setting.
TL/DR: Mastery operates at 30% of the tooltip effectiveness.
Edited by Lylthe on 12/27/2010 2:59 PM PST
I've been following this thread from the beginning. I tried to get the haste to 1680. That number is not unreachable, right now I'm at 421, but I also don't want to sacrifice much of spirit and int (Earthen Ring items have lots of crit on it). Since I have archive a closed number to 1680, would you recommended me to swap gems and other enchant to get to the magic 1680?
My experience in heroic dungeon (of course I'm a bit over geared) is that with group group, my mana is always at 80%, but with bad group, I found myself to run OOM pretty fast. I cast a lot of DL (I know you might tell me that I should cast HL instead) but because they take so much dmg. I hope that by stacking haste, I'll feel more comfortable to cast HL more than DL. Right now I'd say 50/50
Matuzak, just came back to wow that's why I didn't know plate matters that much to our INT. I just sold the plate boot to vendor. Don't know what to do now lol, but thanks for letting me know. Hopefully Blizz is nice enough to restore it back to me.
1680 is a viable amount of haste to shoot for provided you're in mostly T11. Not everyone is yet. And I've met paladins in T11 that are doing perfectly fine without haste. I've also seen paladins that are perfectly successful stacking the wazoo out of haste. So I'm hesitant to say, "You need <XXXX> amount of haste or you fail!" because it's not true.
My guild is basically on hold for the holidays, so if any raiding paladins want to help out with the analysis on Mastery, I would love some Recount/Skada numbers that include:
1) Boss fight and duration.
2) Breakdown of effective healing (including shields). Total Effective healing done, and amount each spell provided.
3) Breakdown of Overhealing done: Total Overhealing and amount each source provided.
I'd like to determine:
- What ratio of our healing in a raid environment will proc shields?
- What ratio of procced shields are actually consumed?
When we have these 2 numbers, we can definitely say, how good or bad mastery rating is.
I don't know what addon to use to track this best... any suggestions? If Recount works I must have it set up wrong because my absorbs tab shows nothing at all. :|
I would love to contribute and know for certain if I really am hindering myself by not going full crit.
Even so, I know that having a ton of Mastery and having counterpart Paladins that avoid Mastery means our healing works better together (less overheal, better tank healing). Seems like a good enough reason to keep it for now anyway, regardless. :D
Assuming that mastery works for 50% of our healing, and 60% of our shields are consumed, then Mastery works @ 30% effectiveness.
So the Effective Healing bonus of Mastering Rating is:
MR = Mastery Rating =
Effective Healing Bonus = (MR/17920)*1.25*0.3 = 0.00002093*MR
CR = Crit Rating
Effective Healing Bonus= (CR/17920)*0.5 = 0.0000279 *CR
Total Bonus "Healing" =
F(CR, MR) =
(1.0000279*CR [Due to crit] ) * (1 + .1 [Base shields] + 0.00002093*MR [Mastery Rating Shields] ) =
1.10003069*CR + 1.000048830583947*CR*MR
If assume we have a set Item Budget, IB, such that:
CR = K*IB
MR = (1-K)*IB
Where K is the ratio of the Item Budget that crit rating will use:
F(CR, MR) =1.10003069*K*IB + 1.000048830583947*K*IB*(1-K)*IB
=1.10003069*K*IB + 1.000048830583947*K*IB^2 - 1.000048830583947*K^2*IB^2
Back to Calculus:
Partial derivative with respect to K
1.10003069*IB + 1.000048830583947*IB^2 - 2.000097661167894*IB^2*K
Find Local Maximum
K = (1.10003069*IB + 1.000048830583947*IB^2) / (2.000097661167894*IB^2)
So what does this mean?
At low item budgets, K is a little bit bigger than 0.5, so you want to have a little bit more crit than mastery. At high item budgets, K approaches 0.5, so keep your crit rating equal to your mastery rating.
We're going to want to double check this math since it's contrary to what many people believe.
Front page has been updated as well.
Haha, thanks for the bumps. :)
Edited by Lylthe on 12/27/2010 12:59 PM PST
A couple important caveats about the crit:mastery ratio.
I do NOT consider IoL procs. I do NOT consider conviction (assumes you have enough crit to keep conviction rolling already). I'm also ignoring overheal. This doesn't take into account RNG variance.
This DOES take into consideration that not all heals will proc shields, and it DOES take into consideration that some shields are just wasted.
I also may need to reconsider the 30% crit required to keep Conviction up. WoL reports are showing uptimes much higher than expected, so I'll probably be forced to write a Monte Carlo sim.
Edited by Lylthe on 12/27/2010 12:58 PM PST
I was kind of a big fan of Mastery on 4.0.1. I saw Shields accounting for some 20% of my overall healing. In current raids, and I have downed every normal mode boss atm, the shields are performing rather poorly. I am currently reforging out of mastery and crit and into haste and spirit.
I see no point theorycrafting much for heroics. You just need to get the best gear possible and run with people you know and healing 5 mans is /yawn easy. Speaking in the context of current raid content, it is worth focusing on what works best. You do need sustainability.
Haste is a good stat because it helps a lot of our spells but in particularly it boosts the effectiveness if Holy Light, the spell we're casting the most in raid content, and Holy Radiance, easily the most wonderful spell in our arsenal. With enough spirit, casting Holy Light is essentially free. When you get the cast time down enough, you see a significant percentage of your overall healing from Holy Light. Technically, my Divine Light out performs my HL, as it should with the mana cost, but overall, they are competitive. Through most phases of damage, I can fill people up with nothing but Holy Light/shocks/Wogs, with Holy Radiance (and Light of Dawn to some extent) taking care of AoE burst healing.
The more spirit you have, the more you can sustain casting larger heals.
Don't get me wrong, crit and mastery are good stats, especially Crit in relation to Holy Radiance. But you NEED haste and spirit. Intellect is still and always will be king. I'd sacrifice any secondary stat, within reason, to gain more Int. But Haste and Spirit are very important.
Aani, I'm well aware of everything you've said. All of it is quite true. But the whole point of my last few posts wasn't about how much haste and spirit we need. That's been settled. We need enough to avoid reflex fail, and we need enough to avoid OOMing.
These last few posts have been, after all that's settled, how should you balance the dump stats, crit and mastery? That's what I've been mathing, and that depends heavily upon, how much of our healing will proc mastery using "typical" spell selection? And when it's been procced, how much shielding is actually used?
And while I would love to have raid data, I don't have any, and I can't analyze anyone else's data, because World of Logs doesn't track mastery. So I got the next best thing, heroics.
So while there's nothing wrong, I do want to keep discussion focused on collecting data and review of the math, because it's something that needs to be done and I haven't seen anything on EJ or anywhere else trying to figure it out.
50% of healing allows for mastery
60% of shields are consumed
Ideal 1:1 crit rating to mastery rating ratio derivation
Wowmetersonline is tracking Illuminated Healing! WOOHOO!
This guy posted some logs (though his focus is considerably different from mine, I'm not complaining)
So anyways, I dug around some more. This is based on an EU 25 Nefarian kill (no one can claim these guys are bad, nor can they claim this is an inapplicable situation.)
I can't type this paladin's name, I'll call him "Oenbop" as the closest Englishization, but the important thing is, he's logged out in holy gear and has NOT gotten any gear upgrades since the kill (thank you Armory stalking).
Oenbop has 9.33 Mastery, or 11.6625% shields.
The logging's not perfect, especially total shielding. I'm expecting 377846 Total Shields, but the combat log is showing 341600 total shields. So the log likely has some errors, but I'm going to just trust it for the sake of analysis.
I am going to trust that 235,786 of effective shielding was done. If all shields were absorbed fully, that means rating shielded 33612 damage.
Total Healing: 7329364
Total Mastery Eligible Healing: 3238835
44.2% Total Healing is Eligible for Mastery. My heroic data predicted this to be 48.47%.
The consumed shield percentage was either: 69.02% base on the logs
or 62.40% base on my predictions.
So the old ballpark heroic predictions of 50% eligible healing, and 60% shield consumption was close, but not quite right.
For simple math, 50% and 60% are nice easy numbers, for more accurate, I'd recommend 45% and 65%. However, in the end, mastery is still operationing at ~30% of the tooltip listed effectiveness. So my old math is probably still good (provided my method of analysis is good).
Threats of violence. We take these seriously and will alert the proper authorities.
Posts containing personal information about other players. This includes physical addresses, e-mail addresses, phone numbers, and inappropriate photos and/or videos.
Harassing or discriminatory language. This will not be tolerated.