Did I make a math error or miscount? Or you think the logs that I selected are simply not good representatives?

I think it should be fairly easy to make an automated program to parse this information from a combat log--though, having multiple druids might mess it up, because it looks like it doesn't show who is refreshing the Lifebloom stack.

Also, if someone could post a link to a decent Sarth3D with a single resto druid, I'd like to take a look at the numbers that resto druids putting up that Blizzard has pinpointed as an example of problematic efficiency.

Patchwerk who only strikes one tank is not that ideal for a LB situation (and as it only hits one tank you have higher overheal figures) I would assume so you really need to try and base it off some Sarth3d logs to get the numbers in a fight more favorable to the spell.
Problem is that it is hard to compare HoT overhealing to normal spell overhealing due to the nature of the differences between them.

I've become somewhat resigned to the thought of the change going through and am just going to see how mana ends up to see if it is worth exploiting blooms and Nourish. Once we have a month or two of data we can see how things have panned out in the grand scheme of things and present some figures for Blizzard to mull over.

I don't have a Sarth 3D kill log, but here is a Sarth 2D log:
WoW Meter Online - Combatlog Replay in Engrish
Wow Web Stats

The failures are all the Sarth 3D attempts. Looking at those numbers that I'm putting out, I can understand why a nerf is due.

Well, it's not really about how much Lifebloom overheals, but rather how much it heals now and it's understandable they want to tone it down. However, yesterday at Thorim I was rolling Lifebloom on 2 tanks with the occasional Rejuvenation on them (whenever there were many adds) and some Wild Growth healing in between as well. I had to hold back a lot or I would run oom before Thorim comes down. I also ran oom on the trash before him, and that's with my pretty mana-oriented gear setup. Can't say it was fun to heal this way.

Here a S3D kill, except the log seems to be split up (it was a 1-shot, the "try" and the "kill" should be merged).

Wow Web Stats

My Sarth3D Stats, if can be of any use: http://wowwebstats.com/yndl4sn6652fw

(Resto druids here are me-Isideh- and Eldril)

Here are stats from one of our S3D tries.

Wow Web Stats

We have two resto druids. Raicho's style is similar to other stats posted here - 50% LB, 30% WG, 16% RJ and minor others. My style (Graze) is a bit less orthodox and LB usually accounts for ~20% of my healing. On this particular attempt I had 50% RG, 20% RJ, 20% LB, and 10% SM. This varies depending on what's going on but the 20% healing through LB is a fairly constant thing. We have fairly similar level of gear (mostly 25-man stuff with occasional 10-man pieces). He geared balancing for crit/haste - most of his gear is probably "best-in-slot" pieces, while I geared straight for haste wherever I could without sacrificing too much on regen/spellpower. We had similar assignments on the attempt I linked.

I understand that if people have 70% of healing done with LB they may have an issue if the changes come through, but based on the fact that my LB % is relatively low and I get similar numbers than other resto druids I raid with, I don't think that's going to be such a big deal. People will just have to shift to a different healing style a little. No more pure LB stacking. Blizz said they want to make healing more interesting. I'm not sure if hitting the main spell of majority of the population with a big nerf is the way to do it but that's more of a philosophical discussion. I think only certain people enjoy healing in the long run and I think that's more of a character trait than game mechanics and nothing that Blizz can do about it is ever going to change that.

Actual kill was 2D - we don't quite have it in 3D yet. So, essentially I didn't have nearly as much work as I would have otherwise and my numbers are way off on that.

I looked through other WWS parses I have and I have quite a few parses with relatively low LB numbers - naxx clears, malygos, other OS attempts - high LB numbers for me are more of an exception than a rule.

Yes we're kinda low on paladins right now - our holy ones especially are on hiatus - but otherwise we'd heal with one holy paly who'd usually be dedicated to sartharion tank. I'm usually healing the drake tank or the adds tank(s) + raid.

All I'm saying is that just because some mana numbers on LB get changed, doesn't mean that nobody will bring trees to raids anymore. It just means that the trees will have to use some other spells now and then. Also I don't see much point in maintaining 6 stacks of LB each healing for 1100ish a sec if I can rejuvenate 10+ people for 2k / 3secs. And I'm pretty sure we can still roll stacks of LBs on a tank or two. All it may mean is that some other classes will have to pick up some slack and I suppose that's the goal. I sometimes top meters with 5-10% over the next healer. Now, meters are not the most important thing - of course keeping people alive is. However, someone has to cover all those missing hit points. If I top meters with 20% overheal and the next person has 40% overheal, then it usually means that my hots cover a lot of missing health that the second person is trying to heal through as well. So, we're esssentially doing a lot of double-healing. Heck, in a typical Naxx farm clear, if you add overheal from even 5-6 healers then most likely it will come out over 100%.

Things like LB nerf and other changes seem to target this exact problem. I'm pretty sure Blizz wants to see overheal numbers go down. One way to do that is to make things hit harded. But that would mean that they would have to escalate stats on tank gear way more than one (or more like half) a tier. So instead they make healers work harder. I think that overall it's a good direction for the gameplay - same reason why +stam enchant on a weapon is a bad thing.

This officially lowers the number of points we need in Balance meaning we can do 11/0/60 because a 3 second period on Grace is going to be so lacking in a world where mana matters and we cannot spam those spells for any period of time without dire consequences. Time will tell over the course of the PTR if it remains worthwhile to pick up but quite honestly it doesn't seem like it will be for the time being.

I think Grace will still be useful in situations where we spam a single tank, like on Patchwerk PTR tests. No doubt that this change makes us even worse tank healers than before, because sometimes we are forced to keep up hots, and may miss the Grace window (and because as you said we can't spam as much). Given the Lifebloom nerfs, and given encounters we have seen so far on the PTR, it just seems likely that resto druids will be mostly raid healing. Their tools seem well suited to the damage patterns (Ignis, Hodir, etc.). I think this is unfortunate, I always thought of druids as generalist healers like priests, while paladins are tank specialists, and shamans are raid specialists. The way it is now, shamans are becoming better generalists and druids are pushed more towards raid healing.

The two good pieces of news is we have the points for Tranquil spirit now, which makes Nourish rather efficient with the glyph and the tier bonus, and that Blizzard apparently doesn't mind us using shifting out of tree as a root breaker (useful on Hodir, etc.), although I guess this may be more of a PvP change.

Healing Hodir last night was a similar experience, for both resto druids in my guild. We found ourselves playing extremely conservatively, keeping full hots on the main tank and taking every possible opportunity to cheat into the five second rule that we could. "Not fun" doesn't even come close to describing the current state of resto druids, "outright painful" would be closer.

Note: Hodir has only one tank to heal, and if your raid plays smart, very little raid damage. With this lifebloom I can't begin to imagine a fight that has raid damage plus 2-3 tanks to heal simultaneously.

I think the 25% crit reduction from the Change to Improved Regrowth (and the 25% increase to crit with Nourish) plus the addition of a Nourish Glyph would have ended the reign of Glyphed Regrowth as our Top tank healing spell anyway. This change to Nature's Grace more than likely put the last nail in the coffin. I'm actually okay with that. I believe the new buffed Glyphed Nourish stands to heal for more AND is a more flexible healing spell. Although I have yet to run the numbers, last I checked the debated between Glyphed Regrowth and 4T7 was fairly close with the advantage going to Regrowth, a swing of 50% crit (-25% to Regrowth, +25% to Nourish) likely puts them at even, or with an Advantage to Nourish.

For those people that want to analyze their own logs, I wrote a simple python script to calculate these metrics:

This code makes one deliberate error in favor of the player when calculating efficiency, for simplicity sake: it assumes every tick heals for the 3-tick value.


Just analyzing Ridley's stack on Bombistick (Wow Web Stats)

start: 36'30.137
finish: 39'43.310
casts: 20
tickval: 1417
ticks: 55
bloom: 5041
overheal: [4779, 1227, 1057, 946, 671, 783, 3, 746, 986, 1273, 474, 458, 550, 525, 478, 730, 691]

So, even in Sarth, the actual efficiency of Lifebloom is looking to be very low compared to the theoretical maximum. The actual HPM it is delivering in this instance is 9.1--is this high or low compared to other healing strategies? The main reason why Lifebloom is performing so poorly with respect to theoretical maximum is because the target is often at full HP. But, as you can see, this doesn't seem to be the case with Bombistick. Once you start having multiple people below full health, or a tank that isn't reliably topped off, the effective HPM of Lifebloom skyrockets.

One thing to keep in mind is that Blizzard may be implementing this change to avoid Lifebloom from being overpowered in future encounters that aren't yet in Live. So, we may not be able to independently come to the same conclusion, because they are working with numbers to which we don't have access.

Is anyone concerned with the current Val'anyr proc? Favors everyone except druids.

* Val'anyr Hammer of Ancient Kings - Equip Effect -- Your healing spells have a chance to cause Blessing of Ancient Kings for 15 seconds allowing your heals to shield the target absorbing damage equal to 15% of the amount healed.

I wouldn't say that, considering we can heal a lot of people really quickly. Damage absorbed is damage not needed to be healed, afterall.

Where did you read that?

I'm hoping it will be similar to Atiesh, in that there will be a different version for each type of healer class, and you pick the one most appropriate for you. Maybe that's just the holy paladin equip effect, or something.

Otherwise I'm a very sad druid.

Best case would be that hot ticks will give us the highest uptime of the buff* and then we can focus on using Nourish more when it procs. Even then I doubt it would be as good for us as it would be for say a Paladin or Shaman but we will see once more is known about it.

I think the following example is the strongest possible evidence that LB shouldn't be changed.

Following Jurik's (excellent) example, I took a look at the hpm for this log. I filtered the log and only recorded casts on the four tanks, since it seems that Blizzard is focused on our tank healing. NOTE: There are two druids rolling LB here, and I combined their numbers (since the log doesn't make it clear where a refresh comes from).

LB Ticks 405547
Tick OH 25489
Casts 160
Bloom Heal 131673
Bloom OH 106020
Eff Healing 405711
Eff HPM 6.928125
Eff HPS 911.7101124

So the two druids COMBINED to do 911 HPS at an efficiency of less than 7 HPM. This was on the hardest content while rolling LB on 3-4 tanks!

For comparisons sake, I calculated the Eff HPM of Regrowth on some of the above logs to be between 25 and 30.

[e] I was using Graze's log to calculate the Eff HPM for Regrowth. The chart seems to not count all casts, so I browsed the log to get an accurate count of number of casts.

I think the killing stroke to Blizzard's reasoning for the nerf would be to calculate effective HPS and HPM for all healing done by a druid on the Sarth 3D encounter. Then you could compare that to logs of other classes on the same fight and show that druids aren't pulling ahead because of lifebloom rolling on multiple tanks. In fact, with this new data it appears that lifebloom is far less powerful and it could be argued that lifebloom is even now underpowered, though I won't because lifebloom does its job in cushioning spike damage.

We should have thought of doing it this way long, long ago. Now we can see past the silence of the combat log to see just how "amazing" lifebloom really is. It even makes me wonder how good it really was back in BC.

This just in: heals usually cause overheal. Lifebloom causes less overheal than other spells - and not just because some of it isn't registered. It's because it ticks every second for a small amount. So don't run off comparing the effective HPS of LB on a single WWS to say the theoretical HPS of holy light. It won't end well.

Of course not. I don't propose comparing apples to oranges here. Effective vs. effective. Then again, Blizzard will probably say that it hasn't become overpowered in 3D Sarth because of the lack of regen nerfs, so I suppose that it probably wouldn't be the "killing stroke". Perhaps have a druid with great gear heal with the other healers being undergeared? Probably not.

Thank you! My long-standing feeling was that in the current edition lifebloom is neither the most efficient nor most effective way to heal. It's a necessity if nothing else - I keep lifebloom rolling on the tank not because of it's HPS value (which is mostly ZERO because in a lot of encounters the tank is topped off 98% of the time and there are fractions of seconds between when he takes damage and the next tank heal lands). I keep lifebloom rolling on the tank because it's a backup heal - if the tank runs OOR or the healers get silenced, or the healers have to move and there is a brief time there is no healing going on, lifebloom will keep on ticking. In certain fights lifebloom is nearly 100% overheal except for those few precious moments when it becomes lifesaving blessing. Regrowth/nourish/healing touch are much easier to micromanage in terms of their effective HPM because they are cast when the damage already happened. If I want to give a heal buffer to a large group of people, I'll use rejuvenation. If I want to quickly recover a group of people, I'll use wild growth. If I want to help one particular person, I'll cast regrowth or nourish - depending on whether I want a HoT on them or not. And that's the final point: when you cast 1 heal / second, there is no time to worry about HPM - you use spells for their functionality rather than their stats.

Of course it's still an issue when trees can outheal other classes in certain encounters by a factor of 2:1, but I believe this has nothing to do with our "perceived efficiency", and much more to do with the fact that we can do a lot of healing while moving with our instant HoTs - Sartharion 3D is a busy fight with a lot of movement - classes heavy in heals with cast time are automatically gimped just because of that - that fight is naturally tree-friendly, as are a lot of other fights for that matter - Blizzard makes a lot of fights difficult by adding heavy movement. Take logs from Patchwerk and the numbers for different healing classes are nowhere near as wide-spread - in that case it's more of a luck in terms of whose heal landed first when the tank gets hit. If you want a fight with lots of raid damage, take Sapphiron - here druids start to shine as well, but other classes with exception of paladins do not too shabby either - numbers in my group are usually depend on particular assignment rather than class. In my opinion, if Blizzard wants to push the "bring the person not the class" idea, they should help priests, shamans, and paladins in busy fights rather than fight with efficiency of trees. If the other classes do better, tree numbers will go down - and then finally maybe we will be able to reliably lower the number of healers necessary in raids.

I'm somewhat expecting a nerf to Rejuvenations efficiency to come in before 3.1 goes live but perhaps they will change their mind on Lifebloom to compensate it.

RJ has always been considered the most efficient heal and as Priests get reigned in on mana with us it will end up seeming so much stronger and yet Blizzard have stated they are fine with it. I highly doubt their theoretical encounter where we roll multiple LBs on several tanks taking very optimal damage will be able to compare to the amazingly disposable RJs that are able to be thrown around so carelessly against the inevitable mass raid damage which prevails on the majority of actual encounters.