[7.2] Legion Holy Priest Discussion

[Djriff]

Yeah I've found heal to be way to slow, even when I'm trying to save mana some other will come snipe it because no one ever pays attention to incoming heals/they have more haste than I do.

That aside, it's not like I'm running into mana issues most fights, so spamming the expensive heals isn't all that punishing.

[Mazi]

I've come across a handful of trinkets over the past few days; not particularly sure which ones are best:

-850 Alch Trinket
-850 Promises deck
-840 Heightened Senses
-850 Vial of Nightmare Fog

I'm assuming Alch trinket + Vial for throughput, and Alch+Promises if more mana is needed?

Promises > Alch by a long shot for mana.

Promises + Alch if you need a lot of mana (ursoc?)
Promises + Vial otherwise

Heightened Senses isn't very good, especially with Vers on it

- - - Updated - - -

Yeah I've found heal to be way to slow, even when I'm trying to save mana some other will come snipe it because no one ever pays attention to incoming heals/they have more haste than I do.

That aside, it's not like I'm running into mana issues most fights, so spamming the expensive heals isn't all that punishing.

Pretty much any time I think I would cast heal, I cast smite instead :P

[VigilantRose]

I notice that benediction is popular even without a binding heal build , enlightment is popular aswell , it makes some sense tbh , apotheosis is a great cd but in a lot of fights you dont really know when its the perfect moment to pop it , benediction is a sustained strong heal and can be used on cd , i also have 2 strong relics that support pom and they are wasted if i dont use it enough , i think i will try benediction and pom spam today.

For Mythic 5-man, I'll generally get about 10%-15% of my healing from Prayer of Mending/Renew/Holy Mending (legendary legs, no Renew or Divine Hymn artifact trait). I almost never hardcast Renew and I don't run Enduring Renewal. In terms of cast-time efficiency, this places it a bit below Holy Word: Serenity and about equal to Holy Word: Sanctify. In terms of mana efficiency, Prayer of Mending is my best spell.

This is also taking into account that the effectiveness of Benediction ramps up dramatically the more intense the healing becomes. If we're just walking down a hall with easy fights and simple tank damage, it isn't very good. In some boss fights, it's almost constantly pinging.

[Vampiregenesis]

Hey guys , is haste much worse than crit? what would you pref between littlehaste/alotmastery vs alotcrit/littlemastery? also is crit much worse than mastery? i have some items with alotofcrit/little mastery and i wonder if i should try to replace alot.

[Danner]

Hey guys , is haste much worse than crit? what would you pref between littlehaste/alotmastery vs alotcrit/littlemastery? also is crit much worse than mastery? i have some items with alotofcrit/little mastery and i wonder if i should try to replace alot.

Depends on how you look at it. I'd give you actual numbers, but I'm not in front of the game right now. So this is all... not backed by numbers. But the intent should be accurate.

Key point:
A given item of a given item level have a fixed amount of rating points on it. Those rating points can be crit, mastery, haste or versatility, or a combination of those.


Versatility
1% versatility is 1% extra throughput.
This stat is free extra healing. Pure extra throughput. There is literally nothing wrong with that.
The only problem of versatility is the opportunity cost: To get 1% versatility you need ot pay some 400 or so versatility rating.
Had those points been say crit rating instead, you could have gotten MORE than 1% crit rating.
This is the reason versatility is often ranked lowest in terms of secondary stats. You get less throughput per rating, than you could have gotten from other stats.

There is however a synergy between stats (all your crits, mastery etc get stronger with versatility), meaning there is a point where you want some but not much versatility, and versatility is your best itemization.

Versatility also helps you survive a little more, which is useful, but... if you need that extra survival, you're probably doing things a bit wrong. I'm not doing deep end mythics, so maybe that survival is actually useful there? I cannot say.

Mastery
1% mastery is 1% extra throughput. Over time.

Mastery is free extra healing. Compared to versatility, it's slower, as it heals over time - and that often means it is overhealing instead of actually healing.
It's also much less rating to get that 1% mastery, meaning it's a far better stat than versatility.
The major downside of Mastery is that there are spells (any healing over time spell, like renew) that simply is not affected by mastery.
If you are doing nothing but spamming renew, mastery is by far your worst stat (only the direct heal component of renew is subject to our mastery).
If you don't, then mastery is pretty darned good!

Crit
1% crit is 1% extra throughput.

Crit is free extra healing. It is instant, but - it's unreliable. Sometimes it triggers, sometimes it does not. This makes it a conceptually bad stat for healers, as we want reliability onwhat we do. And those crits as a result tend to be overhealing a lot. In practice however, crit is pretty good this time around because everyone tend to be rather damaged a lot of the time, and those crits make the major difference between a tank slowly dying while your flash heal spam being unable to keep him up - and topping off the tank so you can spare a heal on someone else. I personally love this stat to death mainly because it also helps me smite those mobs in world quests harder. Mastery simply does not.

Ratingwise, it's slightly more expensive than mastery if I remember it right. You spent more crit rating to get 1% crit than you would spend mastery rating to get 1% mastery.

Haste
1% haste is 1% extra throughput.

Haste is NOT free extra healing. Instead, it allows you to cast more spells in a timeframe. If you do, you get 1% more throughput for 1% extra haste. But this means your mana consumption also goes up. Which is kinda bad if you're tight on mana. And make no mistake - haste stacking will make you tight on mana. I mean, it's not like you MUST chaincast all the fricking time, in which case haste just makes your spells land faster (and save more lives) - but we all know that is going to be the result.

Ratingwise, haste is slightly cheaper than crit and slightly more expensive than mastery if my memory serves me right.

--

TLDR:
- Worry about increasing your item level. More stats > right stats. Also, more ilvls -> more intellect, which is the king of all stats.
- Once you get to the item level ceiling, then item specialization matters more.
- The icy veins holypriest guide recommends mastery->crit->haste->versatility as the priority order. This is based on expert opinions.
- Ask Mr Robot recommends crit->mastery->versatility->haste as the priority order. That is based on simulation results.
- I keep a little of everything for the synergy, but I mainly focus on crit->mastery/haste/versatility (really don't care which) this time around.
- That's my personal choice though, and you should look at your playstyle and preferences and see what suits you.

[Vampiregenesis]

Nice ty interesting post , so crit can actually be even better than mastery and haste is pretty much at the level of versatility , good to know.

[Nephthysis]

Depends on how you look at it. I'd give you actual numbers, but I'm not in front of the game right now. So this is all... not backed by numbers. But the intent should be accurate.


Versatility
1% versatility is 1% extra throughput.
This stat is free extra healing. Pure extra throughput. There is literally nothing wrong with that.
The only problem of versatility is the opportunity cost: To get 1% versatility you need ot pay some 400 or so versatility rating.
Had those points been say crit rating instead, you could have gotten MORE than 1% crit rating.
This is the reason versatility is often ranked lowest in terms of secondary stats. You get less throughput per rating, than you could have gotten from other stats.

There is however a synergy between stats (all your crits, mastery etc get stronger with versatility), meaning there is a point where you want some but not much versatility, and versatility is your best itemization.

Versatility also helps you survive a little more, which is useful, but... if you need that extra survival, you're probably doing things a bit wrong. I'm not doing deep end mythics, so maybe that survival is actually useful there? I cannot say.

In the era of Artifact Weapons, it's also important to remember that Versatility is an additive bonus to the base value of your heals. It does not synergize with other sources of +% healing like Follower of the Light and Reverence. (I only had one belt to test this with but it still appears to be the case.) Whereas 30% crit makes all those artifact traits 30% stronger because its a multiplicative bonus at the end of the formula.

Plus, as Danner said, it just takes too much Versatility rating to get a decent % out of it.

[Noradin]

In the era of Artifact Weapons, it's also important to remember that Versatility is an additive bonus to the base value of your heals. It does not synergize with other sources of +% healing like Follower of the Light and Reverence. (I only had one belt to test this with but it still appears to be the case.) Whereas 30% crit makes all those artifact traits 30% stronger because its a multiplicative bonus at the end of the formula.

Almost all direct healing inceases are now additive.

[Nephthysis]

Those two sentences contradict themselves.

How are they contradictory? Versatility and Follower of the Light have zero impact on each other.

Almost all direct healing inceases are now additive.

Yes, except Crit, which was the point. (Also, +% healing received buffs if anyone cares.)

[Noradin]

How are they contradictory? Versatility and Follower of the Light have zero impact on each other.




Yes, except Crit, which was the point. (Also, +% healing received buffs if anyone cares.)

That first part of the post should not exists, I did edit it not add to it.

Crit is not a direct healing increase, it does not change the base value like artefact traits and talents (and mastery in the case of some classes/specs) do.

[cruumash]

Is 1% crit really the same as 1% haste though? If we examine the absurd, 100% crit means every spell is always doubled in output. At 50% haste, you could cast every spell twice in the same amount of time (for twice the mana). But I don't think the GCD will let you get to that point whereas crit has no caps.

I'm sure the modelers can break it down better, but it seems to me that past a certain point, crit maintains its value while haste adds nothing.

The weight of haste vs versatility is the one that seems the most interesting when making gear choices because I think haste loses to crit or mastery hands down.

[VigilantRose]

Notes on stats:

• Various stats are effectively 'capped' - haste at 50%, critical at 100%, etc. This is not generally relevant since you won't reach the caps anyway.
• Despite be frequently modeled as such, the stats are not actually the same. Haste does not increase mana efficiency except with HoT (where it is abnormally potent since it both reduces casting time and increases healing performed). Critical tends to work best with small spells and is often hampered by various talents/traits that abnormally increase critical - the payoff for Serenity is almost non-existent with critical, for example, due to the conjunction of its enormous size and the artifact critical boosting trait.
• The more you have of a stat, the less valuable it becomes. This is merely an artifact of how the math works. Essentially, you have a stat priority where a certain stat is best up to a certain point where it becomes equal to the next best stat - and then those two stats advance in lockstep until you reach the point where they're equal to the third best stat, etc. Because you start with enormous amounts of critical and mastery compared to haste and versatility, this tends to mean that specs emphasizing those two stats shift over at earlier levels than specs that start with haste/versatility as ideal.
• Damage can actually matter. On a Discipline Priest, I wouldn't normally prioritize Mastery even if it were better than Haste/Critical/Versatility simply because the slight benefit in healing is offset by the massive loss in damage. On a Holy Priest, this is less of an issue but still worth taking into account.
• Damage reduction can actually matter. If you're in a raid, Versatility is normally a weak stat because a healer only takes 1/20th (or less) of the total damage but provides 1/5th (or more) of the total healing. However, I'd argue that if your goal is Mythic+ 5-mans, Versatility might well be your preferred stat because it allows you to potentially survive in situations where you otherwise wouldn't.

In terms of Holy Priests specifically, I would probably prioritize Mastery/Critical for raiding and Versatility/Haste for Mythic 5-man. Mastery/Critical efficiently add to your throughput. Haste maximizes your throughput in situations where mana isn't a concern and Versatility boosts your survivability.

[Danner]

So, I figured I'd give the hard numbers as well.

---

Rating costs (taken from the game, basically just divide your rating by the % gain at your char sheet):
1% versatility costs 400 versatility rating
1% Crit costs 350 Crit rating
1% haste costs 350 Haste rating
1% mastery costs 280 Mastery rating

As you can see, mastery is significantly cheaper than the other stats. In terms of raw healing throughput, it's a pretty awesome stat. As long as you don't use renew.
Haste and Crit actually give the same stat gain per point. I thought haste was slightly cheaper than crit, consider me corrected!

---

The current Ask Mr Robot stat weights for holy priests is:

Intellect 17.76
Critical Strike 10.71
Mastery 10.19
Versatility 10.11
Haste 4.55
Leech 4.14

These numbers should be interpreted as one rating point of critical strike gives you roughly 10.71 extra HPS.
Now, this is not a linear scale! Because these rating points have synergies, YOUR specific stat ratings may vary from these numbers - basically, you should aim for a ratio of stats matching the stat weights!

However, these stat weights also depends on your rotation and encounter. So treat these numbers as a guideline rather than a rule. But as you can see, haste is really getting the shaft this time around, according to Ask Mr Robot - while the other three stats are somewhat evenly matched.

--

There is an elephant in the Ask Mr Robot stat weights. Haste.
Is haste really that bad? I am not really convinced. The biggest issue with haste is the mana cost factor. Secondary, it doesn't really do much for our cooldowns. And Holypriest is currently a very cooldown-based spec. But less than half as good in terms of throughput? I fail to really see why that should be the case.

[Vampiregenesis]

So, I figured I'd give the hard numbers as well.

---

Rating costs (taken from the game, basically just divide your rating by the % gain at your char sheet):
1% versatility costs 400 versatility rating
1% Crit costs 350 Crit rating
1% haste costs 350 Haste rating
1% mastery costs 280 Mastery rating

As you can see, mastery is significantly cheaper than the other stats. In terms of raw healing throughput, it's a pretty awesome stat. As long as you don't use renew.
Haste and Crit actually give the same stat gain per point. I thought haste was slightly cheaper than crit, consider me corrected!

---

The current Ask Mr Robot stat weights for holy priests is:

Intellect 17.76
Critical Strike 10.71
Mastery 10.19
Versatility 10.11
Haste 4.55
Leech 4.14

These numbers should be interpreted as one rating point of critical strike gives you roughly 10.71 extra HPS.
Now, this is not a linear scale! Because these rating points have synergies, YOUR specific stat ratings may vary from these numbers - basically, you should aim for a ratio of stats matching the stat weights!

However, these stat weights also depends on your rotation and encounter. So treat these numbers as a guideline rather than a rule. But as you can see, haste is really getting the shaft this time around, according to Ask Mr Robot - while the other three stats are somewhat evenly matched.

--

There is an elephant in the Ask Mr Robot stat weights. Haste.
Is haste really that bad? I am not really convinced. The biggest issue with haste is the mana cost factor. Secondary, it doesn't really do much for our cooldowns. And Holypriest is currently a very cooldown-based spec. But less than half as good in terms of throughput? I fail to really see why that should be the case.

I was looking exactly for something like this , now i feel better with my versatility items like horn of valor , it felt bad this week to always loot 860-865 with vers on them but not anymore. I cant believe haste is so bad.

[VigilantRose]

Rating costs (taken from the game, basically just divide your rating by the % gain at your char sheet):
1% versatility costs 400 versatility rating
1% Crit costs 350 Crit rating
1% haste costs 350 Haste rating
1% mastery costs 280 Mastery rating

I get 325 Haste rating for 1%.

Now, this is not a linear scale! Because these rating points have synergies, YOUR specific stat ratings may vary from these numbers - basically, you should aim for a ratio of stats matching the stat weights!

It's not a ratio and there are no significant synergies (for Holy Priests). It's a priority list up to a certain point. A visual way to understand the math is to consider a rectangle with X and width Y, where your goal is to maximize the total area of the rectangle by incrementally adding to either X or Y. Without going into all the math, many people should recognize that the point of maximum area is where you have a square - where X = Y.

WoW stats interact in the same way - it's just that you've got 4 dimensions instead of two and some linear scaling factors (add the same amount to Critical is not the same as adding it to Mastery). You're looking for the total impact from each stat to be equivalent, accounting for the costs of that stat. Because the cost is linearly scaled, this creates a situation where (depending on what you value), one stat is simply better than the others until a certain value. At that point, it's tied with the next best stat - you invest equally in both. Eventually, those two stats become tied with the third stat and you invest in all three equally. Most of the time, you'll never get beyond your first two stats. If your best stat starts at zero, you may never even reach the second stat.

So your 'optimal point' is actually with certain stats leading other stats by a fixed amount. For a given spec, you might want 15% more Critical than Mastery - and you'd want 15% more Critical than Mastery whether your Critical was 15% or 500% (well, not 500% given the hard cap of critical, but you get the picture).

However, these stat weights also depends on your rotation and encounter. So treat these numbers as a guideline rather than a rule. But as you can see, haste is really getting the shaft this time around, according to Ask Mr Robot - while the other three stats are somewhat evenly matched.

Traditionally, people geared for raids on the premise that raid gear trivialized everything else (PvE-wise). So while your top notch raid gear might not be exactly what you need to go kill mobs solo in the world, the sheer iLvl advantage was sufficient to offset any secondary stat imbalances. Essentially, you gear for the hardest thing you do. I'm not sure this is still true.

What I'm coming to believe is that, for most players, Mythic+ 5-man - not Mythic raiding - is the 'hardest thing they do'. The number of players in the hunt for world first on raids is incredibly small compared to the number of players who are just running non-raid content with some buddies. It's far easier to find 5 competent players who can play at similar times than it is to find 20 competent players alongside a competent raid leader to keep things organized - and if you're like the 99% of the player base without a preset group of 20 to raid with, it's more effective to assemble a 5-man team than simply try to pug your way through Mythic raiding. The consequences for failing your 4 buddies in a Mythic+ 5-man are also more severe than failing the 19 random strangers you're pugging EN with.

So while 'gear like a top-end raider' might have made sense in the past, I don't believe it's particularly good advice in Legion because almost no one is trying to be a top-end raider (and those that are probably keep their own counsel).

Is haste really that bad? I am not really convinced. The biggest issue with haste is the mana cost factor. Secondary, it doesn't really do much for our cooldowns. And Holypriest is currently a very cooldown-based spec. But less than half as good in terms of throughput? I fail to really see why that should be the case.

Haste doesn't work with cooldown unless it reduces the cooldown. In that case, you'll speed up the entire cycle of casting. Since haste effectively reduces the cooldown of Holy's primary cooldowns, it works just fine. If you've got 0% haste, an 8:1 cycle of Flash Heal:Serenity resolves in 13.5 secs. If you've got 50% haste, the same healing is provided in 9 sec - a 50% increase in hps. (Note: There are some nuances here, but for the most part haste works just fine with Holy Words and Prayer of Mending).

In terms of 'less than half as good', this is largely due to picking an arbitrary starting point to run the calculations and then trying to translate them into 'stat weights' that poorly represent what is actually going on. Think of that rectangle. If you've got a rectangle 1 inch wide and 10 inches tall, adding an inch to the width raises the area by 100% while adding an inch to the height only adds 10%. But if your rectangle is 10 inches wide and tall, adding an inch either way will add 10% - it's not that width is better than height (they're exactly equal 'cost' in this example) but it looks enormously better because you have a 'width deficit' on your rectangle.

Indeed, this is a problem with most 'stat weights' you see. They go through an enormous number of calculations that yield results primarily reflective of their arbitrary selection of starting point. If you have a decent spreadsheet for a spec and want to see what you should be investing in, you have to enter your current stats or what you're getting out are completely meaningless numbers - it's like you're buying shoes for your wife based on your own foot size.

Most of the time, the discrepancy between stats is sufficiently large and the arbitrary points are the ungeared stats (so 0% Versatility but 8 points worth of Mastery) so you won't have to worry too much until you hit higher levels of gear. But you should be aware of where the model starts to break down.

[Danner]

I get 325 Haste rating for 1%.

That's what I thought it should be - Haste being slightly below Crit in rating cost.
Maybe I did a booper in the math when calculating the ratios. Thanks for correcting it!

As for the rest, I do not at all disagree with your input here. I'm sort of thinking in the same veins. (Apart from the mythic vs raid idea, that was an interesting thought!)
I cannot quite get my head around haste supposedly being THAT MUCH worse than other stats.
The mana cost is a thing to have in the back of your head when stacking haste, but it's not really a showstopper.
So I'm thinking something is wrong in those weights.

I did try to run my own simulation on Ask Mr Robot to figure out the optimal stat weights for my character - maybe they were different than the preset?
But... not only did I not find any stat weights, I also found that the simulation was measuring my HPS by spamming renew like there was no tomorrow.
Despite me having a sizeable amount of mastery on my gear (which then does literally nothing), and flash heal providing both a better HPS and HPM.

So... I guess I'm back to my old impression - treat Ask Mr Robot as a tool to help you improve, not a rule to help you improve.

[cruumash]

Notes on stats:
[LIST][*]...Critical tends to work best with small spells and is often hampered by various talents/traits that abnormally increase critical - the payoff for Serenity is almost non-existent with critical, for example, due to the conjunction of its enormous size and the artifact critical boosting trait. ...

With Echo, however, you get to roll those criticals into the future.

[Djriff]

I'm not sure if it's the same for holy, but shadow's haste rating per % actually changes as you get more haste, it's maddening trying to calculate stuff. That could be the difference you're seeing.

As for haste being good for holy, only 2 spells have any type of synergy with haste, Renew and Echo of Light. Everything else benefits more from crit/versatility. Since haste doesn't increase ticks and only makes stuff tick faster + makes you cast faster at the expense of mana you have to leverage whether or not it's worth going into.

Crit on otherhand while RNG give us extra heal at the cost of no extra mana, same with versatility and mastery.

[VigilantRose]

With Echo, however, you get to roll those criticals into the future.

At a very inefficient rate. That "+1% healing for +1% critical" becomes more like "+0.3% healing for +1% critical" when you're counting on Mastery in this way. Also, if the original heal overheals with the critical, the Echo of Light is almost certain to be wasted as well.

- - - Updated - - -

I'm not sure if it's the same for holy, but shadow's haste rating per % actually changes as you get more haste, it's maddening trying to calculate stuff. That could be the difference you're seeing.

I'll file this under "I'm very skeptical but don't have any evidence to contradict" for right now. This would definitely be a significant change from historical norms but I haven't personally investigated it.

As for haste being good for holy, only 2 spells have any type of synergy with haste, Renew and Echo of Light. Everything else benefits more from crit/versatility. Since haste doesn't increase ticks and only makes stuff tick faster + makes you cast faster at the expense of mana you have to leverage whether or not it's worth going into.

Actually, haste isn't even that good for Holy. Echo's size is fixed by the spell creating it - if you heal for 100 with 10% mastery, you're getting 10 points of Echo healing no matter what your secondary stats are. Renew is technically synergistic with haste. But it's only synergistic because you get both a total healing increase and a cast time decrease. However, almost anyone who is getting significant amounts of Renew healing isn't casting it - they're proc'ing it from Benediction. So you don't get the full impact of the haste double-dipping like a Rejuv-spamming Druid would.

So when I commented "maybe Versatility/Haste for Mythic+ 5-man", the emphasis was really on the 'maybe'. When you get hit by an attack 5% larger than your total health pool, the fact that your healing output is 1.25% better is less important than the fact that your damage reduction is 5% less. In a raid setting, "oops, I should have moved more quickly" gets you some dirty glances from healers who have to rez you/pick up your slack. In a 5-man Mythic+, you probably just killed a keystone.

In terms of haste, it's more a recognition of the cost coupled with the limitations of Critical/Mastery. Haste, despite its lack of synergy, will almost certainly add more raw healing to a Holy Priest than Critical - in any environment where mana isn't much of a concern. Mastery will generate more healing in purely theoretical sense, but its lack of effect on either damage or Renew causes me some concern for 5-man (where Holy Nova can be a significant damage source).

What I'm trying to get at is that the normal way to theorycraft these issues is to imagine a 'glass cannon' model where you've got an infinite amount of health available to heal, nothing to do but heal and no concerns about movement/survival. We use a simplistic model like this because it avoids a long, long debate about the relative value of secondary features for every raid encounter and trash fight in the game. What I'm trying to point out is that those details can actually matter sometimes. So while it's fine to blindly accept whatever Mr. Robot tells you to do (it's a relatively small shift in your total healing anyway), if you're concerned about where the actual bleeding edge lies, you at least need to give lip service to such concerns.

[Djriff]

I'll file this under "I'm very skeptical but don't have any evidence to contradict" for right now. This would definitely be a significant change from historical norms but I haven't personally investigated it.

I was getting very weird results when i was doing some calculations in a spreadsheet for shadow. I'd be happy to discuss it with you in a different topic and explore the issue more.

Actually, haste isn't even that good for Holy. Echo's size is fixed by the spell creating it - if you heal for 100 with 10% mastery, you're getting 10 points of Echo healing no matter what your secondary stats are. Renew is technically synergistic with haste. But it's only synergistic because you get both a total healing increase and a cast time decrease. However, almost anyone who is getting significant amounts of Renew healing isn't casting it - they're proc'ing it from Benediction. So you don't get the full impact of the haste double-dipping like a Rejuv-spamming Druid would.

Well sure, i was more commenting on the fact that it does increase the tick rate, but not in a meaningful enough way to stack for haste.

So when I commented "maybe Versatility/Haste for Mythic+ 5-man", the emphasis was really on the 'maybe'. When you get hit by an attack 5% larger than your total health pool, the fact that your healing output is 1.25% better is less important than the fact that your damage reduction is 5% less. In a raid setting, "oops, I should have moved more quickly" gets you some dirty glances from healers who have to rez you/pick up your slack. In a 5-man Mythic+, you probably just killed a keystone.

I personally find Crit to be far more versatile (haha) in Mythic+ than versatility. Sure there are some bosses i may need the extra versatility on, but I feel that you can just brute force heal the majority of stuff. Damage reduction is always good, but if you're not taking any damage then it's a useless stat.

As for raids, the only time you take unpredictable damage is if you make a mistake and stand in the fire. Sure damage reduction will take the edge off of it, but the overall hit to your healing really isn't worth it. Every other scenario of damage will be predictable and the healers would be healing you anyways. Whether you take 100k damage or 150k damage will not change which spell other healers will be using.

In terms of haste, it's more a recognition of the cost coupled with the limitations of Critical/Mastery. Haste, despite its lack of synergy, will almost certainly add more raw healing to a Holy Priest than Critical - in any environment where mana isn't much of a concern. Mastery will generate more healing in purely theoretical sense, but its lack of effect on either damage or Renew causes me some concern for 5-man (where Holy Nova can be a significant damage source).

Sure, if you want to remove mana from the equation, haste will almost always be ahead for every spec (minus maybe druids). The problem is that mana is always a concern, less so in mythic +, but at that point you're trading drink time for clearing time when you could have just stacked crit/mastery. Crit and Mastery also have a special synergy as your Mastery EoL can benefit directly from crit through the application of EoL (think a 2m Serenity) and each tick can crit as well. Sure it's slow healing, but it's a truckload of slow healing for absolutely free. Haste can't match that with mana cost in mind.

What I'm trying to get at is that the normal way to theorycraft these issues is to imagine a 'glass cannon' model where you've got an infinite amount of health available to heal, nothing to do but heal and no concerns about movement/survival. We use a simplistic model like this because it avoids a long, long debate about the relative value of secondary features for every raid encounter and trash fight in the game. What I'm trying to point out is that those details can actually matter sometimes. So while it's fine to blindly accept whatever Mr. Robot tells you to do (it's a relatively small shift in your total healing anyway), if you're concerned about where the actual bleeding edge lies, you at least need to give lip service to such concerns.

I agree, DPS theorycrafting does the same thing. Then movement/damage patterns are considered in real time, but honestly the only major factor to stat weights should be damage patterns. If your EoL is overhealing a lot, then mastery will be devalued for the majority of the fight, but if during the critical phase EoL saves the group/raid then it's amazing. It's hard to quantify all of that while using a spread sheet and you have apply a bit of common sense to it, ie not blindly accept what Mr. Robot, Icy-Veins, or anyone tells you and double check stuff yourself.

Everyone heals a little differently and while 1 heal style may work for you, your gear may not sync up the best with it and you would need to adjust accordingly.