This doesn't make mathematical sense. Even if an exact closed form solution doesn't exist, simulators can do a good approximation, and apparently they have. If Shaman can properly model the old version of windfury correctly, then we can model spell hit.

Besides, what if you were just 1 point under hit cap, and the only way to reach it would involve losing 100 spell power? I guarantee that trade will be a net loss in damage. It's fine for the average raider to "reach hit cap, then work on other stats", and that's generally the correct move. But if you want to absolutely min/max your character, you have to understand that there may be times that being slightly under hit cap is the best choice. Hit rating is a stat that's generally more valuable than alternatives, but it's not infinitely more valuable. One hit rating is apparently just worth 2 spell power (according to an unconfirmed source).

Hmm - I don't see why our dots shouldn't have a closed formula - the closed formula would be incredibly messy, but there is nothing stochastic in any of the shadowpriest mechanics.

When you have a DoT application resist and have to reapply it, all the timing numbers get skewed and shifted as a result. And the actual loss dependd on what other cooldowns get delayed as a result. Sometimes a failed Vampiric Touch refresh will simply cost you half a mind flay and other times it will force you to defer your next Mind Blast as well. Plus having a resisted Mind Flay will cost you only one cooldown instead of two. Even if you were planning on casting a clipped 2 tick Mind Flay, getting a resist might shift up when you are allowed to cast your next non-flay spell.

I'm pretty sure my priest is in the shadowpriest mechanics

Back on topic, I'd argue that there's value in being hitcapped beyond theoretical DPS. Eliminating resists also eliminates the reaction time spent recognizing and reapplying dots--I can spend less mental power on reapplying missed dots and more on fight mechanics and watching a movie on my second monitor situational awareness.

Variable latency and collisions with other spells blows away the idea of a closed formula or things lining up nicely etc.

As for +hit rating, simulating the value of +hit rating isn't that hard. Haste is the real problem child stat for Shadow Priests. It's not a nice easy linear function. Especially when you add in things like the 4pc Tier 8 set bonus.

Yes, I agree with your point about the advantages in terms of reaction time from knowing in advance that all of your spells will hit (except possibly at the start of the fight, or Imp. DP misses which don't matter much and don't change your rotation anyway).

It's actually something that SimCraft doesn't yet really model, but I pointed it out to Dedmon and he agrees that it's something he's considered too and that we should do something about it.

Right, and this means types of fights needs to be modeled into the simulator. There is a fine line here somewhere and it's not very static. A simulator would have vastly different single target nuke numbers for Vezax depending on tactic, and another single target nuke fight like Algalon

I wrote up some analysis on the returns on Spirit for Shadow Priests for the WoW forums in vain hope that perhaps Blizzard will fix it. The math and conclusions might be of interest to people here. Most new shadow priests profoundly underestimate just how terrible spirit is, so I think this is a point worth beating home.

Here's the simple rule of thumb for gear upgrades though. Pieces with spirit are generally two tiers worse than pieces without spirit. Here's the math backing up the claim:

Empirical Proof from Current Items:

Consider these necklaces, each socketted with +23 spell power:

Sapphire Amulet of Renewal - ilvl 239 (Iron Council 25 Hard)
48 int
51 spirit
35 haste
109 spell power (including +11 from spirit)

Chain of the Ancient Wyrm - ilvl 213 (Malygos 10 Normal)
43 int
30 hit
37 haste
79 spell power

The net change from replacing your 25 man hard mode iron council loot with a 10 man Malygos quest reward is:
-5 int
-51 spirit
+30 hit
+2 haste
-30 spell power (including loss from spirit)

So that effectively lets you trade 30 spell power for 30 hit rating. That's actually an upgrade if you are at least 30 points below the hit cap.

Now consider this comparison to another necklace with a socket, this one only one tier lower:

Pendant of the Shallow Grave - ilvl 226 (Thorim 10 Hard)
43 int
41 crit
31 haste
97 spell power

Net change:
-5 int
-51 spirit
+41 crit
-4 haste
-12 spell power (including loss from spirit)

So this necklace makes you lose 12 spell power and 4 haste, but gives you 41 crit. That's just a flat out upgrade.

It's not just necklaces. I could go through every other slot and give examples, but I'll just give one more quick comparison with rings:

Glowing Ring of Reclamation - ilvl 226 (Flame Leviathan 25 Normal)
49 int
44 spirit
39 crit
83 spell power (including 9 from spirit)

Signet of the Kirin Tor - ilvl 200 (Vendor)
34 int
36 crit
28 haste
59 spell power

Net change:
-15 int
-44 spirit
-3 crit
+28 haste
-24 spell power (including loss from spirit)

So you lose a bit of mana and trade 24 spell power and 3 crit for 28 haste. This is a very minor downgrade, but the items are roughly comparable.

Now compare the Inscribed Signet, one tier higher:

Inscribed Signet of the Kirin Tor - ilvl 213 (Vendor)
38 int
37 crit
38 haste
66 spell power

Net change:
-11 int
-44 spirit
-2 crit
+38 haste
-17 spell power

This is a trade of a bit of mana, 17 spell power and 2 crit for 38 haste. That's a clear upgrade.

And you will see this clear pattern for pretty much any set of items you look at. Pieces with spirit are a full two tiers worse than pieces without spirit. I don't mind spirit being a bit worse than other stats, but this amount of terribleness is just too far over the line. Fully half of the cloth drops are meaningless to shadow priests because they have spirit.

Mathematical Proof from Item Budget Values

Consider the trade off between spirit and spell power from item budget points. A gem socket can provide 23 spell power from a runed cardinal ruby or 20 spirit from a sparkling majestic zircon. That 20 spirit becomes 22 after blessing of kings, which turns into at best 4.4 extra spell power. Assuming you don't need mana regen from spirit (and you don't with a 2 minute dispersion), this means the efficiency of spirit relative to spell power is 4.4 / 23 = 19%. As a stat, spirit is 19% as effective as spell power.

Now consider the trade off a rogue makes between attack power and strength. They can either take 20 strength (22 after blessing of kings) or 40 attack power. The effectiveness of strength to a rogue is 22 / 40 = 55%. In other words, strength is more than twice as good for rogues (55% of their main stat) as spirit is for shadow priests (22% of our main stat).

You may wish to edit the first paragraph on your WoW forum post to change Vampiric Touch to Vampiric Embrace for the group healing thing.

By the way, I agree with the points you raised about spirit scaling and it's current rather woeful returns at present.
It reminds me of crit rating scaling in TBC. A value so small that some people just ignored it.

Bumping it up to around the 60% mark (or say 66% mark with BoK) would be ideal as it would put spirit in roughly the same place as haste and crit are. A little bit less, but that would be okay, because it does after all boost our regeneration too, which is of course a non-issue in T9 gear especially, but if you're forced to Disperse in your current gear then it's still a useful component to the stat. Time spent Dispersing for mana is time you're not casting Mind Flay and is a fairly large loss of DPS. I spent a while yesterday tweaking the logic SimulationCraft uses to determine when to cast Dispersion and Shadowfiend so that in the scaling tests (esp. those that made use of SW: D) it wouldn't hit the point where it would decide to cast Dispersion when intellect etc. was lowered enough and thus lose a lot of DPS and thus artificially inflate the +int and +spi scale factors.

I wouldn't ever expect spirit to be *as* good as crit/haste, because it also provides regen, and Blizzard has specifically stated they don't want to go the route of making a regen stat also give great throughput.

Even if players don't care about regen, it's still something Blizzard considers. 30% spirit -> damage would at least bring priests closer in line with moonkins for spirit value. Spirit tap and imp. spirit tap could probably use an overhaul since they really didn't fare well after the overall out-of-5-second-rule regen nerf (and weren't that good to begin with), so maybe some change to those talents to give it some other additional benefit, like how warlocks gain a secondary DPS benefit from spirit in reduced number of life taps.

Keeping suggestions and complaints in line with a line of thinking Blizzard is already following I think is more likely to produce positive results than just saying "spirit should be as good as crit/haste". It doesn't have to be "as good" as long as it's not "nearly useless", which is the current state.

If anything, the PvP Dispersion buff was another huge blow to spirit as having value for shadow priests. As someone who only recently dropped spirit tap/improved spirit tap, any mana problems I would have suffered on longer fights (basically just Yogg and Steelbreaker) was pretty much completely erased by more aggressive/intelligent shadowfiend/dispersion usage.

You should check out the DPS scaling on Intellect for Enhancement Shamans. ;P

This is from a scale factor test I just ran on the latest SimCraft using the Enh Shaman T8 BiS profile:

Scale Factors:
Shaman_T8_16_55_00 Str=0.70 Agi=1.07 Int=0.98 SP=0.67 AP=0.63 Exp=1.15 ArPen=0.99 Hit=1.30 Crit=1.15 Haste=1.28 Wdps=3.49 Lag=0.00

The curious part about all of this is that our other cloth dps tends to embrace spirit more readily than spriests. As a whole, our itemization would be closer to a ele shaman and disc priests than a warlock or a mage, yet the gear doesn't reflect how awful Twisted Faith scales with spirit.

As a whole, when improved spirit tap procs, our spellpower doesn't receive the same benefit as intended. My solution would be: improve the spirit->spelldamage ration by doubling or tripling it to 20-30%. Seeing as the average spirit ranges between 400-500... a priest would get 40-50 spellpower (with a max of 53-67 spellpower during Imp. Spirit Tap procs), whereas a warlock and boomkins would receive 120-150 +damage from fel armor and Imp Moonkin form the same (theoretical) gear. Mages, typically turn spirit into crit, the same 400-500 spirit would turn, 225-250 crit... According to power points, 225 crit is a greater boost than 45-55 spellpower by a large sum.

Insofar as gimping spirit tap and imp spirit tap, I doubt blizzard would affect that seeing as it's the greatest leveling tool for priests, it may not serve a raiding purpose necessarily, it's far from useless, which is why retooling twisted faith (how about making it a 3/3 talent 10%/3% increase respectively) I believe would be the best course for first, increasing our max dps and our gear appropriations as proscribed in tier gear. It's not like if you'd stacked spirit like a holy priest, you'd be able to heal effectively and dps in a competitive raiding level.

I guess the only drawback would be in a situation like

Talent Calculator - World of Warcraft

where you go shadow/holy to build on spiritual guidance... (although, it'd be a minimum loss of 54 SP (imp Inner Fire)+5% damage from SW: D, DP, and MF via twin discipline), for the most part, decrease longevity lacking meditation and general viability.

It's not as if spirit isn't being well utilized as a secondary/non-primary stat. Whereas a warlock or mage would be like, has spirit? SURE ILL USE IT. As a spriest, you'd have to be more careful in current content with itemization. But the non-spirit lacking gear that is made up for in other stats, is normally made up for in... +mp5 gear. Not very useful either.

EJ, do you think blizz is causing this as a general concept because of our 6% hit gift? I don't believe so.

While I share your thoughts regarding Spirit, I can't agree on your post as a whole. You argument against spirit as a dps stat for spriests, but this begs to ask the question if our dps is too low. Or to be more accurate, does Blizzard think our dps is too low? If so - is Spirit the stat we should scale better with? Should we scale better with haste instead perhaps? It's not exactly a stat we shine in.
Quite frankly I don't think you^W we will catch Blizzard's attention unless we can argument and prove that we need Spirit on our gear to not go oom and that this causes us to do too low damage in full CC.25 gear.

T9.245 has 223 Spirit on it. If the community as a whole could re-itemize we would surely spend these points in Spellpower, Crit rating or even Haste. If this would make us struggle with mana, would intellect be the stat to invest some points in. Who knows, no one has really seen CC.25 hardmodes with 3.2 Replenishment yet.

Again I'd like to point out that I agree with your thoughts regarding Spirit as a dps stat, I just can't imagine Blizzard seeing it as a problem based on your post.

One of my core points was that increasing the value of spirit wouldn't actually increase our damage-- it merely makes more gear available to us. Right now, shadow priests are just ignoring high level pieces with spirit when they can get other pieces without spirit. If spirit pieces weren't so bad, we would have more upgrade options, but the DPS gain from going from 213 to 226 level gear would remain the same. You could just happen to use, say, your T8 pants for that slot instead of the pants from Malygos.

Looking throu this thread there are clearly different opinions that work. Have been playing around with a spec and have come up with this.

Talent Calculator - World of Warcraft

It differs slightly from the build described in thread 1 but i have a few questions about this idea.

Improved shadowform? is it worth it, should you need to be using fade in a raid? probably not. This gives you added points to move further down the Disc tree and pick up Mental agility. Is this not a better format?

There's a couple of problems with this build.

#1) Mental Agility isn't that good.

It reduces the cost of Devouring Plague and that's basically it. You should almost never cast Shadow Word: Death and your Pain will be refreshed by Pain and Suffering. So those 3 talent points save you less than 200 mana per minute. Given that you spent 5 talent points to get there, this is less than 4 mp5 per talent point spent. You would be better off with 5 points in Spirit Tap. (Incidentally, 5 points in spirit tap is also a very bad idea due to low returns, but that has been discussed to death earlier in here).

#2) Improved Shadowform is pretty important

The talent is all about 70% pushback resistance. If you are in any kind of AoE situation, it hurts a ton for your Mind Sears to get pushed back. And there's a lot more random pushback effects than you would guess at first glance.

Given that you're spending 2 on Improved Shadowform, the question is where the other 3 points go. Shadow Affinity and Silence seem to be the two most common choices, though in the end it just doesn't matter that much. Silence is probably the best choice for raiding though-- a ranged interrupt is invaluable, even if it's on a long cooldown.

Yes you are missing something.
Other caster have a much better damage return on spirit that it's not two tier worse but none or only one.
What tedv recommends is to boost the damage return on spirit to give us more room in choosing good raiding gear in equal tiers.This would not boost our overall dps as the best items will still be the one without spirit. But it would be easier to get good equipment.

Has anyone been able to generate a spreadsheet similar to the one hunter's use?

A couple of people on shadowpriest.com have made up some spreadsheets. The most popular sources of such stuff though are SimulationCraft and Rawr.ShadowPriest.

As I understand the Simulationcraft and Rawr.shadowpriest these are really just to understand the "value" of your stats. What they wont do is tell you that your over/under the hit cap if you swap a piece of gear out which is what is so nice about the Hunter spreadsheet.


TY for the response I will look deeper into Shadowpriest.com.

Hi, this is my first post here, but I was hoping for some help from ya'll. I've got an incomplete formula for figuring out how much Haste Rating is required for extra ticks of Mind Flay between Mind Blast cool downs. Right now it doesn't account for latency, but that's because I don't really understand how latency slows down Mind Flay.

Under 0 latency, the formula is 3279{1 - (5.5/N)} where N is the number of Mind Flay ticks you're going for and is an integer higher than 5. Obviously this doesn't account for human reaction time and latency, but it's a good baseline for determining the minimum haste rating to shoot for, assuming I didn't mess up my math.

What I'd like to do, is factor in latency now, but I need to understand how that needs to be factored. Is the latency value displayed in game one way communication, or is it round trip? Do I need to account for latency on each cast of Mind Flay? And do I need to account for it as a one-way trip, or do I need to account for communication to and from the server?

And this is Blizzard's penalty for regen stats at your gear(not that I agree with them). I just don' think you can convince Blizzard we avoid Spirit gear since we don't need the regen from it. We don't know how CC.25 hardmodes will be, but mana was pretty tight in certain Ulduar hardmodes, especially Yogg+0.
Quicker regen does mean more damage and I believe we would see more accurate theorycrafting by looking at how important mana potency is in hardmodes