• Dear Guest,

    You're browsing our forum as a Guest meaning you can only see a portion of the forum in read-only mode.
    To view all forum nodes and be able to create threads/posts please register or log-in with your existing account.

    TwinStar team

Paladin Spell Crit for holy Paladins

Chainsaw

Authorized
Joined
Apr 1, 2016
Hi all,


lately I've read through some of my holy paladin calculations I have done on another realm and I want to share them with you. The basic questino for me was if it is more valuable to obtain crit items or just go for mp5+addheal. In the following calculations I'm talking about AVERAGE values, not about min/max values. Also the healing coefficients I use are 0,43 for Flas Of Light (FOL) and 0,71 for Holy Light (HL). Another thing to mention for the calculations is that web sources say 54 int is 1% spell crit for paladins, the theorycraft addon says it's 29.5 int for 1% spell crit (maybe a Kronos admin could reveal how it's working on this realm). I will do the calculations for the 29.5. What I won't do is considering the 12% more healing from talents, because it won't make a difference for the calculations (it doesn't matter if the *1.12 scaling applies to the healing value of 1% crit or the healing value of add heal).


First of all we need to aks ourselves the question what makes 1% spell crit on healing spells seem so special? Yes, obviously that talent, that gives you 100% of the spell's mana cost back, after you get a crit with a spell. Also something that sounds very pleasing is the fact that critical heals do 1,5x healing instead of 1x healing, so these lovely crits give you 0,5x bonus healing.


To do proper calulations we need a formula to work with. The addheal you have on your gear is multiplied with the healing coefficient of the spell used. So let's start with the average healing of our FOL and HL:


Code:
FoL1: 67.5 (healing 62-73, so average is -> 62+(73-62)/2 = 67.5)
FoL2: 103.5
FoL3: 154.5
FoL4: 209.5
FoL5: 283.5
FoL6: 363.5
 
HL1: 43.5
...
HL8: 1317.5
HL9: 1680.5 (AQ20 book)


For future calculations we'll call the base average healing of a spell b and the total average healing done by a spell t. With zero addheal the average healing of FOL1 would be t=b.


Code:
t=b


t=average total healing
b=average base healing


Now we assume to have 100 addheal. So according to the FOL coefficient of 0.43, we get 43 addheal, which will make our FOL heal for 67.5+43=110.5.
The calculation is easy here, it's t=b+k*a=67.5+0.43*100=110.5.


Code:
t=b+k*a


t=average total healing
b=average base healing
k=healing coefficient
a=addheal bonuses


Involving crit is also not that complicated. A crit gives 1,5 times healing (in other words (1+0.5)*healing). The crit rate c consists of the % crit on gear and intellect/54 or intellect/29.5.
Be aware that both the intellect talent and Blessing of Kings increase your intellect by 10%. They get multiplied, so 1.1*1.1=1.21 out of 2x10% we get 21% more int (since patch whatever most modifiers stack multiplicative).
c=0.01*(critrate on gear + (1.21*intellect without blessing of kings and talent)/29.5)
Adding our new knowledge to the formula it will look like that:


Code:
t=(b+k*a)*(1+0,5*c)


t=average total healing
b=average base healing
k=healing coefficient
a=addheal bonuses
c=0.01*(critrate on gear + (1.21*intellect without blessing of kings and talent)/29.5)


Here we go! Let's continue with our calculation by adding 1% crit! t=(67.5+0.43*100)*(1+0.5*0.01)=111.1.
Wow! Only 0.6 more healing from that 1% more crit, compared to the 110.5 before! Not as much as expected.

How much addheal would we have to stack instead of that 1% crit to reach 0.6 more healing (or: how much addheal that 1% crit is worth)? Easy calculation!
67.5+0.43*(100+x)=111.1
43+0.43*x=111.1-67.5
x=(111.1-67.5-43)/0.43
x=1.4


As you can see from the FoL1 formula without crit and addheal t=b=67.5 compared to the same formula for FoL6 t=b=363.5 the difference in healing is roughly 300. Now if we want to get to 363.5 healing (average base healing of FoL), we can calculate the amount of addheal needed for FoL1:

363.5=67.5+0.43*x
x=(363.5-67.5)/0.43
x=688

So we need nearly 700 addheal, just to compensate the rank difference and the 105 mana FoL1 costs less than FoL6. But what has this to do with spell crit?
If we are using FoL1, we get (critbonus*critchance*average base healing) 0.5*0.01*67.5=0.3375 healing just from base heal per 1% crit, if we are using FoL6 we would get 0.5*0.01*363.5=1.8175 healing per 1% crit. This means the higher the healing rank cast, the higher the benefit from crit. BUT for EVERY rank of FoL, 1 healing = 2.3 add heal on equipment (calculation: 1=0.43*x <=> x=1/0.43=2.3). So since we are healing with the 35 mana FoL1 most of the time, it's much more beneficial to get 2.3 add heal instead of 1% crit.

Another example, a BWL/AQ geared paladin with 1000 add heal and FoL1. Benefit from 1% crit is 0.5*0.01*(67.5+0.43*1000)=2.5 healing, which is 2.5/0.43=5.8 add heal. Even with 1k add heal, 1% crit makes only a very minor difference.
Please be aware of the fact that higher ranks of HL and FoL give you more add heal bonus, but they also cost much more mana and - as far as I know - higher ranks of your spells are only used in emergency situations.


Everyone who understood the calculations until now, should be able to do the transfer to do these calculations with holy light. For all those who are too lazy to do them, you can take a look at the spreadsheet:

SPREADSHEET


The mp/5 from 1% crit calculations are the following (also found in the spreadsheet):

(mana cost of the spell)*0.01*(5/casting time)

To make it a little more realistic, add 0.05s to your casting time, so for example out of 1.5s for FoL you get 1.55s. Let's do the calculation for FoL1:

35*0.01*(5/1.55)=1.1/mp5

Please note that this ONLY APPLIES WHEN YOU ARE CASTING PERMANENTLY WITH NO BREAK.
Let's say you have to move for 2.5s every 20s. So your pure casting time is 17.5s, which is 3.5 times the mp5 value. If you were able to cast the full 20s without moving, you would have 4 times the mp5 value.

3.5*1.1mp/5=3.85
4*1.1mp/5=4.4

So in the movement scenario you would have regenerated 3.85 mana, in the non movement you would have regenerated 4.4 mana over 20 seconds (which is nothing compared to regular mp/5 on gear).

Check the Spreadsheet if you want to see how much mana the other healing ranks of your spells bring you per 1% crit.
Please feel free to comment and share your feedback with me.

Best regards!
 
Changelog:


* Gear comparison tool at the document's bottom! (13 April 2016)

* Fixed an error in the Item B mp5 and mpc formulas, that accidentally took item A int instead of item b int. (16 April 2016)

* Added support for the 12% healing and 10% int talent, target with or without Blessing of Light, you with Blessing of Kings (10% int) and/or Zg buff (15% int), and Libram of Divinity. Thanks to Vnm for your input! :) (20 April 2016)
 
Last edited:
What I won't do is considering the 12% more healing from talents, because it won't make a difference for the calculations (it doesn't matter if the *1.12 scaling applies to the healing value of 1% crit or the healing value of add heal).
Does this talent not work the same way as it does for other classes? The 12% only apply to the base value of the spell.
If so crit is more beneficial to this talent then +heal.

(base * 1.12 + add heal) * crit
base * 1.12 + add heal

Add heal only adds on top while crit also multiplies the base value.

Other then that, pretty solid work mate.
 
Does this talent not work the same way as it does for other classes? The 12% only apply to the base value of the spell.
If so crit is more beneficial to this talent then +heal.

(base * 1.12 + add heal) * crit
base * 1.12 + add heal

Add heal only adds on top while crit also multiplies the base value.

Other then that, pretty solid work mate.

Well that's something I should test with my 35 paladin and with my mage. Thanks for this input!

If what you say is true, crit will gain more weight than add heal.
 
Nice work. I have done my own calculations, you can view them here:
https://docs.google.com/spreadsheets/d/1dk0fQz1iap9WWRwLIYgnR0Qdmt9zUrK2EWEnGmzjSb8

You get enough int/mp5/crit from gear so you shouldn't have any mana problems with proper consumables. Stacking stats which increase your HPS the most (the PURE HEALING stat weights on my spreadsheet, item list on page 3 uses them for VALUE1) should be the way to go.

EDIT: Healing Light only increases the base heal listed on the tooltip.
 
Last edited:
Nice work. I have done my own calculations, you can view them here:
https://docs.google.com/spreadsheets/d/1dk0fQz1iap9WWRwLIYgnR0Qdmt9zUrK2EWEnGmzjSb8

You get enough int/mp5/crit from gear so you shouldn't have any mana problems with proper consumables. Stacking stats which increase your HPS the most (the PURE HEALING stat weights on my spreadsheet, item list on page 3 uses them for VALUE1) should be the way to go.

EDIT: Healing Light only increases the base heal listed on the tooltip.

Thank you for also sharing your knowledge with us!

Have you any source or evidence for the 12% on base heal? Using my formulas it can be easily tested if you put on much add heal gear (600+) and calculate the estimatrd healing minima and maxima for fol1 with your amount of add heal, one time mhltiplied with 1.12 and one time not. If your healing exceeds the max value of the fol without *1.12, you know the *1.12 applies to add healing as well.

Edit:
Healing of Fol1 is 62-73. So if you have 600+heal on gear, and the 12% healing talent, you should get the following values:

min heal: 62*1.12+(1.5/3.5)*600=326.58
max heal: 73*1.12+(1.5/3.5)*600=338.90

If the 12% apply also to +heal on gear, you will get greater numbers than max heal. that's the point.
Your new max healing (crits excluded ofc) would be (73+(1.5/3.5)*600)*1.12=369.76

That's the way you can easily test it.
 
Last edited:
Yes, for example(just tested ingame) my R1 FoL heals for 496-505 with 872bonus healing, libram of divinity(flat 53 bonus on FoL) which agrees with math:

872*15/35+75(R1 average with 12% bonus from talents)+53(libram)=501,7avg healing
 
Yes, for example(just tested ingame) my R1 FoL heals for 496-505 with 872bonus healing, libram of divinity(flat 53 bonus on FoL) which agrees with math:

872*15/35+75(R1 average with 12% bonus from talents)+53(libram)=501,7avg healing


Well taking the average healing is not as precise as calculating your max. heal value and looking if you heal more. Also - what I would do - is healing naked one time with and one time without tome to see if the 53 healing really apply 100%.
I believe you are definately right, but I have no way (= I have no 60 paladin with tons of gear) to check it 100%.
 
Yes, the libram adds flat 53 bonus to FoL on this server, same as BoL does for FoL/HL. Some addons (I think Heal Stats at least) calculate libram as 53 bonus healing though and you will heal slightly more than the values given by that addon.

If the libram worked other way, my R1 FoLs would have healed ~30 less. R1 FoL's min-max healing range is only 12 (69-81 from the tooltip with Healing Light), its easy to see that the libram works the way i described.
 
Psojed you sure? Just tested ingame before servers went down:

http://i.imgur.com/OZn2jLX.png

Not a big sample size, but should be enough to prove the point. First 6 casts of R1 FoL with libram, no healing gear on, average 128,833..., then without libram 5 casts, average 75,6.

Difference is 128,83-75,60=53,23.
 
I went ingame to check again, and it really adds the full value. Guess my math went to sleep early :biggrin:
Sorry for the misinformation.
 
I am adding support for the libram of divinity, the 12% talent and blessing of light to my spreadsheet, just as you told me it works. But I'll reserve the option to remove several changes if they prove to be wrong.

Would be very nice, if you could do more precise max value tests with and without 12% talent, libram of divinity and blessing of light (with and without talent, with and without spell coefficients for the latter two, just to be sure) and present them in a table.

Thank you in advance and best regards

Spreadsheet updated, thank you Vnm.

Now you can also find out the value of 1 if you just create an imaginary item with 0 int and one with 1 int and compare them.
 
Last edited by a moderator:
The spreadsheet I linked already has support for all of those and much more, you can download it and change them as you like and see if it gives same results as your calculations. I don't really feel its worth my time doing thousands of casts per rank and writing down everything, especially without any kind of combat log analyzer.

When I was working on my spreadsheet I did test nearly everything(libram, blessing of light, healing coefficients, and buff stacking for example) that they work like they should work on this server, though I didn't save the data because at that time I was doing this only for myself. Only thing I didn't test extensively was if 29.5int=1 spell crit%, though that would again require tens of thousands of casts to get precise data.
 
The spreadsheet I linked already has support for all of those and much more, you can download it and change them as you like and see if it gives same results as your calculations. I don't really feel its worth my time doing thousands of casts per rank and writing down everything, especially without any kind of combat log analyzer.

When I was working on my spreadsheet I did test nearly everything(libram, blessing of light, healing coefficients, and buff stacking for example) that they work like they should work on this server, though I didn't save the data because at that time I was doing this only for myself. Only thing I didn't test extensively was if 29.5int=1 spell crit%, though that would again require tens of thousands of casts to get precise data.

29.5 int = 1 spell crit , but don't forget about "divine intellect" which increase the Intellect up to 1.1 instead 1
 
Backwards calculating it from empirical data that was collected during vanilla wow paladins need arround 60int/crit.
Pretty similar to other classes.

inb4 Talhulers post
 
Here's the discussion that took place a few years ago, about 29.5 vs 60 int per crit that favors the conclusion that paladins need 54 int per crit.
 
Here's the discussion that took place a few years ago, about 29.5 vs 60 int per crit that favors the conclusion that paladins need 54 int per crit.

http://wowwiki.wikia.com/wiki/User:Pauladin/Paladin_spell_critical_rates

When you have low intelligence, your crit chance rises quickly as you gain intelligence. As you get higher and higher intelligence, it becomes harder and harder to keep gaining crit chances. At the highest levels of intellignce, an incremental increase in intelligence does less and less for your crit chances.

Just to make it clear.

http://paladin-guide.blogspot.mk/2005/12/stats.html

Increases your chance to score a critical hit with spells. A number between 27,5 and 29, 5 points of Intellect equal 1% chance to crit with spells. See this post (the post is direct link toward original blizzard forums) for details.
 
Here's the discussion that took place a few years ago, about 29.5 vs 60 int per crit that favors the conclusion that paladins need 54 int per crit.

In the wowwiki post, the user made a conclusion at the end that the int/crit% ratio isn't static, while the person in your link uses the same data, EDIT2: read too fast. He assumes there is certain base crit% chance.

I haven't seen this wowwiki post before, and for me the conclusion looks very promising. His/her int/crit% ratio is (30/350)*int+10. But then again, why would paladins' have nonstatic int/crit% ratio when all the other classes' ratio is static?

And on top of all that, we don't know if the ppl who coded the server went with 28.5, 29.5, 30.0, 54.0, 60.0 or nonstatic int/crit% ratio without extensive testing on this server.

EDIT: http://www.lurkerlounge.com/forums/printthread.php?tid=4286
some more talk about Tseric's comments. Again ppl saying ~29.5int=1crit% for paladins. Though some ppl say that the info given my Tseric might be a bit wrong (and going by the comments he said earlier that everyone has 5% base spell crit but then posted different stuff which conflicts that information).
http://forums.elitistjerks.com/forums/topic/9434-paladin-bugs/
Here aswell one of the posters says that they calculated int/crit% ratio to be 30 before 2.0.

EDIT3: Ok I read more into the thread you linked. I'm starting to agree that 106int is the expected value (you have 5% spellcrit then) and it is ~60int/crit%.

EDIT4: Did some testing: http://i.imgur.com/OZCkoiC.png
R1 FoL, ~207heal per cast(crit included), ~1.9k casts, only buff is GBoW, 5% crit from talents.
If someone wants to continue testing, you are welcome to do so. Or PM Chero to look at the code and tell us the values & math :wink:
 
Last edited:
So.. in conclusion?


You should really just make a thread and list every change you've made for paladins here on Kronos.
 
In conclusion use the spreadsheet to see what item is better. You can roughly say in gerneral: crit is better for mana reg and gives better healing on high rank holy light and add heal is better for Fol spam.
 
In conclusion use the spreadsheet to see what item is better. You can roughly say in gerneral: crit is better for mana reg and gives better healing on high rank holy light and add heal is better for Fol spam.


..... I really don't care what you think.

I care about the actual number they have placed in their data base. What is the int per crit on Kronos?
 
Top Bottom