Forum Settings
       
Reply To Thread

Crit vs. MP5 CalculationsFollow

#1 Jun 17 2008 at 12:47 PM Rating: Good
***
1,599 posts

OK - I was working on a potential update for the Pre-Kara Holy Guide (of course, it will need the blessing of everyone on this forum - especially Dil), and wanted to add something in about Crit vs. MP5. Since it's a touchy subject, I figured that we should let the math do the talking. So I went through, and tried to do the math. For those of you that hate math, please just ignore this post. For those of you OK with it, please review it and let me know if I did it right.

So here is what I was thinking:

Crit gives Paladins back mana as 60% the base cost of the spell. The problem with Crit is that it is not reliable. Yes, if you cast 1 million heals, then you are prpobably going to be pretty close to the average. But in a fight, you may get 10-15 heals cast without a crit (even if you have 20% crit). If you are relying solely on crit for mana return, you may be out of luck. So, basically, the return is great, but based on luck. Also, your mana return is much greater for Holy Light than Flash of Light. So unless you spam Holy Light, you will not get good mana return by stacking too much crit without MP5.

MP5, for this discussion, is Mana returned every 5 seconds "WHILE CASTING". MP5 is good in that it is a reliable and continuous source of mana regen. However, it is harder to stack than crit, so it generally scares people.

Your best bet is to balance the stats so that you are flexible with all aspects of Paladin healing. Isolating yourself one way or another will just gimp you in the long run.

So here are the conclusions of my calculations.

** Max rank Holy Light crit will give you back 504 mana per crit. So 1% crit = 12.59 MP5 for HL spamming with Light's Grace.
** Max rank Flash of Light will give you back 108 Mana per crit. So 1% Crit = 3.6 MP5 for FoL spamming.
** Max rank Holy Shock will give you back 390 mana per crit. I won't even do this calculation since you should never spam Holy Shock.

Calculations (if you care):

You spam Holy Light with Light's Grace, making it a 2 second cast.

Let's say you have 20% crit.
On average, you will crit 20% of the time (I know that statistically, you can go higher or lower, but I'm going to use average to make my point.)
1.00 / 0.2 = 5 .... So you should crit every 5 casts (on average)
5 casts = 10 seconds (since each cast is 2 seconds)
So, on average, you will get 504 mana back every 10 seconds (i.e. 5 casts)
504 mana / 10 seconds = 50.4 MP1
50.4 MP1 = 252.0 MP5

Now, let's say you have 19% crit.
1.00 / 0.19 = 5.26316 .... So you should crit every 5.26316 casts
5.26316 casts = 10.526 seconds
Now, you get 504 mana back every 10.526 seconds
504 mana / 10.526 seconds = 47.88 MP1
47.88 MP1 = 239.41 MP5

So the difference between 20% crit and 19% crit with Holy Light spamming is (on average):
252.00 MP5 - 239.41 MP5 = 12.59 MP5

whew.....

It can be more, and it can be less - depends on your luck.

Anyway, I duplicated this calculation for Flash of Light, and you get 3.6 MP5 difference for 1% crit. So as you can see, FoL spamming does not get good MP5 return with crit. However, Holy Light appears to get a lot of return.

Does this look correct?

Thank you all for reviewing this. And please, be critical. I can take it. I'd rather get beat on and get it right :)
#2 Jun 17 2008 at 1:22 PM Rating: Good
*
146 posts
We really need it in graphical format, with mp5 per 1% crit as a function of crit, then you can find the level that is right for you.
#3 Jun 17 2008 at 2:05 PM Rating: Good
**
808 posts
The trouble with this kind of approach, even if we knew the optimal blend of FoL and Holy Light usage, is that it treats spell crit only as a source of mp5. Somehow, somewhere, you need to give crit some credit for what it actually does- heal your target for an awesomely huge number. Unfortunately for the theorycrafting, the value that anyone would put on crit from that perspective has to be inherently subjective and arbitrary.

Other than that, your calculations appear mathematically correct.


#4 Jun 17 2008 at 3:52 PM Rating: Good
***
1,004 posts
It seems to jive with other recent theorycrafting about this.

No one really accounts for real-time use though. You don't just spam one spell in practice, so the actual value should fall somewhere between those two figures.

Also.. I recently read an article in which crit was described more in terms of "Effective Mana" rather than mp5, just as certain factors add longevity to your actual HP. I liked this because crit doesn't function as a source of mana return if you're silenced, stunned, OOM, etc. In the last case especially, mp5 is important.

Also, mathematically you can skip the whole mp5A - mp5B thing where both represent a different crit %. You'd get the same value if you did the calculation once with a 1% crit chance (rather than twice, with a 19% and 20% chance) because lo and behold, that's what you're trying to determine. Also, you could just divide your crit% out of the large mp5 value to the same effect.
#5 Jun 18 2008 at 4:54 AM Rating: Good
***
1,599 posts
Thanks all for the replies.

HoyadinFTW wrote:
Somehow, somewhere, you need to give crit some credit for what it actually does- heal your target for an awesomely huge number.


Absolutely. I was so focused on mana regen, that I forgot to add the awesome healing power of a crit. Thanks.

Losie wrote:
No one really accounts for real-time use though. You don't just spam one spell in practice, so the actual value should fall somewhere between those two figures.


Good point. Do you mind if I quote you on this?

Losie wrote:
Also.. I recently read an article in which crit was described more in terms of "Effective Mana" rather than mp5, just as certain factors add longevity to your actual HP. I liked this because crit doesn't function as a source of mana return if you're silenced, stunned, OOM, etc. In the last case especially, mp5 is important.


I'm not sure what you mean by "Effective Mana". Do they mean to treat it like 1% crit would be equivalent of you having 500 extra mana (or whatever, I'm just using arbitrary numbers). I don't know if I'd agree because your mana return from crits is based on time too (i.e. over time, you get more crits). Maybe I'm misunderstanding what you're saying.

All I was doing was trying to give a guideline for people who are deciding on whether they want a piece of gear with MP5 or with crit. Every Paladin runs into this while gearing up for Kara. I just wanted to make sure that people understood the value of MP5.

I've heard a lot of people argue that Paladin don't need MP5 because they get their mana back from crits. This is not true, and I wanted to put that myth to rest.

SO what if I wrote something like:

==Crit vs. MP5==
Oh, this is a touchy subject...

Crit increases the heal by an additional 50%, and gives a Paladin back mana as 60% the base cost of the spell. The increase in heal is awesome, and you will always smile everytime you see a crit. The problem with Crit is that it is not relaible. Yes, if you cast 1 million heals, then you are probably going to be pretty close to the average. But in a fight, you may get 10-15 heals cast without a crit (even if you have 20%). So relying on crits for anything (i.e. increased healing or mana return), then you are relying on luck. The higher your crit, the less luck you need :)

MP5 is Mana returned every 5 seconds "WHILE CASTING". MP5 is good in that it is a reliable and continuous source of mana regen. However, it is harder to stack than crit, so it generally scares people.

Your best bet is to balance the stats so that you are flexible is all aspects of Paladin healing. Isolating yourself one way or another will just gimp you in the long run. Think of Crit being your left leg, and MP5 being your right leg. You need both legs. If you have one leg shorter than the other, then you are gimping around.

Not everyone believe this, so here is some data:

The following calculations are only true if you are spamming a particular heal, and never have to move, are never silenced, never casting cleanse, never drinking pots, never getting hit, etc... In reality, your mana return from crit will be much lower.
** Spamming Flash of Light: 1% Crit is approx 3.5 MP5
** Spamming Holy Light: 1% Crit is approx 12.5 MP5

You don't just spam one spell in practice, so the MP5 value should fall somewhere between those figures.
Also, you never get to just stand around casting. You have to move sometimes, get hit sometimes, stunned/feared sometimes, Cleanse sometimes, etc... This all reduces your mana return from Crits.

So in reality, your mana return from 1% crit will generally be 2-5 MP5, depending on the situation. As you can see, stacking MP5 is much better for mana return when compared to crit. However, crit is still VERY important, so please make sure to stack both. Target the stats I listed above, and you willl be just fine. Once you hit those values, you can stack whatever you want. You will have enough Crit and MP5 to get you through Kara.

FYI - the stats I recommended are:

To start Kara, you want the following un-buffed:
** +1100 heals
** 15% crit
** 60 MP5 - while casting
** 8k mana
** 6k health

To do a full Kara run, you'll want the following un-buffed:
** +1500 heals
** 18% crit
** 80 MP5 - while casting
** 9k mana
** 7k health


Edited, Jun 18th 2008 9:26am by YJMark
#6 Jun 18 2008 at 10:39 AM Rating: Good
*
91 posts
Good stuff, Mark; it probably won't settle the crit vs. mp5 debate, as paladins are an argumentative lot, but provides food for thought. I've been thinking about some ways to extend your calculations over a little more time, but I haven't quite broken out my abacus just yet.

However, I have to disagree with this:

>>MP5 ... is harder to stack than crit, so it generally scares people<<

When I started Kara, I was barely ready (it was OK, my group knew this, and were in similarly bad shape). My healing was about 1050, my mana and health pools were limited, my spell crit was low. One thing I did have, however, was mp5. Having most of the gear on Dilbrt's excellent guide, plus a coupld of 'of the Elder' pieces, judiciously enchanted, my mp5 unbuffed was up in the upper 70's/low 80's; with Blessing of Wisdom into the low 100's. I conciously aimed for mp5 at this point, knowing that my mana was a little low, but I didn't find it difficult to do. Just thought I'd share that, others may well disagree, don't know.
#7 Jun 18 2008 at 11:01 AM Rating: Good
***
1,599 posts
meagherly wrote:
However, I have to disagree with this:

>>MP5 ... is harder to stack than crit, so it generally scares people<<


I do see what you're saying. If you follow Dil's guide, then this is true. However, most pre-Kara Healadins I've met didn't follow that guide. I'll make sure to remove that statement, as it only really pertains to my personal encounters.

Thanks for the feedback.


Reply To Thread

Colors Smileys Quote OriginalQuote Checked Help

 

Recent Visitors: 139 All times are in CST
Anonymous Guests (139)