+ Reply to Thread
Page 1 of 2 1 2 LastLast
Results 1 to 20 of 34

Thread: The block meta vs the armor meta. Trying to figure out which is better.

  1. #1
    Join Date
    Dec 2008
    Posts
    1,753

    The block meta vs the armor meta. Trying to figure out which is better.

    The conversation in the shoutbox got me wondering as to what meta was better for me as a paladin tank.

    So, I decided to run some numbers on it. I created a little file in my excell clone, put the numbers from my stat screen into it, and had at it.

    Assumptions and disclaimers
    Only melee damage is taken into consideration.
    I used my own stats, buffed with Kings, Devotion Aura and Horn of Winter.
    The tooltips given in WoW are correct.
    Since I don't have information on all the penalties we get against bosses, I'm assuming a level 80 mob. (If someone could point me in the direction of a cata armor effect formula and a reliable source if we still get the old 0.6% penalties, I would appreciate it, and I'll update this post.)
    Holy shield is active.
    There's a 5% miss chance for bosses.

    Note: The meta's 5% more block value translates into 1% more block value. The base 30% block value is multiplied by 1.05 and then rounded down. Holy shield and the meta do not interact. (Go and check it if you like.)


    My stats
    13.49% dodge.
    13.27% parry.
    43.26% block.
    59.36% physical damage reduction from armor without the armor meta. (about 38k armor)
    59.78% physical damage reduction from armor with the armor meta. (about 700 more than about 38k.)
    41% block value with the block meta.
    40% block value without the block meta.



    Method used
    I took the number 100 to represent the raw damage of the melee swing.
    A) I multiplied it by 1-armor%
    This step reduces the raw damage by armor.
    B) I multiplied it by 1-(dodge%+parry%+miss%)
    This step represents the average damage reduction from avoidance.
    C) I multiplied it by (block%/ (1-avoid%) * (1 - block value%)
    First I calculated the percentage of unavoided blocked blows. Then I multiplied that percentage by the damage you take from those blows.
    D) I added (1-(block% / (1-avoid%) ) )* (what I had after step B)
    This is the damage from unblocked unavoided blows.

    The number I got from this represents the average percent of damage I take from melee.


    Conclusion
    With the armor Meta, I take 20.49% of raw damage.
    With the block Meta, I take 20.52% of raw damage.

    According to this, you take about 0,2% less damage from melee when using the armor Meta.


    Additional notes
    If your avoidance goes up, the effect of the block Meta goes up. Addin 5% to my block, dodge and parry percentages made the block meta pull ahead of the armor meta.
    Without an armor formula, I can't change the numbers for armor at the moment - metas are a tad too expensive for me to go swapping them in and out.
    I am well aware that this math ignores things such as nonphysical melee attacks, physical abilities that ignore the combat table and holy shield uptime.
    I could easily put some warrior numbers in them to see what the effect would be. Just supply me with the numbers and I'll run them.


    TL;DR: Go away, if you can't bother reading my post I don't want your opinion here.
    Last edited by Martie; 12-16-2010 at 05:30 PM.
    Quote Originally Posted by Kahmal
    ...there is no true progression for a casual anymore, just hand outs.

  2. #2
    Join Date
    Mar 2008
    Posts
    785
    The prevailing and generally right way of thought in tank gearing is that when almost equal, go with the stat that provides constant benefit. Armor will always be useful in a physical damage situation, block will not be.

  3. #3
    Join Date
    Jan 2009
    Location
    United Kingdom
    Posts
    1,644
    Yeah, have to agree with Edgewalker. Even if Block and Armor are similar, I would definitely go with Armor as it will apply consistently to all attacks. The only time I could see talking about the Block meta is if you get to a point where you are unhittable (or nearly unhittable) anyway.
    Maintainer of Rawr.ProtWarr theorycrafting tool. Feel free to PM suggestions or feature requests!

  4. #4
    Join Date
    Mar 2010
    Posts
    169
    with enough mastery im assuming the block meta will be better for warriors due to the critical block mechanic judging from the numbers martie posted though, however i dont see it being alot better and would probably stick with the armor meta for now solely for the constant reduction, at least until tanks can get full raid gear

  5. #5
    Join Date
    Dec 2008
    Posts
    1,753
    I'd love to run some numbers for critical block as well, should be easy enough if I have some stats to go from.

    I'm using the block meta right now - as a jewelcrafter, I found it rather hard to get the two yellow gems required for the armor one. I'm postponing most definite gemming decisions untill I know a little more about boss damage (since boss damage determines how much stamina I really want.)
    Quote Originally Posted by Kahmal
    ...there is no true progression for a casual anymore, just hand outs.

  6. #6
    Join Date
    Dec 2008
    Posts
    51
    Since the old block meta has a tooltip saying 1% block and that still improves your blocked amount from 30 to 31%, I'm guessing the new meta gem has a bugged tooltip or isn't working the way it should be. If it did provide a 5% additive block amount, it would be very powerful, probably overpowered even. With just 1% additive block amount, it's a very different story.

    Maybe Blizzard were originally going to make the new block meta much better than the old one, but backed off on it later? Or maybe they just copied the old tooltip from the old meta gem (5% block value, before 4.0)? Anyway, as it stands now, the armor meta seems to win in entry-level gear while the block meta will be best later due to the scaling of mastery. This may happen even before you get to the passive block cap, because healer mana matters a bit now, and therefore the average damage taken metric isn't quite as unimportant compared to plain old EH (which is still important).
    Once you get passively unhittable, the block meta will improve EH, which should be the final nail in the coffin for the armor meta.

  7. #7
    Join Date
    Jan 2010
    Posts
    127
    Has anyone done some numbers to see when exactly, the block metagem surpasses the armor one in effectiveness?
    It seems to me most people say "when we reach the block cap" but it could very well be before that. Afterall, you need quite abit of armor to equal a blocked attack in terms of mitigation, even if a block is random.

  8. #8
    Join Date
    Dec 2008
    Posts
    51
    Quote Originally Posted by Brage View Post
    Has anyone done some numbers to see when exactly, the block metagem surpasses the armor one in effectiveness?
    It seems to me most people say "when we reach the block cap" but it could very well be before that. Afterall, you need quite abit of armor to equal a blocked attack in terms of mitigation, even if a block is random.
    I did some napkin math that indicated the block meta would reduce my average damage taken by 1.3% while the armor meta would reduce it by 1%. That's as a warrior with an ilevel average in the 330s, and before buffs (which I believe would favor the block meta slightly, at least unless you're popping armor elixir and potions).

  9. #9
    Join Date
    Dec 2008
    Posts
    1,753
    The block meta would out-do the armor meta quite quickly for paladins with a good amount of mastery - for melee attacks.
    Remember that the difference is really tiny. Armor works in more situations than block, and that makes it slightly superior, even if the block meta would be better against melee only. Taking two yellow gems may be an issue for some people, tough.

    Again, give me some numbers for a warrior tank and I'll run them to see if it's different for them.
    Quote Originally Posted by Kahmal
    ...there is no true progression for a casual anymore, just hand outs.

  10. #10
    Join Date
    Mar 2010
    Location
    Staten island, NY
    Posts
    458
    i did some chicken scratch math thismorning and it appears that based on curremt armor levels, 5% block wins hands down, for at least prot warriors, dont know about paladins. even if your popping armor pots consistently, the return on block is awsome right now. even the 2% spell reflect dosnt work.
    Last edited by praetoria; 12-19-2010 at 09:32 AM.
    "Dear Santoro, Your wisdom has enlightened me. Thanks!"

  11. #11
    Join Date
    Dec 2008
    Posts
    51
    Quote Originally Posted by Martie View Post
    The block meta would out-do the armor meta quite quickly for paladins with a good amount of mastery - for melee attacks.
    Remember that the difference is really tiny. Armor works in more situations than block, and that makes it slightly superior, even if the block meta would be better against melee only. Taking two yellow gems may be an issue for some people, tough.

    Again, give me some numbers for a warrior tank and I'll run them to see if it's different for them.
    Well, the assumptions would have to be checked first to ensure reliable results. That's why I only got to napkin math.

    I'd assume that warriors block for 30% on normal blocks and 60% on crit blocks, and that the block meta would make those numbers 31% and 62%. Another, more reliable assumption would be that the armor formula here, down the page, is correct.

    Numbers for a warrior tank should be easy to get, just armory someone (not me please, I still have 2 greens equipped )

  12. #12
    Join Date
    Dec 2008
    Posts
    51
    Quote Originally Posted by praetoria View Post
    i did some chicken scratch math thismorning and it appears that based on curremt armor levels, 5% block wins hands down, for at least prot warriors, dont know about paladins. even if your popping armor pots consistently, the return on block is awsome right now. even the 2% spell reflect dosnt work.
    You *did* figure in the fact that the block meta doesn't provide the advertised 5% block value, right? Also, the 2% spell reduction we already knew would be situational/poor, since we've seen similar metas in the past. Not much has changed regarding magic damage.
    Last edited by Sakkura; 12-19-2010 at 09:56 PM.

  13. #13
    Join Date
    Mar 2009
    Location
    Rochester, NY
    Posts
    70
    You can do this analytically. See this post. Using your avoidance, block, and armor values, I get that the armor meta is 25% more effective than the block meta against blockable physical damage (i.e. melee attacks). I also find that by adding ~5% avoidance and ~5% block, the block meta pulls ahead.
    "Theck, Bringer of Numbers and Pounding Headaches," courtesy of Grehn|Skipjack.
    MATLAB TPS 4.0, Talent Spec & Glyph Guide 4.0

  14. #14
    Join Date
    Dec 2008
    Posts
    51
    Quote Originally Posted by theckhd View Post
    You can do this analytically. See this post. Using your avoidance, block, and armor values, I get that the armor meta is 25% more effective than the block meta against blockable physical damage (i.e. melee attacks). I also find that by adding ~5% avoidance and ~5% block, the block meta pulls ahead.
    Your formula only works for paladins though. I wish it was that simple for warriors.

    You have:
    D = (32572.5/(armor + 32572.5)) * (1 - avoidance - 0.6 * crit block - 0.3 * block)

    Da = (32572.5/(armor + 32572.5)) * (1 - avoidance - 0.6 * crit block - 0.3 * block)

    Db = (32572.5/(armor + 32572.5)) *(1 - avoidance - 0.62 * crit block - 0.31 * block)

    Using the mentioned assumptions. Further, the benefit from each meta is given by:

    1-Da/D = 1 - (armor + 32572.5)/(1.02 * armor + 32572.5)

    1-Db/D = 1 - (1 - avoidance - 0.62 * crit block - 0.31 * block)/(1 - avoidance - 0.6 * crit block - 0.3 * block)

    And I really don't want to get into differentiating this. WoW is also not being nice to warriors by not handing us our crit block chance in straight numbers.

  15. #15
    Join Date
    Sep 2009
    Posts
    1,900
    We went through this in the shoutbox earlier which inspired this thread. There is a crossover point where the block meta pulls ahead of the armor meta, that is when you are unhittable or have rediculously high armor reduction (like almost armor capped). This is for warriors only, as paladins have 10% more block which will change the figures and the crossover point. Critical block is a cherry on top, but doesn't really change the numbers.
    RNGesus - Saving you unreliably since BC.
    http://i689.photobucket.com/albums/v...ellvarsig3.jpg

  16. #16
    Join Date
    Dec 2008
    Posts
    51
    Well, more like a crossover line/curve or plane/surface I'd say, given that we are looking at two or three variables at least (armor, mastery, and avoidance total).

  17. #17
    Join Date
    Mar 2009
    Location
    Rochester, NY
    Posts
    70
    Yeah, you warriors have it rough. However, it's not quite as bad as you think, because your formulas aren't constructed properly.

    Your damage taken formula could be written as follows:

    Code:
    D = K/(Ar+K)*[1-Av-0.3*B*(1-C)-0.6*B*C]
      = K/(Ar+K)*[1-Av-0.3*B*(1+C)]
    Av, B, and Ar are all the same as my notation (Avoidance, Block chance, and Armor respectively). C is your critical block chance.

    Da and Db are constructed similarly:
    Code:
    Da = K/(1.02*Ar+K)*[1-Av-0.3*B*(1+C)]
    Db=K/(Ar+K)*[1-Av-0.31*B*(1+C)]
    1-Da/D is identical to paladins. 1-Db/D is very similar, but with 0.3 instead of 0.4 and B*(1+C) instead of B:

    Code:
    1-Da/D = 0.02*Ar/(1.02*Ar+K)
    1-Db/D = 0.01*B*(1+C)/[1-Av-0.3*B*(1+C)]
    Thus, the result of setting the two equal (and letting x stand for the break-even block reduction percentage like I did in my derivation) is exactly the same as that for Paladins, but with the same substitutions:
    Code:
    x = 0.02*Ar/[B*(1+C)]*[1-Av-0.3*B*(1+C)]/(1.02*Ar+K)
    The tricky part is figuring out an average value for C. That would depend on mastery, talents, and some sort of model for Shield Block usage and Hold The Line. However, you might be able to just empirically determine C from parses.
    "Theck, Bringer of Numbers and Pounding Headaches," courtesy of Grehn|Skipjack.
    MATLAB TPS 4.0, Talent Spec & Glyph Guide 4.0

  18. #18
    Join Date
    Aug 2010
    Posts
    209
    The unblocked hit is the new crushing blow...
    Or is that what we were saying when 3.0 hit?

    I agree with Kojiyama, gear for consistency if not worst case.

  19. #19
    Join Date
    Sep 2009
    Posts
    1,900
    The unblocked hit is the new crushing blow...
    At current gear unless you are in full T10 already then you are getting hit repeatedly and often.
    RNGesus - Saving you unreliably since BC.
    http://i689.photobucket.com/albums/v...ellvarsig3.jpg

  20. #20
    Join Date
    Jan 2009
    Location
    Karlsruhe/Germany
    Posts
    4,008
    Even in full T11 you are going to be getting hit quite a lot, we won't be seeing avoidance totals above 50% for quite a while yet (if ever).

+ Reply to Thread

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts