+ Reply to Thread
Page 1 of 2 1 2 LastLast
Results 1 to 20 of 36

Thread: Dodge vs Parry

  1. #1
    Join Date
    Feb 2009
    Posts
    12

    Dodge vs Parry

    I'm not sure if this has been addressed but here it goes.

    I do understand that you get more a more dodge percentage for every rating compared to parry rating but considering that DKs get 4% from weapon enchants and another 10% from blood talents and more rating from +STR you get, is it better to stack parry gems over dodge ones?

    The argument for this is basically because of a general rule someone taught me when trying to gear up a tank.
    It's better to stack up 1 avoidance stat rather than 2 because the chance of either happening doesn't stack.

    basically

    Wrong Equation
    Dodge + Parry = Chance of Dodging or Parry

    Right Equation (or i think it is. someone can correct me. but i know for sure the equation above is 100% wrong)
    Dodge + Parry - (Dodge*Parry) = Chance of Dodging or Parry

  2. #2
    Join Date
    Jul 2007
    Posts
    16,438
    your basis for avoidance is wrong.

    your overall avoidance chance is chance to miss + dodge + parry. There is no diminishing effect between the two of them.

    Dodge > Parry even for DKs. You just will naturally have higher parry than other tanking classes because of your innate parry gains.

    READ THIS: Posting & Chat Rules
    Quote Originally Posted by Turelliax View Post
    I will never be a kaz.. no one can reach the utter awesomeness of you.
    http://i.imgur.com/3vbQi.gif

  3. #3
    Join Date
    Mar 2008
    Location
    Bellingham, WA, USA
    Posts
    474
    Dodge and Parry are independent events. Each one's chance is completely unaffected by the other under normal circumstances. Your total avoidance is the sum of your miss + dodge + parry chance. Increasing one of them does not reduce the others unless some how you are already unhittable in which case youd stay unhittable with any more gains.

    Because each stat increases slower (due to diminishing returns) its better to split between the two.

    If your wanting pure avoidance you'll be better off with gemming dodge rating rather then parry, unless you reach very very high levels of dodge where parry might start to become better.
    Galushi - 85 Dwarf Warrior - Shadowsong US Alliance.
    Branigan - 85 Goblin Warrior - Kilrogg US Horde.

  4. #4
    Join Date
    Jul 2007
    Location
    Canadia
    Posts
    3,523

  5. #5
    Join Date
    Feb 2009
    Posts
    12
    ok, so after tracking down a few friends of mine who are math majors and a few of my old math professors i came to the conclusion that regemming my gear with dodge rating is defintely better... by less than .4%

    yes.. point four percent.

    I would regem my gear but considering that it's so miniscule i've just decided that my healers will just have to suffer. qq more healers ^^

    by the way, the avoidence math miss+parry+dodge is technically wrong.

    its miss+parry+dodge-(miss*parry*dodge) but the (miss*parry*dodge) turns out to be some ridiculously small number that it's probably ignored in the simplified version of avoidence.

    anyways, thanks to those who replied because it made it clearer for me to make my calculations.

  6. #6
    Join Date
    Jul 2007
    Location
    Canadia
    Posts
    3,523
    You're right that the difference is less than 1% - more like you'll use dodge gems instead of parry gems as you replace gear, definitely not worth regemming for at this point.

    I'm curious where you get the (D*P*M) term from - it makes no sense in a table based combat system.
    Got a question? Try here: Evil Empire Guides and here: Tankspot Guides and Articles Library first!

  7. #7
    Join Date
    Feb 2009
    Posts
    12
    probability theory from precalc ^^

  8. #8
    Join Date
    Apr 2008
    Posts
    7
    Quote Originally Posted by Aradvanced View Post
    probability theory from precalc ^^
    As Satrina says, the model in WoW is table based, so a roll to 'hit' could look like:

    0000-0900 : miss
    0901-2900 : dodge
    2901-4700 : parry
    4701-9999 : hit

    you dont calculate each avoidance seperately, in any order.

  9. #9
    Join Date
    Feb 2008
    Location
    Aarhus, Denmark
    Posts
    73
    Quote Originally Posted by Aradvanced View Post
    ok, so after tracking down a few friends of mine who are math majors and a few of my old math professors i came to the conclusion that regemming my gear with dodge rating is defintely better... by less than .4%

    yes.. point four percent.

    I would regem my gear but considering that it's so miniscule i've just decided that my healers will just have to suffer. qq more healers ^^

    by the way, the avoidence math miss+parry+dodge is technically wrong.

    its miss+parry+dodge-(miss*parry*dodge) but the (miss*parry*dodge) turns out to be some ridiculously small number that it's probably ignored in the simplified version of avoidence.

    anyways, thanks to those who replied because it made it clearer for me to make my calculations.
    This is because in normally in Probability theory we are normally interested in those calculations where they are done in turns.

    Much like saying "What are the chance to get a 6 when you roll a Dice 3 times" versus "What are the chance of getting 6, 5 or 4 rolling the dice ones"

    In the first situation, you would have 1/6 chance per roll, thereby increasing your chance to get a 6 for each roll you do. Much like a NPC would increase it's chance to hit you for each time a roll was done.

    But in the second situation you have one roll, and 3/6 chance of hitting either one of those numbers.

    Wow works similar to the second example... Most properly and most certainly because it saves rolls, thereby saves computations, thereby saving time. All in all a performance choice...

    In Probability theory, we don't normally concern our self to much with this approach, because it's not really interesting (or it is as simple as it is, thereby not something we need to learn there, that's something more of basic math) . But we always use them for the calculations because it's very basic.

    (This is kinda an ambiguous example, but is the best way I could think of >.<)
    Last edited by dotJEM; 02-03-2009 at 05:49 AM.

  10. #10
    Join Date
    Oct 2008
    Posts
    4,930
    All discussions of hit table and stat value aside, in current content I haven't gemmed for dodge or parry.

    There's plenty of dodge and parry available on gear (more dodge than parry, though that's not surprising). As DK's we'll get piles of Parry Rating from Strength. I'm generally more interested in getting keeping Def capped, getting strong Stam values, and keeping respectable levels of Exp and Hit rating. I suspect that gems and enchants at this point, provided you have them, will make for a relatively subtle distinction from one choice to the next, more likely than not out-shone by player skills/moves.

    Personally, if I were to try and boost avoidance, I think I'd favor Agi for the joy of dodge+crit+armor, over pure dodge rating. Hardly a "wrong" choice available though, except spellpower. >.>
    The (Old) Book on Death Knight Tanking
    The New Testament on Death Knight Tanking
    -----------------------------------------
    Quote Originally Posted by Horacio View Post
    Who f-ing divided by zero?!?

  11. #11
    Join Date
    Jul 2007
    Location
    Canadia
    Posts
    3,523
    Right. It's been a long time since statistics for me, but it looks like the OP is applying the addition law { P(A U B) = P(A) + P(B) - P(A ∩ B) }, which is fine if there is a chance if A and B both happening. This can't happen in a table system; your result is exactly one of {dodge, parry, miss, hit} (add in block, critical hit, crushing blow as necessary) - but they are all mutually exclusive events

    I believe what you are looking for is simply sigma additivity { P(A U B) = P(A) + P(B) } since all events are mutually exclusive.
    Got a question? Try here: Evil Empire Guides and here: Tankspot Guides and Articles Library first!

  12. #12
    Quote Originally Posted by Satrina View Post
    Right. It's been a long time since statistics for me, but it looks like the OP is applying the addition law { P(A U B) = P(A) + P(B) - P(A ∩ B) }, which is fine if there is a chance if A and B both happening. This can't happen in a table system; your result is exactly one of {dodge, parry, miss, hit} (add in block, critical hit, crushing blow as necessary) - but they are all mutually exclusive events

    I believe what you are looking for is simply sigma additivity { P(A U B) = P(A) + P(B) } since all events are mutually exclusive.
    I believe what satrina says here is correct, there is not a residual probability because they are separate events in the same number space, and due to this mutual exclusivity the probability is nothing other than the probability of each individual event in the probability space added.

    Now if you wanted to get into the math of asking "is the random number generator in WoW really random" then you'd have to add all sorts of wonderful little nick nacks onto the calculation because it's not truely random (but from an engineer's perspective it's such a small margin of error, why waste your time?).

    A mages chance to crit on the other hand is not mutually exclusive due to the dual roll nature (assuming I'm remembering how mage crits work correctly). You roll to see if you hit, then you roll to see if you crit, so if you miss you cant crit, and thus it would be: { P(H U C) = P(H) + P(C) - P(H ∩ C) }. However all melee run off a 1 roll system for hits/miss/dodge/parry/block and thus all outcomes are mutually exclusive.

  13. #13
    Join Date
    Jul 2007
    Location
    Canadia
    Posts
    3,523
    Quote Originally Posted by shushikiary View Post
    Now if you wanted to get into the math of asking "is the random number generator in WoW really random" then you'd have to add all sorts of wonderful little nick nacks onto the calculation because it's not truely random (but from an engineer's perspective it's such a small margin of error, why waste your time?).
    I'd imagine WoW uses a cryptographic RNG anyway, so the answer is most likely yes. (For all we know, they use a hardware RNG on each server)

    However, if they were using a simple pseudorandom RNG... If you take a PRNG and run its sequences for N consumers with thousands of consumptions per second in a completely random order - is each consumer's result still pseudorandom, or does the pseudorandom only apply over the sequence and each consumer's results are now truly random?
    Got a question? Try here: Evil Empire Guides and here: Tankspot Guides and Articles Library first!

  14. #14
    Join Date
    Jul 2007
    Posts
    16,438
    whoa whoa whoa, mages have a 2 dice roll? why would the combat system be different for 1 class?

    READ THIS: Posting & Chat Rules
    Quote Originally Posted by Turelliax View Post
    I will never be a kaz.. no one can reach the utter awesomeness of you.
    http://i.imgur.com/3vbQi.gif

  15. #15
    Join Date
    Jul 2007
    Location
    Canadia
    Posts
    3,523
    All DPS casters use a two-roll system because of partial resists.
    Got a question? Try here: Evil Empire Guides and here: Tankspot Guides and Articles Library first!

  16. #16
    Join Date
    Jul 2007
    Posts
    16,438
    ah, gotcha, that's a different mechanic for me to learn. Got any good reads Sat?

    READ THIS: Posting & Chat Rules
    Quote Originally Posted by Turelliax View Post
    I will never be a kaz.. no one can reach the utter awesomeness of you.
    http://i.imgur.com/3vbQi.gif

  17. #17
    Join Date
    Jul 2007
    Location
    Canadia
    Posts
    3,523

  18. #18
    Join Date
    Apr 2008
    Posts
    51
    I have a DK who keeps telling me that parry is good for a host of crazy reasons. Mostly I think it's a lack of understanding of how DR works, but then he says crazy things like "Parry improves my threat".
    Is there any truth to this threat argument or can it's effect be seen as negligible in most/all models?

  19. #19
    Join Date
    Feb 2009
    Posts
    1
    Quote Originally Posted by Ioseb View Post
    I have a DK who keeps telling me that parry is good for a host of crazy reasons. Mostly I think it's a lack of understanding of how DR works, but then he says crazy things like "Parry improves my threat".
    Is there any truth to this threat argument or can it's effect be seen as negligible in most/all models?
    When you parry an attack, the remaining time on your current swing is reduced by 40% of your weapon speed, unless this would result in a reduction to less than 20% of your swing time remaining
    This results in an average of .24 extra swings per parry; thus parry favors slower, higher damage weapons. source: Tankspot
    Note that with a slow weapon and fast incoming attacks, it is possible to gain multiple speed reductions.
    Your swing timer is always running any time you are able to parry (you can't parry during spell-casting, stun, etc), and since it resets on a predictable basis there are several addons that can display it. Using such a timer and monitoring incoming events (LD50_Abar addon), you will note that parry affects only your current swing, and only the time remaining when the parry occurs.
    In order to display parry events accurately, you must know the correct swing speed (including haste effects) and the elapsed swing time.



    Formulas:Parry - WoWWiki - Your guide to the World of Warcraft


    Yes, this is true in theory, but not worth stacking parry for, because there is no way you can have threat issues at the moment.

  20. #20
    Join Date
    Nov 2008
    Location
    Australia, NSW
    Posts
    619
    Quote Originally Posted by Vethek View Post
    When you parry an attack, the remaining time on your current swing is reduced by 40% of your weapon speed, unless this would result in a reduction to less than 20% of your swing time remaining
    This results in an average of .24 extra swings per parry; thus parry favors slower, higher damage weapons. source: Tankspot
    Note that with a slow weapon and fast incoming attacks, it is possible to gain multiple speed reductions.
    Your swing timer is always running any time you are able to parry (you can't parry during spell-casting, stun, etc), and since it resets on a predictable basis there are several addons that can display it. Using such a timer and monitoring incoming events (LD50_Abar addon), you will note that parry affects only your current swing, and only the time remaining when the parry occurs.
    In order to display parry events accurately, you must know the correct swing speed (including haste effects) and the elapsed swing time.



    Formulas:Parry - WoWWiki - Your guide to the World of Warcraft


    Yes, this is true in theory, but not worth stacking parry for, because there is no way you can have threat issues at the moment.
    Not just threat issues, the DR on parry is such that it's almost pointless stacking it. Have a look at Satrina's article on Diminishing Returns, the best way to fully appreciate the information presented is to throw it into a spreadsheet and draw your own conclusions from there:

    http://www.tankspot.com/forums/f63/4...avoidance.html

+ 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