Item Search
     
BG-Wiki Search
+ Reply to Thread
Results 1 to 12 of 12

Thread: Retaliation Testing     submit to reddit submit to twitter

  1. #1
    BG Content
    Join Date
    Jul 2007
    Posts
    21,133
    BG Level
    10
    FFXI Server
    Lakshmi
    Blog Entries
    1

    Retaliation Testing

    I found some old data on my computer from god-only-knows when. They all came out to about a 40% Retaliation Rate. Anyone else have data on this?:
    2808 hits taken, 95% Retaliation up, 95% hit rate, 1052 retaliations:
    1052/(2808*.95*.95) = ~41% +/- ~2%

    It's not really a binomial, but we can pretend it is I think. As long as we're willing to accept the 95% Acc on Retaliations and that percentage with Retaliation active represents the percentage of hits with Retaliation up. IIRC, this was a 480 delay weapon if that matters. I'm pretty sure this was from breaking Gondo-Shizunori against mandies in the the boyahda tree.

    I guess I should note that this is probably a lowballed estimation of the proc rate, because it assumes I was always facing the Mandy.

  2. #2
    An exploitable mess of a card game
    Join Date
    Sep 2008
    Posts
    13,258
    BG Level
    9
    FFXIV Character
    Gouka Mekkyaku
    FFXIV Server
    Gilgamesh
    FFXI Server
    Diabolos

    I know there was a test of Retaliation + Counterstance somewhere, so if the counter rate on that were ~70% (1-(0.6*0.5)), we could give that additional support.

  3. #3
    Masamune
    Guest

    Retaliations are 100% accuracy always, not 95%.

  4. #4
    BG Content
    Join Date
    Jul 2007
    Posts
    21,133
    BG Level
    10
    FFXI Server
    Lakshmi
    Blog Entries
    1

    Quote Originally Posted by Masamune View Post
    Retaliations are 100% accuracy always, not 95%.
    Retaliations are assumed to be dependent on Hit rate like Counter. Counters and Retaliations that miss are not shown.

  5. #5
    First invited, last in the zone.
    Join Date
    Sep 2008
    Posts
    1,449
    BG Level
    6
    FFXI Server
    Lakshmi

    Quote Originally Posted by Masamune View Post
    Retaliations are 100% accuracy always, not 95%.
    Are you sure you just don't get a message when you miss a retaliation? I mean, absent perfect counter, counters are "100% accurate", in that you'll never see a counter missed message.

  6. #6
    TSwiftie
    Join Date
    Oct 2005
    Posts
    1,920
    BG Level
    6
    FFXI Server
    Fenrir

    Quote Originally Posted by Foldypaws View Post
    Are you sure you just don't get a message when you miss a retaliation? I mean, absent perfect counter, counters are "100% accurate", in that you'll never see a counter missed message.
    Should be easy to test /w blind pots. If someone doesn't get around to testing it, I can get it done this weekend

  7. #7
    Sea Torques
    Join Date
    Dec 2007
    Posts
    582
    BG Level
    5

    Quote Originally Posted by Byrthnoth View Post
    I found some old data on my computer from god-only-knows when. They all came out to about a 40% Retaliation Rate. Anyone else have data on this?:
    2808 hits taken, 95% Retaliation up, 95% hit rate, 1052 retaliations:
    1052/(2808*.95*.95) = ~41% +/- ~2%
    How do you know that 5% of hits taken are without Retaliation up?

  8. #8
    BG Content
    Join Date
    Jul 2007
    Posts
    21,133
    BG Level
    10
    FFXI Server
    Lakshmi
    Blog Entries
    1

    The Buff Active % tab in your parser. I attempted to reactivate as soon as it went down. I also kept track of the number of swings that got through between reactivations (wear off and reactivation message) and it comes out to within 1% of the same numbers.

  9. #9
    Masamune
    Guest

    Wb CDF.

    Regarding my knowledge about retaliation, it is based from all the parses i did on Lesser Colis for pDIF tests. Litterally ALL with 0 exception shown 100% retaliated hits landed. There is no message not even an animation when retaliation don't "trigger".
    In comparison, a counter has 2 checks: one for "triggering" and then one for "landing", the 2nd one showing a message (damage or missed) while 1st one just shows an animation if triggering check succeeded.
    I think Retaliation only has one check, the one for triggering (2nd check being always damage).

    On other hand, retaliation rate hasnot been rigorously tested (compared to currently used accuracy) in same vein as counter rate testings.

    I'm not sure if i could extract some exploitable data from all my Lcolis parses to pin down Retaliation Rate vs. HitRate ? (if there is indeed any relation)

  10. #10
    BG Content
    Join Date
    Jul 2007
    Posts
    21,133
    BG Level
    10
    FFXI Server
    Lakshmi
    Blog Entries
    1

    Yeah, if you had less than 95% hit rate and are confident Retaliation was active and you were being hit from the front. Just go through and divide your Retaliation rates by your hit rates and see if the number you get is more consistent than your retaliation rates without the correction.

    Alternatively, use a Blinding Potion and go melee a DC monster with Retaliation up. If you don't retaliate about half the time, there's an accuracy check.

  11. #11
    TSwiftie
    Join Date
    Oct 2005
    Posts
    1,920
    BG Level
    6
    FFXI Server
    Fenrir

    Quote Originally Posted by Masamune View Post
    In comparison, a counter has 2 checks: one for "triggering" and then one for "landing", the 2nd one showing a message (damage or missed) while 1st one just shows an animation if triggering check succeeded.
    "Perfect Counter" and "Counter" are different.
    A "missed" counter will only show in the logs when using the JA "Perfect Counter" and is not the same as a missed normal counter. When Perfect Counter misses, you still take no damage and you are able to cast a spell through it.
    A miss from the counter JA doesn't show at all in the logs, and you take damage. If you were parse the counter trait by itself, you would get a lower % than expected due to the accuracy check.

  12. #12
    TSwiftie
    Join Date
    Oct 2005
    Posts
    1,920
    BG Level
    6
    FFXI Server
    Fenrir

    Quote Originally Posted by Kirschy View Post
    Should be easy to test /w blind pots. If someone doesn't get around to testing it, I can get it done this weekend
    Got around to testing this.
    I used Lv90 WAR/SAM /w PDT Gear + Earth Staff. Regen Atma + Phalanx II + Defender + Retaliation. I did ~6min for each test. (Blind Pot + Retaliation are 3min each.) Blind Pot/Retaliation were done for a few seconds in between, but I don't think it mattered for this test. I used 5 Mandys in Attowah.

    Synopsis:
    Normal: Acc-98.48%, Retaliation-44.2%

    Blind Pot: Acc-22.67%, Retaliation-11.6%

    Accuracy check takes place in Retaliation.

    Normal:
    Code:
    Melee Damage
    Player            Melee Dmg   Melee %   Hit/Miss    M.HR %   M.Acc %  
    Kirschy                3014   12.78 %      51/15   77.27 %   98.48 %
    
    Other Physical Damage  (Counterattacks and Retaliations)
    Player            CA.Dmg  CA.Hit/Miss  CA.Low/Hi  CA.Avg   Ret.Dmg  Ret.Hit/Miss  
    Kirschy                0          0/0        0/0    0.00     20222         338/0
    
    Melee Damage Taken
    Player             Melee Dmg   Melee %   Hit/Miss 
    Kirschy                   32   62.75 %     765/52
    Blind Pot:
    Code:
    Melee Damage
    Player            Melee Dmg   Melee %   Hit/Miss    M.HR %   M.Acc %
    Kirschy                 766   12.88 %      15/60   20.00 %   22.67 %
    
    Other Physical Damage  (Counterattacks and Retaliations)
    Player            CA.Dmg  CA.Hit/Miss  CA.Low/Hi  CA.Avg   Ret.Dmg  Ret.Hit/Miss
    Kirschy                0          0/0        0/0    0.00      5064          99/0
    
    Melee Damage Taken
    Player             Melee Dmg   Melee %   Hit/Miss
    Kirschy                   21   80.77 %     851/42
    Just to note, both times I had 5% and 5.5% parry rates. (My parry is like ~290). I want your Parry%, Byrth