Seigan Counter-rate Testing

[Nerji]

Went ahead and tested how zanshin effects counter rate with seigan up.
Had about ~20 or so mandies in abyssea attowha to counter against using a low damage great katana.

Melee Damage Taken
Player             Melee Dmg   Melee %   Hit/Miss   M.Low/Hi    M.Avg  #Crit  C.Low/Hi   C.Avg     Crit%
Nerji                   3289   96.85 %  4171/5014       0/26     0.20    261      0/44    9.56    6.26 %

Passive Defenses
Player             Evasion  Evasion %   Parry  Parry %   Counter  Counter %   Intimidate  Intimidate %
Nerji                 4486    42.21 %    1409  22.99 %       539    11.44 %            0        0.00 %

If this calculation is correct =1.96*SQRT(0.1144*(1-0.1144)/(4171+539)) It gives a confidence interval of +/- 0.909%

This type of counter also has an accuracy check (tested using a staff) so this falls within the expected value of 12.5*0.95 (zanshin/4)*accuracy cap

I also tested the counter rate of Unkai Kabuto +2

Melee Damage Taken
Player             Melee Dmg   Melee %   Hit/Miss   M.Low/Hi    M.Avg  #Crit  C.Low/Hi   C.Avg     Crit%
Nerji                   3627   96.64 %  3735/5550       0/23     0.61    243      0/42    6.23    6.51 %

Passive Defenses
Player             Evasion  Evasion %   Parry  Parry %   Counter  Counter %   Intimidate  Intimidate %
Nerji                 4801    45.28 %    1265  21.87 %       766    17.02 %            0        0.00 %

=1.96*SQRT(0.1702*(1-0.1702)/(3735+766)) gives a confidence interval of +/- 1.1%
Most likely the kabuto only gives 5% counter, not the ~10% stated on the wiki

[Nerji]

Tested what the counter rate on Saotome haidate +2 with third eye on is.
I fought one mandagora at a time in abyssea attowha using seigan and third eye

Melee Damage Taken
Player             Melee Dmg   Melee %   Hit/Miss   M.Low/Hi    M.Avg  #Crit  C.Low/Hi   C.Avg     Crit%
Nerji                   3294   90.35 %   158/1555      11/27    19.72      9     29/49   39.44    5.70 %

Passive Defenses
Player             Evasion  Evasion %   Parry  Parry %   Counter  Counter %   Intimidate  Intimidate %
Nerji                 1338    68.09 %     121  19.52 %       224    45.62 %            0        0.00 %

Active Defenses
Player             Shadow  Shadow %   Anticipate  Anticipate %   Retaliations  Retaliation %
Nerji                   0    0.00 %          109       21.84 %              0         0.00 %

Now we have to remove some of the counters due to seigan zanshin counters occurring when third eye is down. To find out the theoretical value of this I divide the total hits I took by zanshin/4*.95 which gives 21 zanshin counters rounded down

So the counter rate with third eye on is 203/(203+109) = 65.1% +/- 5.3%

Taking away the zanshin counter rate 11.875% it becomes ~53% most likely it's suppose to be 50%

[Byrthnoth]

Okay, a few questions about this:
1) How do you identify which counters occur with 3E down vs. up?
2) Seigan + 3E had a base counter rate without AF2+2 legs even before Zanshin-counters were added, so what is the baseline you're subtracting?

Sorry, perhaps this is apparent but I couldn't get it.

[Nerji]

Okay, a few questions about this:
1) How do you identify which counters occur with 3E down vs. up?
2) Seigan + 3E had a base counter rate without AF2+2 legs even before Zanshin-counters were added, so what is the baseline you're subtracting?

Sorry, perhaps this is apparent but I couldn't get it.

1) I calculated how many counters theoretically should have occured when third eye is down by dividing the number of hits I've taken by (1-0.11875) which is the seigan zanshin counters. I do have the logger file for the whole experiment but I wasn't sure how to go about stripping the counters when third eye was up or down, I thought this method might be okay.

2) Ah yeah you're totally right, I only subtracted the zanshin counters, I'll have to find out the base counter rate of just seigan and third eye

[Nerji]

Testing counter of seigan + third eye and using the same methodology as before

Melee Damage Taken
Player             Melee Dmg   Melee %   Hit/Miss   M.Low/Hi    M.Avg  #Crit  C.Low/Hi   C.Avg     Crit%
Nerji                   7044   89.85 %    386/774      11/25    17.25     20     28/43   36.50    5.18 %

Passive Defenses
Player             Evasion  Evasion %   Parry  Parry %   Counter  Counter %   Intimidate  Intimidate %
Nerji                  609    38.64 %     125  13.02 %       172    20.90 %            0        0.00 %

Active Defenses
Player             Shadow  Shadow %   Anticipate  Anticipate %   Retaliations  Retaliation %
Nerji                   0    0.00 %          265       31.74 %              0         0.00 %

Using the number of hits I've taken I determine how many counters there should have been due to seigan zanshin
386/(1-0.125*.95) = 438

So there should have been 52 counters from seigan zanshin when third eye wasn't up which leads us to the seigan third eye counter rate

(172-52)/(172-52+265) -0.125*.95 = 19.3% +/- 5%

So saotome haidate +2 adds ~30% counter rate to third eye

I can understand that some people might not be happy with my methodology since I calculate how many counters should have occurred when third eye was down rather then using the raw data. At best I think it at least gives us a somewhat base value to work from, I have the parsers/logs saved from when I collected the data, but I don't think it'll be of much use since it doesn't say when third eye wears off. In hindsight I should have used autoexec to echo a message

[Byrthnoth]

It's close enough for me, but given your observations I don't think you've ruled out 25% for base and +25% with the pants (which seems like a simpler solution). You have to remember that you're adding variability when you make your assumption, so your actual error is going to be substantially greater than your calculated error.

[Prothescar]

Assuming I'm not reading it wrong, the COunter Rate on Third Eye without Koga is pretty immense with the right gear, and with Koga should be nearly 100% if it doesn't follow normal COunter rules <_<

[dasva]

Yes just for clarification how exactly is all this countering stacking up? Like is it all stacking additively or not and how much is each

[Fake]

Do the AF2+2 pants need to remain equipped for the increased counter rate or only during activation?

[Nerji]

I went back to get some more data on the seigan third eye counter rate

Melee Damage Taken
Player             Melee Dmg   Melee %   Hit/Miss   M.Low/Hi    M.Avg  #Crit  C.Low/Hi   C.Avg     Crit%
Nerji                  28671   91.07 %  1505/4643       9/29    17.96     88     23/52   36.64    5.85 %

Passive Defenses
Player             Evasion  Evasion %   Parry  Parry %   Counter  Counter %   Intimidate  Intimidate %
Nerji                 3880    48.29 %     634  15.38 %       796    23.17 %            0        0.00 %

Active Defenses
Player             Shadow  Shadow %   Anticipate  Anticipate %   Retaliations  Retaliation %
Nerji                   0    0.00 %         1135       32.54 %              0         0.00 %

Using the same method as before 202.8 counters came from zanshin when third eye was down, leaving 593.2 with third eye on

593.2/(593.2+1135) = 34.3% counter rate +/- 2.1%

Remove zanshin counter rate while third eye is up from this

0.34324677 - 0.125*.95 = 22.4% +/- 2.1%

I'm inclined to agree with bryth's suggestion of 25% counter rate, my assumption on number of zanshin counters with third eye down increases the error margin and I also noticed on the parser I got slept 32 times by dream flower which would have lowered the counter rate (if I add an additional 30 counters to the spreadsheet the rate increases by ~1%)

Yes just for clarification how exactly is all this countering stacking up? Like is it all stacking additively or not and how much is each

I'm not sure yet, I'll probably test the saotome haidate +2 with unkai kabuto +2 at some point. I want to find out the activation rate on the unkai set bonus for 2 pieces at some point too

Do the AF2+2 pants need to remain equipped for the increased counter rate or only during activation?

I haven't tested if it works only on activation yet

Edit:
You can't use Saotome haidate +2 only on activation, have to be full timed for the counter bonus. I was getting a similar counter rate to just seigan and third eye

[Nerji]

Went and tested how unkai kabuto +2 and saotome haidate +2 stack in regards to counter, if additive I'd expect
0.25 (seigan third eye)
0.25 (expect value of saotome haidate +2)
0.125 (zanshin counter)
0.05 (unkai kabuto +2)

Which adds up to 67.5%

Melee Damage Taken
Player             Melee Dmg   Melee %   Hit/Miss   M.Low/Hi    M.Avg  #Crit  C.Low/Hi   C.Avg     Crit%
Nerji                  19481   88.59 %   877/3282      13/33    21.04     47     31/65   42.91    5.36 %

Passive Defenses
Player             Evasion  Evasion %   Parry  Parry %   Counter  Counter %   Intimidate  Intimidate %
Nerji                 2365    47.48 %     355  13.69 %       940    42.71 %            0        0.00 %

Active Defenses
Player             Shadow  Shadow %   Anticipate  Anticipate %   Retaliations  Retaliation %
Nerji                   0    0.00 %          384       17.16 %              0         0.00 %

Using the same method before but now taking into account unkai kabuto +2 adding another 5% countering when third eye is off we get

877/(1-(0.125+0.05)*0.95) = 1051.874 Total hits
Therefore 174.87 counters with third eye down
And we get 765.13 counters with third eye on

765.13/(765.13+384)= 66.58% +/- 2.54%

I think it's fair to conclude it's additive unless I've overlooked something

In regards to saotome haidate +2 if we do another calculation to remove zanshin and kabuto +2 counters we can find the expected value

66.58-(12.5+0.05)*0.95= 49.96%

Removing the 25% for seigan third eye counter gives us 25% +/- 2.54% for the haidate

Also something of note, I was wearing unkai sune-ate +2 for the set bonus and saw quite a few counters in my parse that did more damage then my highest crit, so I believe the unkai set bonus applies to zanshin counters

I'd like to get some advice on the best way to find out the proc rate for unkai set bonus, I was thinking about hitting on some fortifications with no multihit gear, using hasso with some extra zanshin gear but what weapon would it make it easy to identify the proc of double damage? An alternative I was thinking about was having a lot of mandagoras hit me with seigan up, would be a lot quicker to gather the data

[Yugl]

The issue with mandies method is that they can guard, so you may end up countering+set bonus, but display less than usual damage. The best way to do this is to use a DMG1 weapon with a Phalanx mule and 100% crit. Make sure their TP moves won't mess the data up and that you have capped accuracy. Avoid MNK/PLD/General Parry mobs as well.

[Nerji]

I originally wanted to test the unkai set bonus proc rate for 2 pieces but after I was only able to get 1 proc in 395 zanshin rounds I decided to determine what the full set rate was then assume the rest.

I fought lizards in abyssea Konschtat highlands with 100% crit rate (funnily enough 1 hit out of 6217 was a non crit some how) using a pitchfork, hasso for zanshin procs, no multi hit gear. I used unkai earring, and back for extra boosts in zanshin

I did notice something a bit weird on the extra attacks tab for kparser, it was reporting a 1.41% triple attack rate, I've heard that if the attack swings are too quick then kparser can pick it up erroneously. I was using sword strap, 17% gear haste, 12.5% hasso, 5% haste samba, and 15% haste spell. I was wondering what you guys thought it might be

Data:

Spoiler: show

Player               # Double Attacks    DA Rate    Perc. DA     # Triple Attacks    TA Rate    Perc. TA
Nerji                            1062    20.03 %     93.40 %                   75     1.41 %      6.60 %

Nerji
  Melee
+^        0:    1
  Melee Crits
         24:    7
         26:   32
         27:  154
         28:  459
         29:  745
         31:  863
 ^       32:  963
         33:  912
+        35: 1266
         36:  767
         55:    2
         57:    1
         58:    1
         59:    4
         61:    3
         62:    1
         63:    1
         65:    3
         66:    2
         67:    6
         68:    3
         70:    8
         71:    6
         72:    7

I'll just use the number of double attacks instead of total multi hit rate

48/1062 =4.324% +/- 1.22%

So the proc rate could be 2/3/4/5%, or 1/2/3/4%, I'm a bit inclined for the later due to my 1/395 with 2 pieces, but I'm open to what others think.

I'll test whether the set procs on seigan counters later, I'm pretty confident it will

[Byrthnoth]

1/2/3/5 is also an option, like the BRD set.