Reply
Views: 3078 | Replies: 6
[ Player Guide ] Testing 10k+ control stats difference

 [

Copy Link

]

  • Registered: 2017-07-24
  • Topics: 822
  • Posts: 939
On 2018-06-02 20:29:03Show this Author OnlyDescending Order
1# Go To

Hey guys!


So I tested this a bit today with a weak player, here is difference between his stats and my own

Screenshot_6

Screenshot_5

So it's a 11k difference in Control but 10k should be enough anyway because in theory every 1k control = 10% more chance to control fail therefore no point after 10k difference.



He was using Fire Main with sleep standard attack and immobile mystery and I was using LM with only heal standard so the duels can last 10 rounds always (I know it makes LM immune but I payed a lot attention to the text and it still says "control failed" even if LM is immune)


testing 6 duels with 14 standard attack(oboro clone) + 4 mystery = 108 control fails

Screenshot_8

Screenshot_9

So my results were ALL control fail the 108 attempts, however I do believe there is a cap of 99% when it comes to control and that I just wasn't lucky to get a control even once.



EXCEPTION TO THE CONTROL STATS:

Screenshot_10

Screenshot_11

Itachi Uchiha is the only ninja that can control always and never fail , no matter the control stats.


He managed to control me 10/10 times even though the difference in control was 11k+ making me perma controlled in the fight

  • Registered: 2017-07-24
  • Topics: 163
  • Posts: 2514
On 2018-06-02 21:47:14Show this Author Only
2#

the number of tries was too low to know for sure the cap is not 95% but 99%. You should run around 500 tries to know it with a more solid basis.

  • Registered: 2017-07-24
  • Topics: 6
  • Posts: 62
On 2018-06-03 10:41:38Show this Author Only
3#
  • Garv On 2018-06-02 21:47:14
  • the number of tries was too low to know for sure the cap is not 95% but 99%. You should run around 500 tries to know it with a more solid basis.

been a long time since i did statistics but there's a 1% chance to get 108/108 fails if the probability was 96%.

108 is a good sample set to say there's a good chance that the underlying probability is between 98% - 100%...


anyway thanks for the testing icex

  • Registered: 2017-07-24
  • Topics: 45
  • Posts: 742
On 2018-06-03 19:08:48Show this Author Only
4#
  • renholder On 2018-06-03 10:41:38
  • been a long time since i did statistics but there's a 1% chance to get 108/108 fails if the probability was 96%.

    108 is a good sample set to say there's a good chance that the underlying probability is between 98% - 100%...


    anyway thanks for the testing icex

Statistics also said that you can fail 108/108 with 99% chance to success.


Since this game has no way to official datamine the API (unless you want to h4ck it), i doubt that we should care if the chance is 96 or 99...




This post was last edited by Shryke on 2018-06-03 19:08:59.
  • Registered: 2017-07-24
  • Topics: 822
  • Posts: 939
On 2018-06-04 01:16:02Show this Author Only
5#

Yeah guys, no way for me to exactly say what the cap would be, but I choose 99% because it sounds more likely from my testing

  • Registered: 2017-07-24
  • Topics: 163
  • Posts: 2514
On 2018-06-04 20:46:05Show this Author Only
6#
  • ICExx On 2018-06-04 01:16:02
  • Yeah guys, no way for me to exactly say what the cap would be, but I choose 99% because it sounds more likely from my testing


Reply
Quicky Post
Reply

Log in in order to Post. | Register