Forum
High stack vs low stack
|
Mause wrote
at 6:49 AM, Tuesday November 9, 2010 EST
Maybe it is just me but i have been testing it out.... from the 25 times i attacked with an 8 stack to a real lower stack of dice.. 5-6 i only won 1 move,,, its said that the rolls are random.. but after my test iam not that sure about it.. how can an 8 stack loose from a 4 stack... how many % you have to loose from it... 8% in theorie.. but in the practical way.. the winchance from 8% becomes loosechance.. Yeah thats random...
|
|
Iumentum wrote
at 8:04 AM, Thursday November 11, 2010 EST It does work, but have a problem using storage within my jquery each function, cause creasemonkey prevent it for security reasons since it thinks its some kind of event it seems. But for now i just store them in an array and writes the results to my console. Just have to cupy that out and make it into the default array if i close or refresh the browser. But don't feel like putting to much energy into this script so kept it simple and dirty so far.
|
|
jurgen wrote
at 8:46 AM, Thursday November 11, 2010 EST I like it simple and I like it dirty so keep up the good work :-)
|
|
StormLord wrote
at 10:52 AM, Thursday November 11, 2010 EST What are the points being made?
1)That the Random Number Generator is biased 2)That Ryan has deliberately coded a bias for high rolls. Both are unlikely. |
|
|
jurgen wrote
at 10:58 AM, Thursday November 11, 2010 EST for many people it's 3) That I am biased, paranoid and superstitious, thinking that I could be jinxed or cursed sometimes and that I want data to prove me wrong
|
|
|
jurgen wrote
at 10:59 AM, Thursday November 11, 2010 EST |
|
|
Iumentum wrote
at 6:47 PM, Thursday November 11, 2010 EST Didn't get to collect as much data i hoped for doing the night. Guess i slept around the wrong table. But here is a taste of the data collected so far.
Defend percentage for 8 stacks: 8v5: 4% (based on 49 rolls) 8v6: 17% (based on 52 rolls) 8v7: 36% (based on 74 rolls) 8v8: 53% (based on 958 rolls) I will collect more data again tonight! |
|
|
jurgen wrote
at 1:41 AM, Friday November 12, 2010 EST expected defend values: 52,9% for 8v8; 32,7% for 8v7; 15,6% for 8v6; 5,2%
given the margin of error resulting in the relatively low amount of data points for 8v7/6/5, these results look pretty promising I'm still curious for more data on the 8v7 results since that is one of those rolls where many people feel it defends more than it should (so far 3% more than expected, which is actually 1/10 more than it should...but you need 200 more rolls to confirm imo) |
|
|
jurgen wrote
at 1:42 AM, Friday November 12, 2010 EST 5,2% is expected defend % for 8v5
|
|
|
Iumentum wrote
at 2:59 AM, Friday November 12, 2010 EST Hopefully after been running all night and most day tomorrow the data will be more reliable. But here is a little update:
8v5: 4.23% (based on 71 rolls) 8v6: 16.47% (based on 85 rolls) 8v7: 34.91% (based on 106 rolls) 8v8: 53.29% (based on 1475 rolls) |
|
|
Iumentum wrote
at 4:39 AM, Monday November 15, 2010 EST Updated:
The page at http://kdice.com says:8v1: 0.00% (based on 124 rolls) 8v2: 0.00% (based on 150 rolls) 8v3: 0.00% (based on 157 rolls) 8v4: 0.00% (based on 152 rolls) 8v5: 5.75% (based on 174 rolls) 8v6: 15.53% (based on 219 rolls) 8v7: 33.69% (based on 282 rolls) 8v8: 53.40% (based on 3489 rolls) |
