Conquer Club

Posted: **Mon Jul 03, 2006 3:23 pm**

meh, from a strictly statistical viewpoint, you wouldn't need hundreds.

Posted: **Mon Jul 03, 2006 3:26 pm**

Meh, from strictly a mass believability stand point, you would need hundreds.

--Andy

Posted: **Mon Jul 03, 2006 3:28 pm**

Pilate wrote:meh, from a strictly statistical viewpoint, you wouldn't need hundreds.

no strictly from a statistical viewpoint you would need as many as is possible to work with

Posted: **Mon Jul 03, 2006 3:38 pm**

Telvannia wrote:
Pilate wrote:meh, from a strictly statistical viewpoint, you wouldn't need hundreds.

no strictly from a statistical viewpoint you would need as many as is possible to work with

That just decreases your error size and increases your confidence level.

Really, how do you guys think real researchers work? Most studies do not use hundreds and thousands of samples.

Posted: **Mon Jul 03, 2006 3:47 pm**

It'd be hard to say anything with just 50 rolls, unless the deviation was very pronounced.

Posted: **Thu Aug 24, 2006 2:27 pm**

is that GreaseMonkey script available for d/l?

Posted: **Thu Aug 24, 2006 4:48 pm**

Pilate wrote:The dice are not purely random in a statistical sense.

There is a 3.6 % chance (36 in 1000) of losing 3 "3 vs 1" rolls in a row. This is significantly slower than the actual chances in CC.

good point

Posted: **Thu Aug 24, 2006 5:27 pm**

http://www.hostingphpbb.com/forum/viewt ... ionofglory

Dice are actually biased for the attacker when attacking with three dice, and Pilate, no there is a 36% chance, not a 3.6% chance of losing when attacking 3 on 1.

That's from Wikipedia.

Posted: **Thu Aug 24, 2006 6:56 pm**

Red Army wrote:http://www.hostingphpbb.com/forum/viewtopic.php?t=15&mforum=scionofglory

Dice are actually biased for the attacker when attacking with three dice, and Pilate, no there is a 36% chance, not a 3.6% chance of losing when attacking 3 on 1.

That's from Wikipedia.

There is a 3.6 % chance (36 in 1000) of losing 3 "3 vs 1" rolls in a row.

Posted: **Thu Aug 24, 2006 7:04 pm**

Mrs. Mad, what would be really cool is if you could publish you Luck analyser and maybe have a eset button and stuff, then if you have bad dice, you can back it up on your own.

Posted: **Thu Aug 24, 2006 9:45 pm**

I don't know anything about programming, but it seems like if the numbers for the defending and attacking dice are deteremined using the same method, the outcome, even if not the same as real dice, wouldn't favor the defender or attacker. Can someone explain why this isn't true?

Posted: **Thu Aug 24, 2006 10:12 pm**

Red Army wrote:http://www.hostingphpbb.com/forum/viewtopic.php?t=15&mforum=scionofglory

Dice are actually biased for the attacker when attacking with three dice, and Pilate, no there is a 36% chance, not a 3.6% chance of losing when attacking 3 on 1.

That's from Wikipedia.

wtf is that your own site of risk like cc and waw?

Posted: **Fri Aug 25, 2006 5:00 pm**

MrsMad wrote:luck analyzer coming soon.....

How do you read/write in a file with javascript? I've searched the web and there isn't a standard. Something that might work on Firefox might not work on IE or another version of Firefox.

Posted: **Fri Aug 25, 2006 5:49 pm**

shes given up with this dice analysiser... I pm'd her as I'm more interested than anyone on this site to find out the truth

Posted: **Fri Aug 25, 2006 6:04 pm**

Could you share the info? I'm interested in this and if she has an easy way of doing this I'd like to know.

Posted: **Sat Aug 26, 2006 3:52 am**

Pedronicus wrote:shes given up with this dice analysiser... I pm'd her as I'm more interested than anyone on this site to find out the truth

aw im pretty interested...i would love to see how i somehow managed to get a 0.1% chance of losing

Conquer Club

The dice: an experiment