nkelsch wrote:talljosh85 wrote:I would like a set, I wish I could find the thread, but a stats professor at ASU (I think) did a proper study on varying
D6 that were mass produced, and compared them to casino dice. The difference was statistically significant, enough that I would be interested in a set. I remember him pointing out not to use the standard dice for leadership checks and the such, because it would be like cheating if using the better dice for hits, saves, etc.
Edit, here it is:
http://www.dakkadakka.com/wiki/en/That's_How_I_Roll_-_A_Scientific_Analysis_of_Dice
The only flaw with that article is that they are attempting to prove an agenda and the results are fabricated as it is statistically impossible to roll 29% 1s without the dice being visibly deformed.
I did a sample of about 600 throws myself with
GW dice, and did not find any signifant statistical anomalies. Granted, the sample was pretty small, but the results creeped closer to median as my test progressed and I sure don't believe that adding another 5000 rolls would have resulted number of 1's magically double. In fact, in my test, 1 was the least common result.