Oh dear, I seem to have kicked open a bit of a can of worms here...
@Direwolf: I apologise if I have caused offence and I certainly didn't intend to insult you or accuse you of cheating. Like I said, don't take it personally! That said, I am still extremely sceptical.
I was only including physical dice rolls in my estimation, not computer simulations. Even assuming as many as 10 million hardcore gamers that have made 100,000 d6 rolls each, that's "only" one trillion dice rolls, relative to the slighly more than a 100 trillion that are needed, statistically, to give 18 6s in a row. I imagine that several billion people in the developing world very rarely roll dice, and that the vast majority of people in the western world only do so occasionally (maybe upto a thousand times) in their life. Anyway, it was only an attempt to illustrate the extreme improbability of the outcome.
What was your reaction when you rolled the 18 consecutive 6s? Did you not stop to consider whether the die was sound? If you kept on rolling 6s, at what point would you begin to question the randomness of the outcome? Can you be absolutely sure that no-one you have ever gamed with might have used a loaded or crooked dice?
@Fruny: I believe you and yes, that particular outcome has an extremely low probability, as does any random sequence of 19 d6s that might turn up, and which could not have been predicted in advance. One should not mistake probability for luck - in this context a 6 is the best outcome for each roll.
@ladyphoenix: Even if a d6 is rolled many thousands of times, the probability of getting 18 6s in a row is still vanishingly small.
I find this an interesting phenomenon... Of course, extremely unlikely and strange sets of rolls do sometimes occur, and it's plausible that a gamer with years of experience may have witnessed one-in-a-million-type events. Equally, it's plausible for all kinds of bias to affect these extraordinary rolls, from the slightly dodgy rolling action, overly rosy or selective recollection of events, and fudging, to outright cheating along the lines of loaded dice or secret rerolling. As I say, I've seen one player use a crooked d10. I've also seen another roll dice without any announcement - if it was good he would announce it as "the roll"; otherwise, he was just idly exercising his hand. Under such circumstances, lucky becomes more likely. For every order of magnitude in a probability above 1:1,000,000 my scepticism grows in proportion. Extraordinary claims require extraordinary proof.[/i]
).
None of us play with perfect casino dice 99.9% of the time. I certainly don't, even if I have 2 laser cut dice that I use for some games (10 USD a piece they costed me, the buggers...). I certainly have no D20 that is perfectly cut. Statistics might be biassed due to us not using perfect dice, but I fail to see why you can question the validity of the rolls made by other gaming group. It is quite flippant, really.
