Mephisto Wrote:No, I don't have any kind of fallacy in my head. If 6/20 of the players are mafian, and you have to run the RNG 6 times because the first 5 players you hit were dead, a simple probabilistic calculation will show you that by the 6th roll the chances of hitting a mafian are high.
That's the very definition of gambler's fallacy, so I have no idea what you're thinking.
We'll play it your way, anyway. I generated a list of 100 numbers in the 1-20 range. Using the playerlist as a guide, 1 is Holypie, 13 is Synthesist, 15 is z0mgc00k13 and 19 is dante9898. This is the setup for Day 8, the last day, where Pie had a 25% chance of dying (theoretically speaking). Assuming you're right, after 100 rolls Pie should have a high chance of dying, yes?
![[Image: gamblerfail.png]](http://dl.dropbox.com/u/4085446/Final%20Destination/gamblerfail.png)
There are 80 instances of me having to roll again, and 20 instances of someone dying. Within those 20 instances, townies died in 16 instances and Pie died in 4. That shows a 20% chance of killing a player, and WITHIN THAT 20%, a 20% chance of Pie dying.
Don't get me wrong, obviously Pie having less chance of dying was random too. That's the nature of the generator. But what I'm trying to say is that you really are using gambler's fallacy.
Your move.