[ExI] any exact copy of you is you + universe is infinite = you are guaranteed immortality

Stathis Papaioannou stathisp at gmail.com
Tue Jun 19 13:29:51 UTC 2007


On 17/06/07, Stathis Papaioannou <stathisp at gmail.com> wrote:

> There's an easier, if less immediately lucrative, way to win at gambling if
> the MWI is correct. You decide on a quick and certain means of suicide, such
> as a cyanide pill that you can keep in your mouth and bite on if you should
> so decide. You then place your bet on your game of choice and think the
> following thought as sincerely as you possibly can: "if I lose, I will kill
> myself". Most probably, if you lose you'll chicken out and not kill
> yourself, but there has to be at least a slightly greater chance that you
> will kill yourself if you lose than if you win. Therefore, after many bets
> you will more likely find yourself alive in a universe where you have come
> out ahead. The crazier and more impulsive you are and the closer your game
> of choice is to being perfectly fair, the better this will work.

Well, I've done some calculations assuming a perfectly fair game where
you bet a dollar and lose or win a dollar with equal probability. Let
x be the probability that you will carry through and kill yourself if
you lose a game and let n be the number of games played. Using a
normal approximation of the binomial distribution (mean = np, variance
= np(1-p), where p is the probability of a win), the value of n needed
to ensure that that the expected number of wins will be one SD greater
than the break even point is (4-4x)/x^2.

This result is quite discouraging. If x=0.1 (i.e. there is a 0.1
probability that you will kill yourself if you lose), you need to play
360 games, after which you will almost certainly (1-0.9^360) be dead
if you are wrong about MWI. If x is lower the numbers are even worse.
You may as well go with John's method.


-- 
Stathis Papaioannou



More information about the extropy-chat mailing list