Wildstar,
I don't play. To me, it would be essentially throwing away $100 that I could use for other things that have a much greater chance than 1/1000 for achieving my ends.
Yes, this is a demonstration that the concept of expected value is severely flawed.
Being inherently lazy, I'm copying part of one of my emails below --
"Say that you could win $100M by correctly predicting the number drawn out of 000-999 of a thousand numbered balls in a barrel. It costs you $100 to play and the expected value is $100M X .001 or $100K, a 1000 to 1 return on the $100 cost. Do you play?
There is no certain answer, but there is a big difference between having only a single trial and having 10,000 trials, if you could afford them. In a single trial, you will just be out $100 with a probability of 99.9%. With 10,000 trials, you will win the $100M with a probability of 99.995%. To reach a win probability of 50%, you need 692.8 trials.
In addition, the subjective marginal utility, even if it were possible to measure it, would not be 10 times greater for a $100M prize than
it would be for a $10M prize, as the law of diminishing marginal utility will kick in.
In the end, the subjective marginal utility of lotteries and the like probably has less to do with prize sizes and probabilities and more to do with entertainment value and dreams."
Regards, Don |
