Jun 15, 2006 22:10
While the relative frequency approach depends upon the fact that an experiment can be repeated under identical conditions, probabilities can be assigned to other situations by treating them as a rational measure of belief.
Thus if one truly believes that an event, E has a probability of occuring one in two times, and we assume that that is true unless proven otherwise, then p = ½. Assuming that E is something as trivial as a coin toss, then it would be safe to assume that p = ½.
Poker and blackjack, however, differ slightly from the example of the coin toss. There are 52 cards in a deck, and depending on how a deck is divided, the fractions can vary. Consider, too, that casinos often use multiple decks and at that, only a part of one. However, it can be done, as the students at MIT have proven time and time again.
This is a statistically viable topic to explore. Interestingly, our experiments are coming up short. I assume that this is due to the lack of an experienced dealer amongst the group. Nonetheless, we persist. The equations will come out at some point.