(January 8, 2015 at 1:44 pm)Heywood Wrote: Probability is an estimate of how likely something is to be true or be the case...or come to pass. I don't need to have computed a probability to know if certain results increase or decrease it. That is part of my point.You are still hung up on observations of results influencing probability. Maybe this will help the conversation.
We all know that the probability of throwing a 1 on a toss of a six sided die is 1/6. Conceivably we could gather all pertinent information and use the laws of physics to predict the outcome: starting position of die, force of toss, angular momentum as it leaves hand, gravity induced hyperbolic arc, air resistance including any wind effect, angle of incidence on surface, energy absorbed by surface on impact, resulting motion characteristics, repeat for bounces, etc. The point here is to illustrate that for something seemingly simple, the variables become to numerous to manage and in some cases impossible to measure. Even though we theoretically can account for everything that is going on, the process is for all practical purposes random. 1/6 probability for any particular outcome.
Now imagine that you are in front of a computer screen. Every ten seconds a number is flashed. You witness ten 1's in a row, but you have no idea what is driving the number. According to you, the probability of the next number being 1 is higher than it was before seeing the first or second 1. Not knowing what was driving the value of the displayed number I too would expect to see another 1 at this point, but that's not probability. It's simply an inference based on previous observations. The next number is a five. The number we see is caused by someone keying in the number corresponding to the throw of a die. Every ten seconds, the probability of you seeing a 1 was 1/6 no matter how many times you had witnessed the number before or the time when you didn't.
Now that we understand what the mechanism is, we can go back and say that seeing ten 1's in a row was highly unusual. If we do this a million times we will see all values near about 167,000 times. The more iterations, the less variance between outcomes to the point that for all practical purposes there is no difference and all can be said to be 1/6th the total number of throws. With a sufficient sample size and without knowing there is a die being thrown we can make certain inferences about our observations: there are only six possible outcomes, each is as likely as the other to occur, the lack of discernible sequencing allows us to conclude that it is random. At this point, because we have the experience of tossing a die, we may even be confident that a die being thrown is what's determining the value we see.
(January 8, 2015 at 1:44 pm)Heywood Wrote: When an atheist claims there are no observations which support the existence of God, they are wrong. There are observations which support the existence of God. They do not prove categorically that God exists but they do support the proposition. Each time that observation is replicated with out being contradicted it moves the probability of the proposition closer to 1.
It would have been more accurate to say that some people make observations and attribute cause to God. Nobody has seen God do anything, particularly since the advent and proliferation of cameras. It's like our previous example of the flashing numbers. After about 100 iterations you really consider the first ten 1's to be bizarre. Not knowing what's causing any of the numbers you conclude it must have been God and then count every other instance of 1 as evidence for God. All the while you simply don't understand what you're observing and lurking in the background is a perfectly reasonable explanation.
