Kieren, how about this:
You are going to flip a coin. What is the probability that it will be heads?
50%.
Right?
Now, let’s say I know ahead of time with absolute certainty that when you flip the coin, it is going to be heads. Don’t worry about how I know this, I just do. And with absolute certainty – I cannot be wrong. Furthermore, I don’t tell you ahead of time.
Now when you flip the coin what is the probability that it will be heads?
100%.
Right?
After all, I knew it was going to be heads, so it has to be. Of course, since I didn’t tell you this, you thought it was a 50% probability. Little did you know it was absolutely going to be heads. Logically, I can prove this because all alternatives (that is, the coin comes up tails) violates the precondition (i.e. that I absolutely knew it would be heads.)
It’s the same thing Fish is saying. Once someone knows with absolute certainty the outcome of any event, then that is what has to happen. It can’t be random. And nobody can choose to make any other event happen. Anything else violates the precondition.
Of course it doesn’t work in retrospect, as someone tried to argue. Back to the coin flip.
This time I have no idea what the outcome is, other than there is a 50% probability that it will be heads. You flip the coin. It’s heads. Now I know with absolute certainty that it was heads because I saw it happen. But the fact that I know this now does not change the original probability. It was still 50%.
Is that more clear?