Seems vey opinionated.
If what others say is All irrelevant, then we are not going to get any meaningful dialogue.
That's right. The reason for that is that neither you nor Rodney have anythign meaningful to say. A meaningful dialogue is simply not possible under those circumstances.
I am aware that many people do not understand, or are put off by statistics, but I don't think there is any need for meaningless dogmatic responses, completely ignoring the statistical arguments which have been made.
No statistical arguments
have been made, and there is therefore nothing there not to ignore. And, no, I'm not one to be put off by statistics; in fact I've taught the subject professionally from time to time. The simple fact is that the p-value is not relevant in circumstances where random chance is not a factor.
As a simple example : suppose you offer to predict the final score of a basketball game. The "chance" of you doing it correctly through random guessing is astronomically small, but if you've already seen the game (and we're watching it together on tape delay), then "chance" is irrelevant. You have special knowledge that in this case allows you to do something that would be arbitrarily difficult under "normal" circumstances.
Similarly, the chance of being dealt a hand of all spades in a bridge game is almost unthinkable -- unless a professional magician is doing the dealing, in which case it's routine.
The "p-value" is thus a test of which is which.
If you have the special skill or knowledge you claim to have, then you can do a task that is arbitrarily unlikely. If you do not have, then the chances of your succeeding are arbitrarily small.
In the case of the Randi challenge --- Randi listens to your description of what you can do (and what accuracy you can expect). He will then propose a task that
if you can do what you say, will be well within your skill level. The task will also be almost impossible to do "by chance" and he will take appropriate precautions against ordinary cheating-type manipulations.
For example -- I claim to be able to look at a tin of soup (with the label visible) and tell you what the contents of the can are. I claim I can do that with virtually 100% accuracy. (And I can, too; unless the manufacturer made a mistake, which rarely happens.) If Randi accepted the challenge, he might say something like : all right, we will give you a set of twenty soup tinss -- some tomato, and some cream of mushroom -- and you have to successfully pick out the ones that are tomato. I claim that I can do twenty out of twenty, so he will allow me two mistakes and see if I can get 18 out of twenty. Simple claim, simple protocol, simple task.
And do you see how it doesn't matter in this case whether he gives me twenty cans, or fifty, or a hundred? From my point of view, either I can read the labels or I can't. But reading a hundred labels is no more difficult than reading fifty, and if I can't manage a hundred, I can't manage twenty --- or two. The only reason I would care about the p-value is if I plan to guess instead of reading the labels.
Now, for a more "realistic" challenge. I have X-ray vision and I can do the same thing, even when the tins are sealed in an opaque container. The task is almost exactly the same, and the same argument shows that it doesn't matter whether he gives me a hundred or twenty tins. If I really have X-ray vision, I just read the labels through the shoe-box. If I don't have X-ray vision, I can't read the first one, let alone the twenty-first.
The p-value is simply not relevant.
So he would prefer to go with 100 cans, but I might object; it will be a long and boring afternoon, and soup tins aren't free. So I say "how about five", but he doesn't consider that a sufficiently stringent test. Here's where the negotiation comes in. It's my time and money vs. his time, money, and confidence.
). However, it may be possible for someone to pick out a red card at a significantly above chance rate, even if that rate is only 51-52% over thousands of trials. Would such a person apply for the prize, when there is not even a hint in the official rules of what level (s)he must perform at even to pass the preliminary test, let alone win the million dollars? The point is that, by not specifying an odds standard, serious challenges for the prize are discouraged. As a person involved in Ganzfeld experiments told me, the MDC is "so hedged and the criteria for success so arbitrarily set up and changeable at his [Randi's] whim that nobody will ever be able to pass his test."
