HAVRILIAK says...

TO: Guy Macon
Thank you very much for your comments. I've tried your suggestion
to prove the point.

No you didn't.

Instead of flipping a coin I did the following calculation.

Your calculation is wrong. Do the experiment and you will see with your
own eyes that the actual results do not come out as you have calculated.

If you flip an unbiased coin 2 million times it will come up heads
1 million times and tails 1 million times.

Once again, a simble experiment will tell you that the above is wrong.

If you believe that if you flip an unbiased coin 2 million times it
will come up heads 1 million times and tails 1 million times, then
you must also believe that if you flip an unbiased coin 20 times it
will come up heads 10 times and tails 10 times. So do the experiment:
flip a coin 20 times and report the result.

Here is my result: 8 heads, twelve tails.

Again I ask you to do the experiment rather than making calculations.
The fact that the results of the experiment won't match the predictions
from your calculations will tell you that you have made an error in your
calculations.

Here is that experiment again. This time do the experiment.

************************************************** ************

You can prove this for yourself. Make a number line like this:

....-9 -8 -7 6 5 -4 -3 -2 -1 0 +1 +2 +3 +4 +5 +6 +7 +8 +9...
^
POINTER

Now start flipping a coin, moving the pointer one place to the
right if it comes up heads, left if tails.

You will find that the pointer gets farther and farther from the
starting point, and less and less likely to get back to zero.

If you suspect a coin bias, flip twice as often and move the
pointer one place to the right on heads-heads or tails-tails,
left on heads-tails or tails-heads. Or you can take my word
that the results will not change.

You can even do a computer simulation if you have a source of
random numbers such as /dev/random/ on a Linux box.
(The pseudorandom generator that comes with your computer
language is *not* random.)

Do the experiment, then report the results.