Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Waltzing Matilda (was: statistics, 10 events tell us what ?

Author: fca

Date: 05:22:34 08/17/98

Go up one level in this thread


On August 15, 1998 at 22:14:04, Bruce Moreland wrote:

Dann picked up the 'are the trials mutually exclusive and independent', to which
the correct response must be 'I never said they were, so do not assume it - i.e.
Interpretation X.

Now Dan (as distinct from Dann) has just posted and he spotted the other trap: I
never said "such" trials, or similar words, just trials of "20 in a row". Could
be 20 heads in a row or 20 tails in a row (call this Interpretation Y).  Which
is why I did not agree with your answers (a) and (b), Bruce: I just clarified
the underlying math.  Of course the principles you used were 100% correct, and
this could be viewed as wordplay (though neither Dan(n) did).

As to whether Interpretation X changes the values of (a) or (b) (depending also
on Interpretation Y i.e. 4 cases for each)... please consider. Besides this (a)
and (b) have no more interest.

Dann has posted an interesting answer to (c).  I refer you to my follow-up post
to that (which will be posted presently)

Ditto for (d), which answer (in the hundreds of thousands!) is rather different
from your answer of "1".  Do you wish to reconsider, Bruce (not that I suggest
you are necessarily wrong)?  Is your reply the mode (most likely single) and is
that what was asked for?

You noted (c) was chess relevant: I feel (d) is chess relevant too.  Say you are
playing Ferret / 533MHz Alpha against someone a little weaker (me, human). Say
p(loss for me) = 0.99.  What is the game number of the first game I am likely to
not lose?  And (or do you think the answer is the same?), given that game is x
is the first game in the sequence I did not lose, what is your best guess for x?

Kind regards

fca





This page took 0 seconds to execute

Last modified: Thu, 15 Apr 21 08:11:13 -0700

Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.