Index ← 1104 CFJ 1105 1106 → text
                               CFJ 1105

  The method Michael used is not sufficiently random and therefore
  incorrect according to the rules.


Judge:        Morendil

Judgement:    TRUE

Eligible:     Antimatter, Blob, Chuck, Crito, General Chaos,
              Harlequin, Kolja A., Macross, Morendil, Murphy, Proglet,

Not eligible:
Caller:       Lee
Barred:       Michael
Disqualified: -
On hold:      elJefe, Oerjan, Swann


  Called by Lee, Thu, 3 Sep 1998 08:02:44 -0500
  Assigned to Morendil, Tue, 8 Sep 1998 10:45:56 +0100
  Judged TRUE, Tue, 8 Sep 1998 12:21:06 +0200
  Published, Tue, 8 Sep 1998 11:35:22 +0100


Judgement: TRUE

Reasons and arguments:

Whoa. I wish the Caller had been more precise in eir Statement - the
method Michael used when ? E's used a few different ones in the past.
I'm tempted to dismiss the Call on that ground, but I don't want to be
too much of a nitpicker, so for clarity's sake : I take the above as
referring to Michael's message relating to the determination of the
outcome in the Call for Revolt made this September 3d.

Michael indicated that e had used the following protocol to select a
number between 1 and 17 for the purposes of determining that outcome,
a selection which, being required to be a 'random' one falls under the
purview of Rule 1079 :

>To determine this in a verifiable way, I take the whole part of the
>Dow Jones closing price for yesterday and take the modulo of this
>number by 17 and then add one to get a number in the range 1 - 17.  I
>determine this method before calculating the result.

In the ensuing discussion, several flaws were pointed out in the
chosen method, some by Michael themselves; the apparent clincher,
however, being a statistical analysis provided by Player Steve
(chi-square test). Michael emself having declined to provide further
proof of the 'randomness' of eir method, in the sense of 1079, and
given the provision of 1079 which places the burden of proof upon the
Player using a given random selection method, I am required to enter a
Judgement of TRUE.


(Caller's) Arguments:

Since Michael did not announce in the public forum the method he was
using before the Dow Jones Closing price for yesterday, he could
looked up the Dow Jones Closing Price and could have tested the
results before announcing them. It is possible that if the results
were not as he had liked that he could have chosen some other
sufficiently published number, like the NYSE composite value, or the
Dow Jones Closing price to the second decimal Place, and tested it
until he came up with a method that seemed random, would be random if
chosen and announced in the public forum before the number was
available to calculate the result.