TOO MANY DOUBLES :(
Charles M. Solen
giving out WAY TO MANY DOUBLES! Yesterday, in a match I got 6 sets of
them in a row and the fellow I was playing got two or three in the middle
of that run.
It's easy for all of You that have been on the receiving end to say
I've never noticed it or "they have been just as many for the other guy"
But... I THINK something is wrong!
Has anybody else noticed this or is it just my imagination?
Sacker on FIBS
--
============================================================================
| Charles M. Solen + sac...@netcom.com on internet |
| Las Vegas, Nevada + |
|___________________________________________________________________________
| Lord, Help Me be the man that My dog thinks I am! |
============================================================================
chrisw
says:
>
> For the past several weeks I've been noticing the dice generator is
>giving out WAY TO MANY DOUBLES! Yesterday, in a match I got 6 sets of
>them in a row and the fellow I was playing got two or three in the middle
>of that run.
> It's easy for all of You that have been on the receiving end to say
>I've never noticed it or "they have been just as many for the other guy"
>But... I THINK something is wrong!
> Has anybody else noticed this or is it just my imagination?
>
> Sacker on FIBS
NO, it is not your imagination! In the past few days I have seen more sets
of multiple doubles than I have ever seen! I had an opponent roll -3-
straight doubles *twice* in one match! I've also had a couple of players
roll 3 doubles while bearing off! Come to think of it, there were quite
a few doubles rolled in our match early this morning. While I have had
my share of doubles in the past, *nothing* has compared to what I've seen
lately! In a 5 point match I played just the other night, my opponent
was stunned at the number of doubles he was rolling!
Something is definitely wrong!
chrisw on FIBS
---
"a politician is an arse upon which everyone has sat except a man."
E. E. Cummings (1894-1962)
Bill Bohn
: > For the past several weeks I've been noticing the dice generator is
: >them in a row and the fellow I was playing got two or three in the middle
: >of that run.>
: NO, it is not your imagination! In the past few days I have seen more sets
: of multiple doubles than I have ever seen! I had an opponent roll -3-
: chrisw on FIBS
I'd like to come forward to say the doubles are VERY odd if I may
use an oxymoron... :>
I played bibi in a match the other day and watch about 10 doubles go in
bibi's favor. with me getting only about 5 ;>
And I've rolled more double 3's in a row that they seem loaded...
Akira
--
"You don't believe in God, you don't believe in me..."
"You Scientists...don't believe in anything!"
-Mr Frost
Erik Gravgaard
: For the past several weeks I've been noticing the dice generator is
: them in a row and the fellow I was playing got two or three in the middle
: of that run.
: It's easy for all of You that have been on the receiving end to say
: I've never noticed it or "they have been just as many for the other guy"
: But... I THINK something is wrong!
: Has anybody else noticed this or is it just my imagination?
In my opinion the dice at FIBS are as normal as one would would see in
any other backgammon club.
Sometimes (also in real life) you see a number of doubles in a row, or a
number of perfect roles or maybe a number of dissaster rolls, but that's
life - and backgammon is a cruel game, sometimes as cruel as life it self.
--
Erik Gravgaard (erikg) --------------------------
Pres. of the Danish Backgammon Life is a series of up
Federation and down equity decisions
er...@inet.uni-c.dk - Paul Magriel
--------------------------
chrisw
this match after the first game and perhaps 1/3 thru the second.
So these results are from approx. 2.5 games played:
Doubles rolled:
chrisw - 15
Opponent - 9
1-1: 5 times
2-2: 4 times
3-3: 5 times
4-4: 4 times
5-5: 4 times
6-6: 2 times
I'm not sure how many were rolled in the first game and in the early
part of the second but this seems to be high. I'll record an entire
match to get a better sampling.
chrisw
Kit Woolsey
dice rolls are as random as one could want. Let's face it -- doubles
will come up 1/6 of the time, which means that if you play enough (or too
much as is the case with most of us), you are going to see sequences of
consecutive doubles. What happens is that we tend to remember the
*remarkable* sequences, while forgetting about the normal sequences where
a bunch of doubles don't occur. In this way, it is easy to convince
oneself that more doubles occur than there should be. This is called
selective memory, and can distort a person's picture of reality.
I am not an expert on random number generators, but I would guess that if
there were something wrong with the FIBS dice generator the results would
be more screwed up than just producing more doubles than there should
be. I'm sure there are some random number generator experts out there
who could say more about this.
However, it is always possible that something is wrong. In order to
determine this, it would be necessary to set up an unbiased test. To do
this, one would have to state IN ADVANCE the bias which they think
exists. For example, one might state: I think FIBS produces more
doubles than it should. Then one would determine a match to start
counting, and from then on keep track of EVERY dice roll (no cheating)
for a large number of rolls, the larger the better (decide in advance how
many rolls you will count). For example, suppose you decided to count
1000 dice rolls. It is clear that the average number of doubles one
would expect if things were truly random is 166 2/3. I'm not much of a
statistician so I can't say a whole lot about confidence levels and all
that stuff, but if more than 200 doubles turned up in the trial I would
think there would be cause for concern. So, somebody who really believes
that there are more doubles than usual should try such an experiment. By
the way, for 1000 trials I'm betting on under 200. Anybody want to bet
over 200?
Kit
Grant Edwards
: Here's a break down of a match I just finished. I began recording
: Doubles rolled:
This is all meaningless unless you also record the total number of
rolls.
--
Grant Edwards | Microsoft isn't the | Yow! Are we on STRIKE yet?
Rosemount Inc. | answer. Microsoft |
| is the question, and |
gra...@rosemount.com | the answer is no. |
James Eibisch
the other day as well. I played one match this morning in which
my opponent rolled the most, or near-most, number of doubles in
one game I've ever seen - 6 out of 7 rolls were doubles, etc. The
day before, I rolled 66 66 66 55 (while on the bar - harrumph).
Many other instances of freaky double-rolling the last few days,
too. Maybe it's all just coincidence, maybe not, but the last 2-3
days have seemed remarkable to me in this respect.
--
James Eibisch
Reading, England
Don A. Hanlen
black box evaluations of marvin's random program, certainly more work than
I care to put out. Besides, even *if* there's a glitch, it affects us all
equally.
The only thing I'd be willing to do is look at marvin's random() routine,
but I've asked many times to no avail. Apparently, it's a hot secret.
One thing that scares me is I've heard it's tied to a clock. If it is,
then I could conceive of Patti's end-game, autoplay scenario being true.
Anyway, like Old Man River, I plan to just keep rolling along.
--
--=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=--
Don A. Hanlen Computer Science
dha...@beta.tricity.wsu.edu Washington State University
(509) 588-4528 Tri-Cities
Christopher Yep
Don A. Hanlen <dha...@beta.tricity.wsu.edu> wrote:
>Patti makes some excellent points. It would take a lot of work to do
>black box evaluations of marvin's random program, certainly more work than
>I care to put out. Besides, even *if* there's a glitch, it affects us all
>equally.
If long strings of high doubles (say, 3-3 or greater) occur more frequently
than statistically expected in the bearoff, then it allows for more
come-from-behind wins in a race. Such a glitch would favor match-underdogs
in a match, because they then would be closer to 50% to win the match.
In general, such a glitch would be more advantageous to less-skilled FIBS
players.
>The only thing I'd be willing to do is look at marvin's random() routine,
>but I've asked many times to no avail. Apparently, it's a hot secret.
>One thing that scares me is I've heard it's tied to a clock. If it is,
>then I could conceive of Patti's end-game, autoplay scenario being true.
Yes, it is tied to the clock (numbers determined by random number
generator + "insignificant digit in the clock"). For further details,
contact John Bazigos, aka "doc" of FIBS.
It is theoretically possible for certain uninterrupted sequences (ie if
all moves are forced and both players have automove set) to occur more
frequently than statistically expected, on FIBS. For example, a long
string of high doubles (by both players at the same time) during bearoff may
occur more frequently than is statistically expected. Whether any sequences
actually do fall under this category is an open question.
Chris
Don A. Hanlen
I have constructed the following chart showing the probabilities
of rolling x doubles in n rolls for n = 1..10 . Numbers are
rounded and I added up each set to make sure I got 1.0 (a good
thing to do when working with probabilities). The function is
the binomial probability distribution (BPD) { [p = 1/6], [the
binomial coefficient is C(nx) = n!/(x!*(n-x)!)] }:
C(nx) * p^x * (1-p)^(n-x) .
Some explanation is called for because it seems that this chart
supports, unequivocally, those who think doubles are out of whack
on FIBS:
Consider a sequence of 8 rolls. Suppose you are interested in
knowing what your chances are of getting 4 in-a-row at some point
in this sequence ( 1/1296 or .00008 for 4 rolls).
o Your chances of 4/8 doubles are .026 or 1/38.
o There are 5 *sequences* of 4 rolls in those 8 rolls,
1234, 2345, 3456, 4567, 5678 .
o A second application of the BPD yields chances of .0038
or 1/260 in *independent* trials of getting 4 in-a-row.
The word, independent, is key. The 5 sequences in 8 rolls are
*dependent*. For example, if you've *already* rolled two doubles
in-a-row, your chances of 4 in-a-row are 1/36, your chances of
4/8 in *this* sequence, starting with 2 doubles, are .2 or 1/5 --
they've increased by a factor of 10. Only the first sequence,
1234, and the second, 5678, are *independent*.
I believe there is a close connection between mathematical
dependence and the selective memory of game players. As has been
said before there are two ways to determine the randomness of the
rolls on FIBS. Exhaustive black box testing -- recording and
analyzing rolls -- or white box testing which would involve an
actual examination of marvin's random function.
Note that I'm neither endorsing nor debunking the notion that the dice on
FIBS are a good approximation of the real world at all times.
* * *
Binomial probability function for rolling pairs of dice and
getting doubles:
P(1 roll) n x P(2 rolls) n x
0.833 1 0 .694 2 0
1/6 --> 0.167 1 1 0.278 2 1
----- 0.028 2 2 <-- 1/36
1.000 -----
1.000
* * *
P(3 rolls) n x P(4 rolls) n x
0.579 3 0 0.4823 4 0
0.347 3 1 0.3858 4 1
0.069 3 2 0.1157 4 2
0.005 3 3 0.0154 4 3
----- 0.0008 4 4
1.000 ------
1.0000
* * *
P(5 rolls) n x P(6 rolls) n x
0.4019 5 0 0.33490 6 0
0.4019 5 1 0.40188 6 1
0.1608 5 2 0.20094 6 2
0.0322 5 3 0.05358 6 3
0.0032 5 4 0.00804 6 4
0.0001 5 5 0.00064 6 5
------ 0.00002 6 6
1.0000 -------
1.00000
* * *
P(7 rolls) n x P(8 rolls) n x
0.279082 7 0 0.2325680 8 0
0.390714 7 1 0.3721089 8 1
0.234429 7 2 0.2604762 8 2
0.078143 7 3 0.1041905 8 3
0.015629 7 4 0.0260476 8 4
0.001875 7 5 0.0041676 8 5
0.000125 7 6 0.0004168 8 6
0.000004 7 7 0.0000238 8 7
-------- 0.0000006 8 8
1.000000 ---------
1.0000000
* * *
P(9 rolls) n x P(10 rolls) n x
0.1938067 9 0 0.16150558 10 0
0.3488521 9 1 0.32301117 10 1
0.2790816 9 2 0.29071005 10 2
0.1302381 9 3 0.15504536 10 3
0.0390714 9 4 0.05426588 10 4
0.0078143 9 5 0.01302381 10 5
0.0010419 9 6 0.00217064 10 6
0.0000893 9 7 0.00024807 10 7
0.0000045 9 8 0.00001861 10 8
0.0000001 9 9 0.00000083 10 9
--------- 0.00000002 10 10
1.0000000 ----------
1.00000000