BackgammonMasters, bots and crooked dice
HarryPotter
played with Snowie the past few years a LOT so I'm upto date on the
latest information. I was at the Las Vegas event in November and
Backgammonmasters was there and I talked to the rep who assured me,
with the timer they have on the game, I would not be playing someone
with Snowie playing beside them at home. So I signed up.
After dumping 5 grand in a month, I've come to the following
conclusions: I'm either having the worst run of bad luck in the history
of the planet, or I'm being cheated. I've played perhaps 300 or more
games, in the $50 and $20 limits. I have the lowest ratings of anybody
there and I've just been playing others with low ratings (1300's and
1400's). Here are the facts:
I've played with maybe a hundred different players. After a while i
began to write down names and rate their play. Most of them play
identically: Their checker play is ok, good even, but their cube play
is allways based on 'live cube' decisions. However, in these limits
the most you can win or lose per game is 4 points, no matter what.
These players drop most cubes which are the least bit gammonish, which
is a huge blunder and especially in games with a dead cube.
There are many automatic doubles (not declinable), and that was my
first suspicion something was wrong. It seemed like much more than one
game in 6 was an auto. So with the cube in the middle in so many games,
if you are doubled, you not only have no redouble vig, but you can't
get gammoned either, so the take is exactly 25%. Yet my opponents would
drop like flies when I doubled a gamonish position when they were 35% a
winner.
Next, they would give me countless free rolls with the cube in the
middle at any value. How could I be losing? At first I just thought bad
luck. There were so many 17-1 shots hitting me at the crucial time. And
I was getting killed in close races. It seemed the opponent was rolling
doubles every 4 or 5 rolls, not one in 6.
As for my cube taking, I began to doubt myself so I took screen shots
of all the positions where I was doubled. Nearly all my actions were
correct and my misses were off never more than 1%. When I'd take a
gamonish but correct take, what would happen was the opponent would
start rolling doubles every 2 or 3 rolls and close me out and gammon me
time after time after time. And once I had a 2 or 4 cube on my side I
would dance on a 3 and 4 point board 4 or 5 times in a row until the
opponent was all over me. I'm not talking about a couple games or 1
session. I'm talking about 300 games or more.
I got really suspicious just this week. There is a chat window so I
started saying hello to my opponent. I did this to dozens of players. I
never got a response. I'd even stop in a bearoff where I was going to
lose and not worry about the timer going off, and message them over and
over and over. NEVER have I got a reply. So I'd wait 2 or 3 minutes,
then make my move. Their roll was always instantanious. Don't you think
one of them might have glanced away for a second after such a delay?
Then I started thinking some other things: These players all play
without hesitation. I know there is a clock but it's not like you are
going to run out of time if you take a few seconds now and then. These
opponents never took any time. And on cube decisions: Some of them were
close. A human would occasionally pause to evaluate. A computer's
pause is the snap of a finger. The cover for this is occasionally the
opponent would play the checkers one way, then bring them back and redo
it correctly. However, the moves they took back were seldom valid
plays. It was almost always like you clicked on the wrong checker by
mistake. Was this a smokescreen? Take away this seemingly human
maneuver and there is nothing human left about my opponents.
Could it be I was not only playing bots (whose cube was bad) but that
they had control of the dice as well? And why were there so many
different players like this? I think it's a human reaction that there
couldn't be that many humans cheating you on one site. But for a
computer that's nothing. They could have hundreds, change their names
often, keep their ratings below 1500 with a click of the mouse, but all
this would have to be if Backgammonmasters was doing it. An outside bot
wouldn't be able to control the dice.
So am I delusional, or incredibly unlucky or what? I'd sure like to
get some feedback from some other players there that have played in the
bigger money games. Please tell me if you noticed any of the same thing
I have.
Zzyzx000
WBA
looks like a bot-populated site.... go figure. i had the same
suspicions and after some games and analysis, i listened to my 6th
sense and decided not to deposit any money. good for me after reading
your thread.
Grunty
> I was at the Las Vegas event in November and
> Backgammonmasters was there and I talked to the rep who assured me,
> with the timer they have on the game, I would not be playing someone
> with Snowie playing beside them at home. So I signed up.
Well, they told you the truth - you didn't run into *people* getting
aid with a program.
> After dumping 5 grand in a month, I've come to the following
> conclusions: I'm either having the worst run of bad luck in the history
> of the planet, or I'm being cheated.maneuver and there is nothing human
> left about my opponents.
>
> [ snip ]
>
> Could it be I was not only playing bots (whose cube was bad) but that
> they had control of the dice as well? And why were there so many
> different players like this? I think it's a human reaction that there
> couldn't be that many humans cheating you on one site. But for a
> computer that's nothing. They could have hundreds, change their names
> often, keep their ratings below 1500 with a click of the mouse, but all
> this would have to be if Backgammonmasters was doing it. An outside bot
> wouldn't be able to control the dice.
>
> So am I delusional, or incredibly unlucky or what? I'd sure like to
> get some feedback from some other players there that have played in the
> bigger money games. Please tell me if you noticed any of the same thing
> I have.
Take a look at the thread "Opinions about BGroom.com ?" Nov 6, in the
subthread where the discussion shifted to the "BGM bots" matter. Review
the dialog between Robi and Piranha. Robi uses to play bigger stakes.
> Zzyzx000
No, I don't think you're delusional or incredibly unlucky.
HarryPotter
left and not willing to give them more. I was thinking I was playing a
robot, but for the 1st time, they answered my chat.
I was playing a typical game, automatic double, doubles thrown about
every 4th roll, when I started thinking: What if BGmasters just changed
the frequency of doubles and that was all they did? That would surely
meess up my checker play strategy. Why would they do it? Well, every
automatic double doubles the stake and doubles their commission. Hum..
Anyways, I asked my opponent if they noticed the abundance in doubles
being thrown (as 3 of the last 4 rolls were doubles and they said no.
I read another thread somewhere where a player confronted management
about bots and the 1st reply was that a bot would not take back a move
and play it over. I found this to be an interesting answer. If you were
the manager of a site and it was on the level and someone asked you if
you had bots, what would be your first reaction? Of course he may have
had similar questions before and investigated and that was his
conclusion. Yet if you wanted to program a bot to disguise their play,
it seems like that little decoy would not be hard to program...
I just know I'm a big favorite vs. whatever or whomever I've been
playing just because of their poor cube action. It's the dice I'm
concerned with. It could be something simple like they increased the #
of doubles thrown to increase the automatic doubles and thus their
rake, or it could be much more sininster like bots playing with someone
monitoring them and selecting the next roll when needed. Or I could be
incredibly unlucky lately. More feedback is appreciated from others.
Thanks
zzyzx000
chouette
dice.
I also consider myself a good player, and realized that it was impossible
to win, with sistematical loses, including those games almost won,
suffering magic dices and incredible series of doubles. And all the games
were played with a lot of different players, with low ratings but perfect
play, without pauses for thinking cube decissions or difficult plays.
mont...@lycos.com
I wonder? Or is this one site that he thinks is in fact "crooked,"
though he can't believe that any other site manipulates software to
"even things out" a bit, keeping the novices coming back for more?
Robi
"Grunty" <grunti...@yahoo.com> schrieb im Newsbeitrag
news:1167313590....@i12g2000cwa.googlegroups.com...
>
> Take a look at the thread "Opinions about BGroom.com ?" Nov 6, in the
> subthread where the discussion shifted to the "BGM bots" matter. Review
> the dialog between Robi and Piranha. Robi uses to play bigger stakes.
True. I don't play anymore there. Don't know what's going on, maybe I'm
wrong, but my feeling is bad. I play
the big freeroll tournament (4.5$ commission) starting in 1 minute :)
Have you seen taht the 100/400 and 125/500 tables are gone.
Robi
Derek Ray
> I played a few games there last night for $7.50, as I've only got $80
> left and not willing to give them more. I was thinking I was playing a
> robot, but for the 1st time, they answered my chat.
>
> I was playing a typical game, automatic double, doubles thrown about
> every 4th roll,
Did you track this, or is this just a random guess?
> Anyways, I asked my opponent if they noticed the abundance in doubles
> being thrown (as 3 of the last 4 rolls were doubles and they said no.
Is it possible that your opponents' perception is more accurate than
your own?
I suggest you Google "selective perception" and read up on it. Unless
you are explicitly tracking this sort of thing, you are never going to
have an accurate idea of what's happening.
> had similar questions before and investigated and that was his
> conclusion. Yet if you wanted to program a bot to disguise their play,
> it seems like that little decoy would not be hard to program...
Correct. It would, in fact, be trivial.
> I just know I'm a big favorite vs. whatever or whomever I've been
> playing just because of their poor cube action.
Have you recorded the matches and fed them into gnubg or Snowie for
analysis? If you are making checker blunders, you can easily get
trounced regardless of your opponent's cube action. Statements like "I
just KNOW" are very suspect, for example; since you haven't done any
analysis, in fact, you DON'T know. All you know is that you're losing.
To understand why you're losing, you must record matches, or the only
thing you have is a bunch of faulty guesswork.
Also remember that cube take/drop info assumes perfect play from both
sides from that point forward, but unless someone actually drops, it
only changes the amount a game is played for. Checker play, on the
other hand, directly influences the outcome of the game.
If I could be confident my opponent would routinely make big checker
blunders in complicated positions, I would consider modifying my cube
play appropriately (take, redouble, and retake, mostly).
> It's the dice I'm concerned with.
Don't be. Regardless of what our resident conspiracy theorists may tell
you, the dice are fine and nobody has ever demonstrated (or _attempted_
to demonstrate) otherwise; it's just a bunch of smoke and hooey.
If I were you, I'd be much more concerned about someone with gnubg or
Snowie on a laptop sitting next to their machine, or even running in
VMWare/Virtual PC or just plain old having it open at the same time, if
the server doesn't monitor for that.
> monitoring them and selecting the next roll when needed. Or I could be
> incredibly unlucky lately. More feedback is appreciated from others.
I would guess one of three things here:
-- You are unlucky
-- You are not as good as you think you are
-- You are playing for much too high stakes, and are running into
players who are playing with Snowie/gnubg by their side, helping them
through the hard positions.
--
Derek
insert clever quotation here
ri...@notyahoo.com
>Where is Rich to tell you all that you are "delusional" or "libellous,"
>I wonder? Or is this one site that he thinks is in fact "crooked,"
Don't know since I never played there. But my response is the same as
it is to you. Unless you analyze enough games one cannot assert with
any reasonable degree of certainty that the dice are fixed. Selective
perception just does not cut it.
>though he can't believe that any other site manipulates software to
>"even things out" a bit, keeping the novices coming back for more?
I have seen no evidence of this nor has anyone (AFAIK) *ever*
demonstrated this by an analysis of a large enough number of games. If
someone can give more evidence than selective perception I will take
them more seriously.
Rich
mont...@lycos.com
often starts with a "hunch" or "impression," and then conducts an
investigation consistent with the scientific method. The problem with
the backgammon issue is that if a site decided to create "bouncey dice"
software, we are not allowed to examine it. In science, there are
cases where qualified scientists cannot get funding to do reasonable
experiments. Now in this case, Rich is claiming that if one gives him
several hundred games from a particular site, he is capable of
determining if dice software is "fair." This is his claim, and it is
his responsibility to explain exactly how he can be sure that his
analytic technique will accomplish this. Moreover, he can play several
hundred games at these sites and do his own analysis - he does not need
anyone else's games.
Thus, the key points are:
1. The sites should be "transparent" and have independent oversight,
but since that does not appear to be on the near horizon, then:
2. A person should articulate how he/she is going to examine a sample
set of games, and then:
3. Play the games, or obtain the games from someone else, and do the
analysis, then present it here.
An important point is exactly what "luck" means in this context. It
needs to be defined precisely. Kit Woolsey made a point about looking
at specific situations, and it seems to me that this is because the
"luck" notion in backgammon is far from self evident. For example, I
can get a bunch of "lucky rolls," and get my opponent in an acd point
anchor game. Now he closes his board and gets a shot and hits it, then
doubles me out. For me, the "luck" that decides the game is what's
important, not the luck leading up to the decisive "lucky" roll.
ri...@notyahoo.com
>In science, the best method of investigation humans have devised, one
>often starts with a "hunch" or "impression," and then conducts an
>investigation consistent with the scientific method.
I agree completely. You have a hunch or impression based upon
selective perception that dice are rigged on certain sites. The burden
of proof is on you to prove it.
> The problem with
>the backgammon issue is that if a site decided to create "bouncey dice"
>software, we are not allowed to examine it.
Huh? Why are we not allowed to examine it assuming that the site
allows us to save match logs? Of course on BGR it would be very
difficult to do so.
> In science, there are
>cases where qualified scientists cannot get funding to do reasonable
>experiments.
This is true. What on earth this has to do with statistical evaluation
of rolled dice is beyond me.
> Now in this case, Rich is claiming that if one gives him
>several hundred games from a particular site, he is capable of
>determining if dice software is "fair."
Within a certain confidence level this is true. Of course the more
games you have the higher your confidence that your result is
reliable. But you will never have 100% certainty.
> This is his claim, and it is
>his responsibility to explain exactly how he can be sure that his
>analytic technique will accomplish this.
I am not about to explain basic statistics to you Monty. You are the
Professor. Do your own homework.
> Moreover, he can play several
>hundred games at these sites and do his own analysis - he does not need
>anyone else's games.
I sure could. But I am not the one making a remarkable claim about
dice being rigged on an on line site. You are. Since you are making an
incredible claim, the onus is on you to prove it. I understand you
would like me to have the onus to provide proof that the dice are not
rigged. It does not work that way "Professor".
>
>Thus, the key points are:
>
>1. The sites should be "transparent" and have independent oversight,
>but since that does not appear to be on the near horizon, then:
Create your own site if you don't like the one's in existence. Don't
play if you are paranoid about the dice.
>
>2. A person should articulate how he/she is going to examine a sample
>set of games, and then:
>
>3. Play the games, or obtain the games from someone else, and do the
>analysis, then present it here.
Once again Monty wants others to do the work for him. Not only does he
want all this free money but he won't even do any work to prove his
hypotheses.
>
>An important point is exactly what "luck" means in this context. It
>needs to be defined precisely.
Why don't you define it Monty.
> Kit Woolsey made a point about looking
>at specific situations, and it seems to me that this is because the
>"luck" notion in backgammon is far from self evident. For example, I
>can get a bunch of "lucky rolls," and get my opponent in an acd point
>anchor game. Now he closes his board and gets a shot and hits it, then
>doubles me out. For me, the "luck" that decides the game is what's
>important, not the luck leading up to the decisive "lucky" roll.
Why don't you contact Kit and tell him that you think that the dice on
XG are rigged and tell him why you believe it to be so. See what he
says.
Rich
mont...@lycos.com
not seem to realize that it is common for a scientist (or other
scholar) to mention to a colleague or graduate student that he/she has
a hunch or impression, and that it may be worthwhile to investigate.
If you don't want to explain what you mean by "luck," that is your
business, but in order to have a reasonable discussion, someone must do
this. As I said, let's take an obvious example: you play several
hundred games at a site, and it seems as though you get a lot of "luck"
early, getting your opponents into ace point holding games, but they
get one lucky shot right after they close their boards, then redouble
you out of the game. How would such games be evaluated, in terms of
"luck?" I put forth the notion of "decisive luck," which means that
the game is just about totally decided, in this case by one bad luck
roll by you, followed by a direct shot that your opponent hits, which
is not all that "lucky" actually, since it has to be at least 11/36.
What KW thinks of XG is irrelevant here: the discussion has to do with
how to evaluate "luck" in backgammon. If this cannot be agreed upon,
then all anyone can do is to present "impressions," and thus the onus
is on you to articulate what "backgammon luck" is and how to evaluate
it in a way that consistent with established principles (math, logic,
etc.). Because I don't know if there is a way to establish a useful
"backgammon luck" doctrine or formula, I can only put forth
impressions. If you are irritated by the exercise of free speech in
this context, then it is up to you to provide the necessary analytical
tools. Obviously, you do not agree, and so there is nothing remaining
but to "agree to disagree," at least for the time being. Raising all
kinds of irrelevant and distracting issues (reading minds, making
diagnoses of mental illness, charging that crimes have been committed,
etc.) demonstrates an inability to partake of an honest intellectual
discussion. I hope that one day you are able to understand this simple
point.
Dan Pelton
doubles, and the fact that they pulled out on short notice of sponsoring
several ABT events with no explanation doesn't lend me to feel comfortable
about the site. I contacted the woman who was representing
BackgammonMasters in Las Vegas to ask if their changes were a result of the
new US law, and she supposedly forwarded my question to BackgammonMasters.
I have NEVER received any reply. Perhaps someone who is involved in the ABT
events effected (Greg C?) could shed more light.
"chouette" <fro...@argensun.com.ar> wrote in message
news:49a613c12c6d3c7c...@localhost.talkaboutgaming.com...
Kees van den Doel
<mont...@lycos.com> wrote:
There is a "luck" tab in the analysis window of GNUBG which shows the
articulated luck, consistent with established principles (math, locic,
and, last but not least: etc.)
Maybe your doorknob can explain it to you better than I can.
Kees (From kvan...@xs2.xs4all.nl Sun Oct 30 0 anno bovani us your
doorknob, if in Maastricht, Gouda, Utrecht, Gezellig met
opmerkingen moeten praten zolang deze onrustbarende terugloop in
SF PL+IO- M9G/Q;-+QV:=IVSZFLFL!EM\IK&E
M;48,TVOAVY"F^JF"E!NFF4NG+KY;F MW":TZ6SJF^09$D!P3IL P Goed
definieeren?)
ri...@notyahoo.com
>There's nothing unusual about my position, Rich. It is you who does
>not seem to realize that it is common for a scientist (or other
>scholar) to mention to a colleague or graduate student that he/she has
>a hunch or impression, and that it may be worthwhile to investigate.
Straw man argument. I have never said nor implied that it is uncommon
for a scientist to mention that he has a hunch or impression that may
be worthwhile to investigate.
If you had said that you had a hunch that the dice on XG were rigged
and that it would be useful to investigate this possibility before
passing judgement on XG I would have absolutely no problem with that.
But that is not what you have said. You have said that it is obvious
that the dice on XG are rigged to make the games more exciting and
that if dice behaved the way that they do on XG in "real life" that
you would not continue to play the game of backgammon.
I understand that you are attempting to suggest that you are simply
giving your impression or hunch that needs testing before coming to a
conclusion. You have clearly come to a conclusion about XG and are now
backpedaling about that conclusion. Apparently you made similar
accusation about CG, accusing them of manipulating the dice even
though the site administators assured you that the dice were generated
by a random number generator.
Once again I have no problem with you having a hunch about CG dice
that needs investigation. But you act as is your "hunch" is the gospel
truth although you are likely going to deny this.
Rich
mont...@lycos.com
truth although you are likely going to deny this."
You need to cite quotations that have led you to this conclusion. In
fact, I was taught in graduate school to make "qualified" statements if
the issue is not clear, as is the case here. Instead, my contention is
that you believe yourself to have some sort of "mind reading" ability -
this is based upon your clear and undeniable diagnoses of various
psychological disorders that you feel you are qualified to render.
Readers will have to decide for themselves here.
As to Kees point about GNU having some sort of "objective" ability to
determine "luck." I set up a game against GNU in which I controlled
the the dice. I allowed it to get me into an ace point holding game,
and I played terrible moves on purpose. Then, as soon as I closed my
board, I gave it a bad roll and then hit (an 11/36 shot). The analysis
said that GNU played at "world class level" and got "good dice, man."
The human opponent played "awful."
Now it is obvious that whether you agree with this arbitrary way to
judge "backgammon luck" or not, my point was that a better notion might
be "decisive luck." In this game, the human played terribly and got a
lot of bad luck at the beginning, then simply hit an 11/36 shot at a
good time - decisive luck. If a site could manipulate the software to
do this kind of thing once in a while, it would really help the players
who aren't very good, because they often find themselves in these kinds
of very bad situations.
Without credible and independent oversight of the sites, this is a
major concern, but in any case, my main point here is that if you use
GNU to determine "luck," you could have great "luck" over the course of
hundreds of games, and yet still you have just broken even, more or
less.
Derek Ray
> As to Kees point about GNU having some sort of "objective" ability to
> determine "luck." I set up a game against GNU in which I controlled
> the the dice. I allowed it to get me into an ace point holding game,
> and I played terrible moves on purpose. Then, as soon as I closed my
> board, I gave it a bad roll and then hit (an 11/36 shot). The analysis
> said that GNU played at "world class level" and got "good dice, man."
> The human opponent played "awful."
First: Please provide the match record when you cite examples like
this, as it allows us to see exactly what you're talking about rather
than rely on your hearsay. Any serious student of science knows that
one of the cardinal rules of an experiment is that it be independently
repeatable; and to repeat it requires that we know the exact steps you
took, which includes dice rolls, moves, and settings on GNU.
You are clearly not a serious student of science, Professor.
Second: What do you believe is incorrect about GNU's evaluation?
-- You admit to having intentionally played "awful".
-- GNU probably did play at world class; it tends to.
-- You say you controlled the dice, so you probably gave it some good
rolls to help it get into a strong position against you; this would
indicate "good dice, man". We can't know for sure without the match
record that you so carelessly failed to post for us.
And most importantly: Did you continue the game after your hit? Just
making the initial hit from that position does not guarantee a win... in
fact, far from it. You will need some further good rolls, as well as
your opponent not to roll any big jokers, in order to come back and win
the game from there. Again, if you had bothered to include the match
record, we could have done some PROPER analysis, rather than your
half-assed conclusions.
At the moment, your post and example case tends to support the premise
that GNU's luck evaluation is _better_ than your concept of "decisive luck".
Looking forward to the match record of this experiment!
mont...@lycos.com
This is so simple I did not think anyone would have a problem
comprehending it, nor creating a similar game. Can't you put GNU on
manual dice and make up a game where you get caught in an ace point
holding game? You give GNU a lot of good rolls to start. GNU will
double and you should drop, but you take, and of course playing bad
moves is fine, because you control the dice. Then you will give GNU a
6,5 when it is bearing off and then you hit the blot, redouble, and it
has to pass. What you get is GNU giving itself a "good dice, man"
rating along with "world class" play, but it loses anyway, even though
you play "awful." Now if a site manipulated the dice with this in
mind, you would have several hundred games, with a fairly good "luck"
rating, but because of what I call the "decisive luck" factor, you
could be even or actually losing money at the site, no matter how much
better you are than your opponent. Thus, if you examined your games
and saw that your "luck" was fairly good, people like Rich might say
that this "proves" that he is correct.
However, all a site would have to do is to manipulate the dice so that
poor players get the "decisive luck" once in a while, and the GNU
evaluation (of several hundred games) would look reasonable, even if
you took the opponent's "luck" into consideration as well. This is
about as basic as things get, and if you don't understand what I'm
saying here, it may be time for you to switch to a different game.
In using the GNU analysis feature to examine a number of recent
matches, what I noticed is that I am often rated "intermediate" in
checker play, while my opponents are rated less, often as beginner or
awful. However, my "bad" plays usually come in very strong positions,
where I play a good move but not the best one, and I've written up
posts here on such positions. At BGR, I've found that such contests
are usually easy to win when I play in long matches, but even against
terrible players, DMP sessions often ended about even. At XG, I've
only played 3 and 5 point matches, and my record in the 5 point matches
is considerably better. Thus, there is certainly an argument that can
be made that if I played 9 pt. matches at XG I would get similar
results that I have at BGR. As I've said, I agree with the person who
said that XG's dice appear "bouncey," but before any claim can be put
forth, there needs be analysis. Backgammon "luck" needs to be
formulated in a way that is sensible, and so far I have yet to see such
a formulation. Here is an area where the scientific method can be
used, in that any formulation put forth can be tested by creating
games, then considering if software could be manipulated to make things
look "fair." This is why I created a game against GNU, but in any
case, the onus is on the person who thinks that he/she possesses a
formulation to articulate it, at the very least, then others can test
it out to see if it would resist attempts to make unfair dice look fair.
Derek Ray
> Derek:
>
> This is so simple I did not think anyone would have a problem
> comprehending it, nor creating a similar game.
I shouldn't have to repeat myself on this issue, Professor.
"Repeatability". It is the watchword of the scientific community, one
that you clearly make false claims to belonging to.
If you had posted a match record, then anyone else could have duplicated
your EXACT steps and done the same thing. The exact steps matter. I
can surely perform a similar experiment, but then we introduce human
bias into the equation,... so that if I obtain different results than
you, we actually prove nothing.
We must both be using the exact same steps and looking at the exact same
match record, and only then can we have a reasonable discussion over
what EXACTLY you think is wrong with Gnu's evaluation in that situation.
> has to pass. What you get is GNU giving itself a "good dice, man"
> rating along with "world class" play, but it loses anyway, even though
> you play "awful."
I note that you did not respond to my direct challenges about what
exactly regarding this evaluation you considered wrong -- especially the
part where I pointed out that a single hit from an ace point holding
game is hardly "decisive" in and of the sense that a player goes from a
mostly-won position to a surely-lost position.
Please respond to the following directly and without evasion:
/self-quote begins/
> What do you believe is incorrect about GNU's evaluation?
>
> -- You admit to having intentionally played "awful".
> -- GNU probably did play at world class; it tends to.
> -- You say you controlled the dice, so you probably gave it some good
> rolls to help it get into a strong position against you; this would
> indicate "good dice, man". We can't know for sure without the match
> record that you so carelessly failed to post for us.
>
> And most importantly: Did you continue the game after your hit? Just
> making the initial hit from that position does not guarantee a win... in
> fact, far from it. You will need some further good rolls, as well as
> your opponent not to roll any big jokers, in order to come back and win
> the game from there. Again, if you had bothered to include the match
> record, we could have done some PROPER analysis, rather than your
> half-assed conclusions.
/self-quote ends/
You didn't answer the questions, Professor. I challenged your
experimental method; you need to defend your method before babbling on
for a couple pages about "conclusions".
> forth, there needs be analysis. Backgammon "luck" needs to be
> formulated in a way that is sensible, and so far I have yet to see such
> a formulation.
Gnu's formulation makes sense and is effective. Post your steps and
match record, and I will duplicate your results, and then we will
discuss it, and I will explain why it works to you.
> Here is an area where the scientific method can be
> used,
Yet you refuse to use this method yourself.
REPEATABILITY. A first year science student learns this as part of his
initial class. Why do you resist it so strongly?
--
Derek
...It must be terrible to be a bitter old man, learning that all the
assumptions you've made for your entire life are actually wrong... and
discovering that wisdom isn't a door, but instead a path leading ever
onward through many doors.
Raccoon
On Dec 28, 8:36 am, "HarryPotter" <zzyzx...@gmail.com> wrote:
> It could be something simple like they increased the #
> of doubles thrown to increase the automatic doubles and thus their
> rake,
If you have played 300-odd games then (figuring an average game at
about 20 rolls per player) you and opponents have rolled about 12,000
rolls. If you have saved your matches then you probably have enough
data to determine whether an unusually large percentage of the rolls
are doublets.
mont...@lycos.com
remember saying that I thought I might be getting close to playing on
the highest level. I will clarify: I don't know, nor care, if I can
ever beat GNU on a high setting in a long match. I am talking about
the level at which the players in the final at Monte Carlo were playing
at this year (I was able to watch several of the games of that match
via TMG). As long as I can hold my own against the best humans in the
world, I will feel that playing for higher stakes may be worthwhile,
even with high rakes and perhaps an occasional cheater to deal with.
Once again, I'll say that I think that I am close to this level, if not
already there. I "fine tune" my understanding of the game almost every
day, and this usually takes up the time I would be using to play a lot
of matches for let's call them "medium level stakes." If I play in a
couple of freerolls, then analyse the games, it often takes up quite a
bit of the time I provide myself to engage in this hobby.
As to the game I created with GNU, as I said, anyone beyond the
beginner level can do the same, and see how GNU evaluates it. It will
take a few minutes only. However, there is no copy and paste for the
moves, only the game file itself can be "moved around." Moreover, it
only takes a bit of common sense to see how a site could create
software to give weak players the one "lucky" roll (which I called
"decisive luck") every once in a while, to keep things fairly even,
thus not losing the weak players.
Raccoon
On Dec 30, 12:20 pm, monty1...@lycos.com wrote:
> Once again, Rich distorts reality, for whatever reason. I distinctly
> remember saying that I thought I might be getting close to playing on
> the highest level.
Monty: "I don't claim to be a "world class player, though I would not
be surprised if I was fairly close at this point."
Monty again: "In using the GNU analysis feature to examine a number of
recent matches, what I noticed is that I am often rated 'intermediate'
in checker play."
I appreciate the optimistic view that an intermediate player might be
Robi
"Raccoon" <racg...@yahoo.com> schrieb im Newsbeitrag
news:1167511657.6...@k21g2000cwa.googlegroups.com...
Simon Woodhead
> Once again, Rich distorts reality, for whatever reason. I distinctly
> remember........
and on and on and on and on
Why don't you just post the match logs ? What's wrong with
you that you can't manage something as simple as that ?
Derek Ray
Hash: SHA1
mont...@lycos.com wrote:
> As to the game I created with GNU, as I said, anyone beyond the
> beginner level can do the same, and see how GNU evaluates it.
Let's play a game. "I did that. I got the opposite result from you.
Therefore, you're wrong."
See how meaningful that was, since we used totally different moves, dice
rolls, and GNU settings? Repeatability, Professor. Every science
student learns it. Perhaps you weren't a science student after all...
Now would you like to post your game log so there can be a productive
discussion? Or are you not interested in productive discussion and
instead would just like to hear yourself talk -- in which case I
recommend a tape recorder.
> However, there is no copy and paste for the
> moves, only the game file itself can be "moved around."
Learn to use the software before running your mouth.
File.
Export.
Select "text file".
Pick a name.
Click OK.
Open that file.
Select all.
Copy.
Paste.
Anyone with half a brain could figure THAT out.
- --
Derek
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2.2 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQFFlvwttQZlu3o7QpERAm23AKDX+uESIdypinbTyJ5rXQM18Vj0ygCglf/K
D5c45hRfRJDqZYcV40S4Zs0=
=43V+
-----END PGP SIGNATURE-----
Unknown
Can you post your game log so we can all see it.
If you didn't keep it, can you perform the same steps over again and
then post the game log here.
It should take you a few minutes only as you said.
HarryPotter
played a fair amount there and came to the same conclusions I did. It
seems I found one before this thread got sidetracked.
There is no provision for saving the games, and the other players play
so fast there was no time for them to be using Snowie.
And I have played vs. Snowie daily for years now, and play at a world
class level (average error rate under 4.0). I know how I was playing
and I know how bad the opponent's cube was.
I can add this info from a reliable source: BGMaster's headquarters is
in Israel and the govt there is shutting down the online poker sites.
BGmasters days look numbered. From the same source, I heard another
player had a very difficult time getting their money out when they
tried. In my own experience, I did make a withdrawal and I got my money
(temporarily as it was) but there were little problems: 1st, it takes
weeks to ask them to send you a check. They had another option to
credit it to your credit card but you had to email them and them call
you on the phone. They didn't call for 2 days after they said they
would and in that time I quit playing and sent them daily email
reminders.
The reason I asked for my balance back was because twice in one night
on a solid broadband connection, I got disconnected and disqualified in
a game worth $200 and one worth $80. I told them my connection was
working ok and they said theirs was ok, but as a goodwill gesture they
agreed to refund the whole amount in question (more than my equity but
they never asked about that). But they didn't do it. When I questioned
as to when, I was told they already had. But in this case I had quit
playing until it was resolved and knew my exact balance and they
hadn't. When I cited the #'s back to them, they then credited me the
amount in question.
This is either very sloppy or they are playing me for a fool. And I'm
surprised they agreed to give me back the whole amount unless they
were pretty sure they would be getting it back very shortly with
interest, which they did many times over.
I'm really confused because there are so many players on there with
high ratings. Are they mostly bots? Or did they get their success by
not playing the lowest rated players? Could it be there are just a few
real players and dozens of bots at any one time?
I also read somewhere that BGMasters caters mostly to a Middle Eastern
Clientele. The history of them is that the game has been played there
for centuries but without the cube which was a 20th century invention.
Many of them, I'm told are poor with the cube. Perhaps this explains
players who play fast and have bad cubes. But it could also mean a bot
programmed to play with a live cube strategy built in even when the
cube is dead. Wh said their programmers worte the whole bot? Maybe it
came in a software package they bought. This was the most common and
glaring error and I was gleefully thinking how easy this would be to
beat this game. I assumed I was playing your typical experienced
chouette player who learned the game from watching others and never
really thought much about the why's of things. Nobody in a normal money
game or chouette can ignore the possibility of getting gammoned or of
getting redoubled if they give the cube away. These are common
occurances on BGMasters.
Again, I'm hoping to get some opinions from others who have actually
played there in the $20 and $50 games
HarryPotter
the lowest rating player I could. So after those many times I just got
the game stolen from me at the lat instant, they winner of $200 would
disappear back to the room and not play me again. I figured ok, they
know how lucky they were. In the meantime, I would get invitations from
other players, some of them also with low ratings. But often I would
refuse them temporarily, because I wanted to check the ratings of the
other players in the room.
Many times, I would see that my best game theoritically, was the player
who had just moments ago invited me and I had refused. So I would
invite them, and as far as I can remember, the nearly unamious answer
was no. I was thinking this was some perry human 'mini-revenge' issue
and didn't think about it until recently. Because after a couple
minutes if they weren't playing someone else they would invite me
again. These kinds of actions, would be easy to program and would tend
to give a human attribute to the bot. And it makes sense. This kind of
thing, like the taking back of a move, is something the programmer
could do without tampering with how the bot actually played the game,
which may have come in a purchased package and beyond the talents of
the local programmer. And that could be why it played with such a bad
cube and had to resort to rolling whatever it needed to beat me.
HarryPotter
cat I guess. I'm really searching for the truth here, not trying to
alabi a big loss. So I tried to play players that I didn't think could
be bots. The first game was text book: Automatic Double, good start for
them, bad start for me, a gamonish double to 4 ($200) but they had most
of their spares on my side of the board, I had one in the air, and them
a 3 point board. All my checkers were in play and stacked with spares.
I may have had a 2 point board. A clear take with a dead cube.
So I took, they rolled double 4's. I danced, they rolled double 5's I
danced and danced and danced, etc. it's happened so many time it;s like
a broken record there.
So I lost again and bought some more. There were too many doubles
thrown. So for the final 10 games I played, I kept track. Here's the
results (doubled thrown-total rolls):
18-57
6-17
2-13
2-10
1-10
5-30
9-24
5-19
15-59
8-24
Totals: 71-293
That's 24.2% doubles thrown
It should be 16.7%
Not enough data to be tooo significant, except it's pretty close to my
gut estimate from last month.
I wonder if the successful players maybe have figured this out and
adapted their play to account for it? I did notice that my rating in
the $20 game sank to 1300 and stayed there. That's as low as you can
go. And you seldfom see a player with a rating over 1600.
Your ratings change from room to room. I'm certain I was in the low
1300 in the $50 room last week, but last night I won a couple games and
it's suddenly over 1500. Hmmm. I also checked my Visa statement and see
I'm down 7 grand. And my rating is over 1500 (where you start) What a
load of crap.
Grunty
> Last night I bought some more chips on BGMasters. Curiousity killed the
> cat I guess.
> [....]
> Your ratings change from room to room. I'm certain I was in the low
> 1300 in the $50 room last week, but last night I won a couple games and
> it's suddenly over 1500. Hmmm. I also checked my Visa statement and see
> I'm down 7 grand. And my rating is over 1500 (where you start) What a
> load of crap.
I had played a few at BGM time ago, low stakes. I'd say they were
meat&bones players, there were rather few players back then.
Earned some bucks from their initial bonus. Then I lost many ridiculous
games to bad players, games easily spotted as rigged - two classical
types: my primes being skipped just-in-time by one or two men, or them
getting an early anchor and within the 4-5 next rolls getting the big
doublet, soon followed by the next decisive big doublet.
Sure such games won't be uncommon to any bg player, but I decided it
was a bit too many of such games, over a bit too little time, against a
bit too bad opponents.
Yes I can recall myself winning some crazy game as well, but nowhere
nearly as many as those crazy games lost to fish.
Two further things reinforced the bad smell.
* BGM had no savegame feature. This made it impossible to replay those
games. In other words, no proofs left.
* BGM had inforced automatic doubles. This is an interesting feature.
They could control at their convenience the frequence and actual
occurrence of the duplication of the stakes. This would allow them to
control the cash flow in a session. Many times, after winning a 2 or 4
game, an automatic double would come up and sure enough, I was doomed
to lose that game, thus restoring the cash balance somewhat.
At the point all that reached my disgust-threshold I just withdrew my
money.
I'd love to have some spare grands like you, to throw into those
slotmachines.
Gregg Cattanach
>
> What KW thinks of XG is irrelevant here: the discussion has to do with
> how to evaluate "luck" in backgammon. If this cannot be agreed upon,
> then all anyone can do is to present "impressions," and thus the onus
> is on you to articulate what "backgammon luck" is and how to evaluate
> it in a way that consistent with established principles (math, logic,
> etc.). Because I don't know if there is a way to establish a useful
> "backgammon luck" doctrine or formula, I can only put forth
> impressions.
The backgammon 'luck formula' has already been established and is in use by
both major bots. The average equity of all 36 rolls is evaulated, the
equity of the roll (if played correctly) is compared to that average, and
the difference, plus or minus because the positive or negative luck for that
roll. Every roll is evaulated that way and the total for the game, match or
individual roll is reported.
Gnu reports two numbers, the luck based on player A's rolls and player B's
rolls. Snowie combines these two values to a single luck number, positive
for the overall luckier player and the same number but negative for the less
lucky player.
Your contention that there is no backgammon luck formula is false. We do
not have to base anything on 'impressions'.
--
Gregg C.
Gregg Cattanach
> As I've said, I agree with the
> person who said that XG's dice appear "bouncey," but before any claim
> can be put forth, there needs be analysis. Backgammon "luck" needs
> to be formulated in a way that is sensible, and so far I have yet to
> see such a formulation. Here is an area where the scientific method
> can be used, in that any formulation put forth can be tested by
> creating games, then considering if software could be manipulated to
> make things look "fair."
Your 'bounciness' theorem is trivial to prove or disprove. Collect a large
sample of XG games and a similarly large sample of live games. Compare the
number of jokers in each, (Snowie or Gnu can both do this and you can adjust
the threshold for joker or 'very lucky'). If the dice are 'bouncy' (I
assume this means more prone to big swings), then XG will have far more
jokers/anti-jokers for both players compared to the live match control set.
--
Gregg C.
Gregg Cattanach
<snip>
> I also read somewhere that BGMasters caters mostly to a Middle Eastern
> Clientele. The history of them is that the game has been played there
> for centuries but without the cube which was a 20th century invention.
> Many of them, I'm told are poor with the cube. Perhaps this explains
> players who play fast and have bad cubes. But it could also mean a bot
> programmed to play with a live cube strategy built in even when the
> cube is dead. Wh said their programmers worte the whole bot? Maybe it
> came in a software package they bought. This was the most common and
> glaring error and I was gleefully thinking how easy this would be to
> beat this game. I assumed I was playing your typical experienced
> chouette player who learned the game from watching others and never
> really thought much about the why's of things. Nobody in a normal
> money game or chouette can ignore the possibility of getting gammoned
> or of getting redoubled if they give the cube away. These are common
> occurances on BGMasters.
Much of what you say may have some basis in fact (but I seriously dispute
the notion of 'house bots'), but be sure you understand the cube
considerations with the 4-point limit. Once the cube has been turned once,
the redouble can and should come quite quickly as it cancels all gammons and
forces the game to be played to conclusion at the 4 level. It bears a lot
of similarities to the cube in a 2-away, 2-away match situation.
--
Gregg C.
Gregg Cattanach
> I too have stopped playing at BackgammonMasters. Way too many
> automatic doubles, and the fact that they pulled out on short notice
> of sponsoring several ABT events with no explanation doesn't lend me
> to feel comfortable about the site. I contacted the woman who was
> representing BackgammonMasters in Las Vegas to ask if their changes
> were a result of the new US law, and she supposedly forwarded my
> question to BackgammonMasters. I have NEVER received any reply. Perhaps
> someone who is involved in the ABT events effected (Greg C?)
> could shed more light.
They claimed that their withdrawal from ABT sponsorships was because of the
US 'ban' on on-line gambling, but I have a pretty strong feeling that they
made the decision because that they were getting very limited bang for their
buck as far as new traffic on their site from the sponsorships, and that the
US law was a simple excuse to end their live-tournament sponsorship deals.
AFAIK, they are still happily taking credit-card deposits from US customers.
I made one last week myself so I could play in all of their holiday
'free-rolls'. (not free, but >>100% return tournaments).
--
Gregg C.
HarryPotter
Grunty wrote:
Well my experiment is over. Your experience was as if I'd written it.
Last night I dumped a grand in about an hour, I don't think I won more
than 1 game. I played 5 or 6 different players and I don't think any of
them were bots. Some of the names were familiar, some have played in
the free rolls (although BETTY FORD seemed to enter all of them... when
does 'she' sleep?). This time, my opponents occasionally took a moment
to think.
But I was like an empty chair, and the doubles kept coming. Theirs were
good, mine were blocked but they happened all too often.
I'm starting to fantacize about humans playing for the house with
special software to either roll ramdom dice or enter a roll themselves
(with instructions from management not to overdue it). As usual, I'm
not making accusations, I'm trying to arrive at the truth. But my
experiments are over. It was too consistant, and too often that I could
be outrolled like I was. This was my first experience at online BG for
money. I'd stayed away before because I was worried about someone using
Snowie to play for them. When I was shown BGMaster's system of time
pressure for each game, and saw how fast the players were playing, I
figured it snowie wasn't a problem. I had no thoughts that the house
would be running a big scam. It's like Vagas cheating you: they don't
need to. BGMasters takes a healthy commission from each game.
But now I have my doubts. I lost about 200 points I'd say altoghther,
my average stake starting at say, $35. With all those automatics, that
doesn't seem like more than a bad streak. Surely I've seen players lose
200 points in a Chouette in one night. But I've never seen the # of
doubles thrown at 25% of the total rolls for a week, and I've never
lost so many games the way Grunty describes it above. I'm going back to
playing Snowie, and the occasional live tournament.
Grunty
> I'm going back to playing Snowie, and the occasional live tournament.
Harry,
maybe too late for you, maybe not, please read on.
I have a long record here as the group's paranoid voice.
For years I've fighted to and been fighted by rationalists,
scientificists, statisticians, all sort of math/sci mindsets, and even
some shills of course, about these obscure matters.
I've played on many sites, on some much longer than others, never
losing a buck. On *every* of them I became convinced of some tricks
going on, though at different degrees.
If I had to recommend just ONE site, I'd name TrueMoneyGames.com.
Not that I don't have issues with them (I long since have, though not
as blatant as other sites). But all in all, they're the best
(technically, graphically and otherwise) and most serious site to play
bg for money. Just my opinion.
I guess at this point your Visa statement could easily permit you to
give it a try. Just take it easy for starters, try not highest stakes,
say play $10 through $30 per point, $30 through $50 matches) and see
"how it feels."
>From my own experience there, I can guarantee you only one thing -- if
you're good *and prudent*, you will likely end up a net winner.
Robi
play some games vs me on another server? 50$ per point is okay for me.
please let me know
"HarryPotter" <zzyz...@gmail.com> schrieb im Newsbeitrag
news:1167764903....@a3g2000cwd.googlegroups.com...
HarryPotter
Robi wrote:
> harry, i have the feeling you are not so good player like you think. wanna
> play some games vs me on another server? 50$ per point is okay for me.
>
> please let me know
>
Robi
We can play for other stakes if you want.
Robi
"HarryPotter" <zzyz...@gmail.com> schrieb im Newsbeitrag
news:1167820975.7...@v33g2000cwv.googlegroups.com...
HarryPotter
Americans don't put the $ after the number so you must be elsewhere.
That's all I need is to travel the world to collect my winnings. Thanks
but from now on I want to see the dice do their dance.
Grunty
"Harry Potter and the BotGammonMasters."
HarryPotter
Gregg Cattanach wrote:
>
> Much of what you say may have some basis in fact (but I seriously dispute
> the notion of 'house bots'), but be sure you understand the cube
> considerations with the 4-point limit. Once the cube has been turned once,
> the redouble can and should come quite quickly as it cancels all gammons and
> forces the game to be played to conclusion at the 4 level. It bears a lot
> of similarities to the cube in a 2-away, 2-away match situation.
>
could get a drop from almost everybody even though they had 35% winning
chances (gammonish positions but so what?) I didn't double hoping to
get between 65 and 75% , and take my 2 points. And if I got too good,
that was fine also.
I have been watching games today and recording the ratio of doubles. So
far it looks ok. Strange there are not many of the same players I was
playing.
HarryPotter
or tails of this please:
In the $20-$80 game I start watching FRANK78 play HOTGUN. but Frank78's
timer is in the red and he's not doing anything. So after about 5 more
minutes, He times out and Hotgun wins a backgammon by default ($80).
Frank78 now didn't have enough $ to stay in that room so I went to the
$50-$200 room and started watching BGFLYER - KICKME. Nothing is
happening. It's BGFlyer's move and his timer is in the red. About 10
minutes later he times out and loses $200, the max. However, a new game
is immediately started between the same 2 players. Kickme goes first
and moves, and BGflyer doesn't play his roll and times out in about 10
minutes and loses another $200. This time however, they both go back in
the lobby of the same room. Within 30 seconds BGflyer is playing
AVIGAIL a short game and he's soon back in the lobby where he plays
Electron, takes a double and finishes the game.Back to the lobby,
BGflyer plays Skyteam. In the middle of their game, BGflyer again times
out in 10 minutes of inactivity and loses another $200. This time
BGflyer disappears after the game ends and Skyteam playes MONO19, then
moves to the $20-$160 lobby and plays 3 games there with average
results, and disappears.
Derek Ray
Hash: SHA1
HarryPotter wrote:
> I just got done watching games for about an hour. Somebody make heads
> or tails of this please:
>
> Frank78 now didn't have enough $ to stay in that room so I went to the
> $50-$200 room and started watching BGFLYER - KICKME. Nothing is
> happening. It's BGFlyer's move and his timer is in the red. About 10
> minutes later he times out and loses $200, the max. However, a new game
> is immediately started between the same 2 players. Kickme goes first
> and moves, and BGflyer doesn't play his roll and times out in about 10
> minutes and loses another $200.
It looks like a money transfer, done in such a fashion that it won't
trip any automated server processes looking for things like "Start game,
resign, start game, resign, start game, resign", etc.
The reasons behind this are left as an exercise for the reader to
determine, but it would help to start from the possibility that BGFlyer
and Kickme may not be two different players.
- --
Derek
insert clever quotation here
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2.2 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD4DBQFFoDdDtQZlu3o7QpERAjbSAJjXfQWYCL2CYFPCe59qnBdx8qm1AJ9uUU9H
RuE43xTv/uZ0KxkxIhiJKw==
=nj/a
-----END PGP SIGNATURE-----
HarryPotter
house)
Derek Ray
Hash: SHA1
HarryPotter wrote:
> That's an expensive way to transfer money (unless they work for the
> house)
Man, you CTs will see ghosts _anywhere_, won't you?
If they (he?) worked for the house, why would they need to wait ten
minutes and let the game time out as opposed to immediately resigning?
For that matter, why would they need to resort to such a subterfuge at
all? Why wouldn't the house just move the money for them?
No. What you saw is a quite common practice on both backgammon and
poker servers, and it's done entirely by the players ... and the intent
is, in fact, to launder money in a fashion undetected by the house.
- --
Derek
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2.2 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQFFoHhztQZlu3o7QpERArq5AKDcee9lUuHJie3X//CpiDGz2/4GDgCg5g4f
nEN2op7gfy/WTvaFwf7U5UU=
=wsU8
-----END PGP SIGNATURE-----
HarryPotter
>
> For that matter, why would they need to resort to such a subterfuge at
> all? Why wouldn't the house just move the money for them?
>
"THE HOUSE" does not have to an entire company of conspirators.My
defination is someone or a group who has access to their software and
may be cheating
Derek Ray
Hash: SHA1
Your definition doesn't match anyone else's in the world, then.
The rest of the world uses "The house" to mean "The server and the
organization running that server", much like referring to "The house" at
a casino means "The casino, its owners, and all its employees" (and
props/shills, if you're in Mississippi).
If you're thinking of a small group of people who are performing these
transfers without the house's knowledge (because that's the only reason
to pull those weird stunts, is to try to avoid any automated detectors
the house is using) then you should say so, rather than trying to be
confusing about it.
- --
Derek
insert clever quotation here
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2.2 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD4DBQFFoQwxtQZlu3o7QpERAk8VAJj3sGUQ6GeJF+uNfmAUPvlOq1tcAJ9kEji4
IJqiHn0kjp5V4F1OAfoHrg==
=sHvP
-----END PGP SIGNATURE-----
kentcov...@gmail.com
> I'm a good BG player. I've won big events and played for 30 years. I've
> played with Snowie the past few years a LOT so I'm upto date on the
> latest information. I was at the Las Vegas event in November and
> Backgammonmasters was there and I talked to the rep who assured me,
> with the timer they have on the game, I would not be playing someone
> with Snowie playing beside them at home. So I signed up.
>
> After dumping 5 grand in a month, I've come to the following
> conclusions: I'm either having the worst run of bad luck in the history
> of the planet, or I'm being cheated. I've played perhaps 300 or more
> games, in the $50 and $20 limits. I have the lowest ratings of anybody
> there and I've just been playing others with low ratings (1300's and
> 1400's). Here are the facts:
>
> I've played with maybe a hundred different players. After a while i
> began to write down names and rate their play. Most of them play
> identically: Their checker play is ok, good even, but their cube play
> is allways based on 'live cube' decisions. However, in these limits
> the most you can win or lose per game is 4 points, no matter what.
> These players drop most cubes which are the least bit gammonish, which
> is a huge blunder and especially in games with a dead cube.
>
> There are many automatic doubles (not declinable), and that was my
> first suspicion something was wrong. It seemed like much more than one
> game in 6 was an auto. So with the cube in the middle in so many games,
> if you are doubled, you not only have no redouble vig, but you can't
> get gammoned either, so the take is exactly 25%. Yet my opponents would
> drop like flies when I doubled a gamonish position when they were 35% a
> winner.
>
> Next, they would give me countless free rolls with the cube in the
> middle at any value. How could I be losing? At first I just thought bad
> luck. There were so many 17-1 shots hitting me at the crucial time. And
> I was getting killed in close races. It seemed the opponent was rolling
> doubles every 4 or 5 rolls, not one in 6.
>
> As for my cube taking, I began to doubt myself so I took screen shots
> of all the positions where I was doubled. Nearly all my actions were
> correct and my misses were off never more than 1%. When I'd take a
> gamonish but correct take, what would happen was the opponent would
> start rolling doubles every 2 or 3 rolls and close me out and gammon me
> time after time after time. And once I had a 2 or 4 cube on my side I
> would dance on a 3 and 4 point board 4 or 5 times in a row until the
> opponent was all over me. I'm not talking about a couple games or 1
> session. I'm talking about 300 games or more.
>
> I got really suspicious just this week. There is a chat window so I
> started saying hello to my opponent. I did this to dozens of players. I
> never got a response. I'd even stop in a bearoff where I was going to
> lose and not worry about the timer going off, and message them over and
> over and over. NEVER have I got a reply. So I'd wait 2 or 3 minutes,
> then make my move. Their roll was always instantanious. Don't you think
> one of them might have glanced away for a second after such a delay?
> Then I started thinking some other things: These players all play
> without hesitation. I know there is a clock but it's not like you are
> going to run out of time if you take a few seconds now and then. These
> opponents never took any time. And on cube decisions: Some of them were
> close. A human would occasionally pause to evaluate. A computer's
> pause is the snap of a finger. The cover for this is occasionally the
> opponent would play the checkers one way, then bring them back and redo
> it correctly. However, the moves they took back were seldom valid
> plays. It was almost always like you clicked on the wrong checker by
> mistake. Was this a smokescreen? Take away this seemingly human
> maneuver and there is nothing human left about my opponents.
>
> Could it be I was not only playing bots (whose cube was bad) but that
> they had control of the dice as well? And why were there so many
> different players like this? I think it's a human reaction that there
> couldn't be that many humans cheating you on one site. But for a
> computer that's nothing. They could have hundreds, change their names
> often, keep their ratings below 1500 with a click of the mouse, but all
> this would have to be if Backgammonmasters was doing it. An outside bot
> wouldn't be able to control the dice.
>
> So am I delusional, or incredibly unlucky or what? I'd sure like to
> get some feedback from some other players there that have played in the
> bigger money games. Please tell me if you noticed any of the same thing
> I have.
>
> Zzyzx000
I've thought for many years that the dice at Backgammanmasters.com are programmed. I play on the free site.