Another angle on cube ownership
muratk
wrote:
> 1. The cube value 1 has the special property that both players can cube, as
> opposed to higher cube values, when the cube is always "owned" by one
> player. Owning the cube is -- all other things equal -- a significant
> advantage: the opponent can't cube you (out) when things turn bad, so that
> you can wait for a swing and thus use the few percent chances to win from a
> bad situation, or at least don't lose at a higher stake.
The above quote comes from the thread "Computer Games Cheating".
Obviously the "swing" that he is talking about refers to "luck" and
"cubing
out" refers to capitalizing on a "temporary swing of luck".
Whether the author realizes it or not, the above statement is an
admission
of cube's magnifying the element of luck in backgammon.
Remembering the arguments that luck (and relatedly, shorter matches)
favor (help) the weaker player, we can conclude that cube favors
(helps)
the weaker player.
If cube ownership helps prolong the game, allowing to wait for another
swing of luck, (i.e. "luck evening out"), then we should be able to
argu
that playing without the cube has the same value as owning the cube
when playing with the cube.
It would be absurd to use the word "swing" when referring to checker
play.
Even though the weaker checker player with an average "cube skill"
can "cube out" the stronger checker player based on "swings of luck",
in the long run the stronger checker player will win, playing with or
without the cube...
Clearly the so called "cube skill" is external to backgammon and only
a means of enhancing the gambling experience...
It doesn't take more skill to play backgammon with the cube and it
doesn't take less skill to play backgammon without the cube.
Personally, like the "Classic Coke", I would like to see more people
return to enjoying the unadultered taste of "classic backgammon"...
MK
Patti Beadles
muratk <mu...@compuplus.net> wrote:
>Whether the author realizes it or not, the above statement is an
>admission of cube's magnifying the element of luck in backgammon.
The doubling cube magnifies both the luck and the skill in
backgammon. It's not just luck.
The other day while playing online, I was in a fairly strong
position where I probably was going to gammon my opponent. I
was playing on autopilot while doing other things, and made an
extremely bonehead play. Rather than playing safely, I left
a triple shot and broke a prime. The move was just d-u-m-b.
My opponent correctly whipped the cube back at me as soon as my
play was done. There was no luck in this swing, just (lack of)
skill.
This is an extreme example, but things like this happen on a smaller
scale all the time.
>Remembering the arguments that luck (and relatedly, shorter matches)
>favor (help) the weaker player, we can conclude that cube favors
>(helps) the weaker player.
Not necessarily.
I believe that cube favors the stronger player, because (barring
pure stupidity like my play above) the size of the mistakes that
players make when dealing with a cube are often much larger than
the ones they make in checker play.
>It doesn't take more skill to play backgammon with the cube and it
>doesn't take less skill to play backgammon without the cube.
This statment asserts that there is zero skill involved in knowing
when to offer or accept/reject a cube. That is clearly false.
-Patti
--
Patti Beadles, Oakland, CA |
pattib~pattib.org | All religions are equally
http://www.pattib.org/ | ludicrous, and should be ridiculed
http://stopshootingauto.com | as often as possible. C. Bond
Doctor Bob
Proper utilitzation of the Cube separates the men from the boys. Even
when I am having bad luck against a computer, I can often get back
into the match by using the cube wisely. My score against GNU is
definitely better than 50%. Do I play better? Like anybody,
sometimes I get a little careless and make mistakes. GNU never does.
Many times I will be looking at his position and think he will play
this because it is best. Invariably he does. However, he is whimpy
when he is ahead and often he underestimates my chances.
The Cube is also great for avoiding bad luck. You need to end games
early when don't have good gammon possibilites.....
Dr. Bob
tc...@lsa.umich.edu
Doctor Bob <r.d.b...@gmail.com> wrote:
>Proper utilitzation of the Cube separates the men from the boys. Even
>when I am having bad luck against a computer, I can often get back
>into the match by using the cube wisely. My score against GNU is
>definitely better than 50%. Do I play better? Like anybody,
>sometimes I get a little careless and make mistakes. GNU never does.
>Many times I will be looking at his position and think he will play
>this because it is best. Invariably he does. However, he is whimpy
>when he is ahead and often he underestimates my chances.
It's true that GNU probably won't beat you as badly as a very skilful
human opponent will, but this isn't so much because GNU "underestimates
your chances." It's because GNU doesn't pick up on your weaknesses and
play accordingly, whereas a strong human will.
There are some players in my circle who always double too late and drop
too early. Against such players I will double as soon as I think they
will drop, even if against a strong opponent I would not double. Similarly
I usually have no trouble with take decisions (at least for money games)
since they never double until it's an obvious monster drop.
Conversely, there are players who take too deep. Against such players
I will try to hold off doubling until their point of last take.
GNU won't play such tricks. So unless you're an extremely strong player
you will probably do better against GNU than against a very strong human
who can pick up on your weaknesses and exploit them.
--
Tim Chow tchow-at-alum-dot-mit-dot-edu
The range of our projectiles---even ... the artillery---however great, will
never exceed four of those miles of which as many thousand separate us from
the center of the earth. ---Galileo, Dialogues Concerning Two New Sciences
Doctor Bob
> In article <3372b5fc-af5d-46c3-b7f4-681e0736c...@g21g2000yqk.googlegroups.com>,
That is why I mix it up sometimes. For example, I may drop a position
which is a weak take, or I just don't feel like waiting for a blot and
then hoping to hit it. Sometimes they will start to try to pick up
quick points by throwing the cube early or will start to take
questionable positions in an attempt to "show me" when to turn the
cube.
Lately, I have been playing matches with the bot so my cube play is
usually heavily influenced by the score.
Finally, in the short term you can't always go by the numbers. There
is definitely luck involved. Sometimes you have to go by feel.......
Dr. Bob
tc...@lsa.umich.edu
Doctor Bob <r.d.b...@gmail.com> wrote:
>Finally, in the short term you can't always go by the numbers. There
>is definitely luck involved. Sometimes you have to go by feel.......
There is definitely luck involved. And sometimes the numbers are not clear.
But to go by feel *against* what the numbers say---well, there's a word for
that. It's called losing.
muratk
> muratk <mu...@compuplus.net> wrote:
>> Whether the author realizes it or not, the above statement is an
>> admission of cube's magnifying the element of luck in backgammon.
> The doubling cube magnifies both the luck and the skill in
> backgammon. It's not just luck.
This should be rather easy to test. Bots already analyze each
roll for luck and each move for checker error. Whenever a cube
action occurs (or "correctly" should occur according to the
bot), we can make it loot at the immediately preceding dice
rolls and checker plays, so that it can keep statistics on
what "triggers" a cube action more often; (i.e. luck or skill)?
> extremely bonehead play. Rather than playing safely, I left
> a triple shot and broke a prime. The move was just d-u-m-b.
> My opponent correctly whipped the cube back at me as soon as my
> play was done. There was no luck in this swing, just (lack of)
> skill. This is an extreme example, but things like this happen
> on a smaller scale all the time.
As you said, your example is extreme but it wouldn't hurt to
count them all, whether extreme cases or smaller scale cases.
One difficulty would be to determine if an extreme case error
like your example is not indeed intentional...
Depending on the strength of the players, one player may want
to have the cube back in his hand so badly that he may make
such a bad move on purpose to sucker his opponent into cubing.
This is the ultimate weakness of the bots. Their strategy is
"metal tasting", very sterile, non-imaginative, non-productive,
etc...
How do you get a bot to properly analyze a sand-bagging move,
if the bot itself doesn't even have such a notion???
Are bots today good enough to differentiate between a genuinely
bad move and a forced bad move? I assume they do and I think it
would be really interesting, for example, to know what percentage
of cube actions arise as a consequence of forced moves. Wouldn't
you folks like to know such things...??
>> Remembering the arguments that luck (and relatedly, shorter
>> matches) favor (help) the weaker player, we can conclude that
>> cube favors (helps) the weaker player.
> Not necessarily.
> I believe that cube favors the stronger player, because (barring
> pure stupidity like my play above) the size of the mistakes that
> players make when dealing with a cube are often much larger than
> the ones they make in checker play.
This can be true if both players believe in the same god but may
not be so true if one is a heathen... :))
As far as I can see, nobody has yet come up with a properly
calibrated measuring stick to be able to make such claims. It
be be true but it may also not be true.
>> It doesn't take more skill to play backgammon with the cube and
>> it doesn't take less skill to play backgammon without the cube.
> This statment asserts that there is zero skill involved in knowing
> when to offer or accept/reject a cube. That is clearly false.
What I mean is making different checker plays in an given
position, based on cubeful/cubeless game, cube value, cube
ownership, (and perhaps other elaborate BS like crawford,
jacoby, goofy, micky mouse, etc. "rules")...
There has been unsuccessful attempts to discuss this in the
past also. I don't know if any human world-class players
are able to do that and if it is something that people can
observe, but I haven't seen any evidence of any bots being
able to do that.
MK
muratk
> The Cube is also great for avoiding bad luck. You need to
> end games early when don't have good gammon possibilites...
Hah hahhh... I find this amusing :)) So, the cube does not
magnify luck for one player but it does de-magnify luck for
the other... :))
MK
Doctor Bob
> In article <11dfe060-69e5-48bb-9d41-14a1a7f17...@z33g2000vbb.googlegroups.com>,
> Doctor Bob <r.d.beel...@gmail.com> wrote:
>
> >Finally, in the short term you can't always go by the numbers. There
> >is definitely luck involved. Sometimes you have to go by feel.......
>
> There is definitely luck involved. And sometimes the numbers are not clear.
> But to go by feel *against* what the numbers say---well, there's a word for
> that. It's called losing.
> --
> Tim Chow tchow-at-alum-dot-mit-dot-edu
> The range of our projectiles---even ... the artillery---however great, will
> never exceed four of those miles of which as many thousand separate us from
> the center of the earth. ---Galileo, Dialogues Concerning Two New Sciences
The numbers are a matter of degree. The second best move in a
position may only have a small effect on the overall probability. I
am just saying that you drop a bad position once in while when you
don't feel like it if you want. You are just guestimating your chances
to take anyway. If you are all stacked up and pissed off, maybe it is
better to clear your head and start a new game.
We are not computers running an algorithm to determine the odds on
taking. We are applying fuzzy logic. There is always something
called "fun". Some positions aren't fun. If there is no money
involved, I might drop especially if the match score is in my favor.
Dr. Bob
Doctor Bob
I definitely use it to end games before they get lucky. I learned
that from JF. One blot trapped behind a four point prime is like
being behind a 7 point prime. Especially when he starts giving your
double 4's and 3's. The first time, I was like: ok, backgammon
happens. A few more times I became a little more suspicious. Where
are those doubles early in the game when I can use them to pickup a
quick point?
Peter Schneider
I'm reading Murat's nonsense through Bob since Murat is in my kill filter.
But sometimes idiots, like children, raise deep qustions, so that this may
be an opportunity to clarify my position on cube and luck.
1. The cube does not enhance the luck factor but it deminishes it.
1a. If the cube is a pass, it's glass clear: The cube ends games where one
player has a great advantage. The other player could, with great luck,
still win; this luck swing is made impossible by ending the game here and
now.
1b. If the cube is a take it's less clear; a luck swing now stays a
possibility and will have a magnified effect. But the likely outcome -- the
favourite wins -- is magnified as well. Since it's the dominant outcome I
assume that the magnification works in favour of the favourite. How many
times have you taken a borderline cube silently cursing it?
2. Handling the cube is, as I argued at some length some time ago with Tim
Chow, more difficult than handling the checkers. The reason is that in
order to find the best move, it is sufficient to just *order* all possibly
resulting positions by their equity; no exact numbers must be produced. All
the contrary with cube decisions: Not only must one come up with a good
estimate of a position's equity, but in a match one also needs to come up
with numbers for gammons and backgammons. To suggest that no additional
skill is involved in producing these numbers simply betrays blissful
innocence.
The Britons who invented the cube were brilliant innovators who did the
game a great service. They turned their fun in placing bets into a
productive extension of backgammon. Every idiot can win a cubeless game.
Best,
Peter aka the juggler
William Womack
wrote:
> Hi,
>
> The Britons who invented the cube were brilliant innovators who did the
> game a great service. They turned their fun in placing bets into a
> productive extension of backgammon. Every idiot can win a cubeless game.
>
I thought the best information for the invention of doubling (which
originally was done without the a cube) was that it was invented in a
club in New York City.
tc...@lsa.umich.edu
William Womack <showme...@gmail.com> wrote:
>On May 7, 4:44�pm, "Peter Schneider" <schneiderp_REMOVET...@gmx.net>
>wrote:
>
>> game a great service. They turned their fun in placing bets into a
>> productive extension of backgammon. Every idiot can win a cubeless game.
>
>I thought the best information for the invention of doubling (which
>originally was done without the a cube) was that it was invented in a
>club in New York City.
By general principles, without any knowledge of specific facts, I feel
pretty confident that "the invention of doubling" is, like "the invention
of television," something that has no crisp answer. The concept of doubling
the stakes in a gambling game must date back to the invention of gambling.
The notion that people must take turns doubling is such a natural idea once
you get some experience with repeated doubling that it must have been
independently rediscovered multiple times. What's left is applying these
ancient concepts to backgammon in particular, which surely must have been
done independently many times.
tc...@lsa.umich.edu
Peter Schneider <schneiderp...@gmx.net> wrote:
>2. Handling the cube is, as I argued at some length some time ago with Tim
>Chow, more difficult than handling the checkers. The reason is that in
>order to find the best move, it is sufficient to just *order* all possibly
>resulting positions by their equity; no exact numbers must be produced. All
>the contrary with cube decisions: Not only must one come up with a good
>estimate of a position's equity, but in a match one also needs to come up
>with numbers for gammons and backgammons.
For those who did not follow my debate with Peter, I will summarize my
position as follows. Playing *matches* as opposed to *money games*
(or "unlimited" games) does force you to be quantitative instead
of qualitative, but this is true both of checker decisions and cube
decisions. If you are not playing a match, however, then you can become
very good at handling the cube without ever estimating absolute equities
or gammon chances or anything like that, simply by learning lots of
reference positions. I personally brought down my money-game cube error
rate significantly by studying reference positions without ever estimating
an absolute equity or a winning probability, and I don't think I am all
that unusual in this regard. Thus I do not believe that handling the
cube is intrinsically more difficult than handling the checkers.
I believe that Doug Zare once did a study and found that most people throw
away about twice as much equity through checker errors than through cube
errors. Part of this is because non-trivial checker-play decisions arise
more frequently than non-trivial cube decisions.
Peter Schneider
"William Womack" <showme...@gmail.com> wrote
<Cough> I was sure enough about what I was writing to write it publicly,
but can't find anything substantial in my memory, bookshelf or the net to
support it, besides Tim's friendly general reservation against single
points of invention. I'm sorry. Thanks for the correction.
muratk
> On May 6, 2:28 am, muratk <mu...@compuplus.net> wrote:
>> Hah hahhh... I find this amusing :)) So, the cube does not
>> magnify luck for one player but it does de-magnify luck for
>> the other... :))
> I definitely use it to end games before they get lucky.
"They" who...?
> I learned that from JF.
Ah, "they JF". When you said "Proper utilitzation of the
Cube separates the men from the boys", perhaps you meant
to say "Proper utilitzation of the Cube separates the men
from the bots" but you mistyped "y" instead of "t"...? ;)
MK
muratk
wrote:
> I'm reading Murat's nonsense through Bob since Murat is in my kill filter.
So, you got to read my "nonsense" by mistake, but why are you replying
to it...?
> But sometimes idiots, like children, raise deep qustions, so that this may
> be an opportunity to clarify my position on cube and luck.
Here I am posting many articles for several days without using
endearing
words for anybody and there comes the arrogant asshole calling me
idiot,
just because I did politely took what came out of his ass and shoved
it
in his mouth (or was it the other way around?? but it's the same
either
way)...
Let's see what historical revelations he will be making... :)
> 1. The cube does not enhance the luck factor but it deminishes it.
>
> 1a. If the cube is a pass, it's glass clear: The cube ends games where
> one player has a great advantage. The other player could, with great
> luck, still win; this luck swing is made impossible by ending the game
> here and now.
The only problem with this bullshit is that it ignores the fact that
the advantage gained by the one player was most likely gained through
luck in the first place!
> Every idiot can win a cubeless game.
I bet you can't...! :)))
MK
PS: I think it's time to discuss different aspects of the so-called
"cube skill" separately. So, I will post several articles about the
issues and suggest variations tests for each. I don't think the test
results will put an end to the debates but I hope that they will carry
the debates to higher levels of intelligence. If you will be
interested
in discussion them, please read each article carefully since I will
use
a lot of cut and paste to avoid retyping but each article will differ
from the other in the details.