Three Monkeyteers of RGB: Tim Chow, BlueDice, Michael...

[m***@compuplus.net]

What makes you primates think that you can pollute the
cyberspace this much...!?

Go have your three-way anal fucking somewhere else, you
obsessive, cage-trained, sick gambling trash... :(

MK

[Tim Chow]

Post by m***@compuplus.net
Go have your three-way anal fucking somewhere else, you
obsessive, cage-trained, sick gambling trash... :(

I love it! Please insult me some more. I live for this!

---
Tim Chow

[m***@compuplus.net]

Post by m***@compuplus.net
Go have your three-way anal fucking somewhere else, you
obsessive, cage-trained, sick gambling trash... :(

I love it! Please insult me some more. I live for this!

This is the limit of my insulting you for free. If you
want more, you'll have to pay for it... ;)

There are limits to what I will do for money also. If
you want to be urinated and defecated on, you'll have
to ask someone else... :(

MK

[Michael]

Post by m***@compuplus.net
What makes you primates think that you can pollute the
cyberspace this much...!?
Go have your three-way anal fucking somewhere else, you
obsessive, cage-trained, sick gambling trash... :(
MK

The villagers over here have a saying:
Should we throw a stone at every dog that barks?

[m***@compuplus.net]

Post by Michael
Should we throw a stone at every dog that barks?

Never mind the villagers, what does XG rollout say
about it...? :)

BTW: it looks like you save "your biggest" bullshit
about "math axioms", etc. for stick's forum, while
Tim Chow, the best person (in fact a genuine math
phd) to discuss such matters is depriving stick's
forum from his wisdom and instead wastes his time
here in RGB discussing positions... What gives??

MK

[Michael]

Post by m***@compuplus.net

Post by Michael
Should we throw a stone at every dog that barks?

Never mind the villagers, what does XG rollout say
about it...? :)
BTW: it looks like you save "your biggest" bullshit
about "math axioms", etc. for stick's forum, while
Tim Chow, the best person (in fact a genuine math
phd) to discuss such matters is depriving stick's
forum from his wisdom and instead wastes his time
here in RGB discussing positions... What gives??
MK

The same thing: Should we hit every blot that barks?
Glad you don't miss a chance to dig in my "biggest" bullshit. :-)

[m***@compuplus.net]

Post by Michael
The same thing: Should we hit every blot that barks?

Ah, now that you reworded it, XG may understand "blot"
that barks better than "dog" that barks.

I got that the question is the same but what is XG's
answer to "Should we hit every blot that barks?"

Surely you can rollout and findout, can't you?

Post by Michael
Glad you don't miss a chance to dig in my "biggest"
bullshit. :-)

"Dig in"? Are you kidding? I wouldn't even know what
a "math axiom" is to begin with, at least regarding
the "cube"...

But surely our math phd Chow would know and it would
be entertainment for me to watch the you of two perform
a discussion on it here in RGB... ;)

Come on boys, don't be shy... :)

MK

[Tim Chow]

Post by m***@compuplus.net
Tim Chow, the best person (in fact a genuine math
phd) to discuss such matters is depriving stick's
forum from his wisdom and instead wastes his time
here in RGB discussing positions... What gives??

Isn't it obvious? It's because only on RGB can I savor the delicious
vituperative that you lovingly heap on me!

---
Tim Chow

[m***@compuplus.net]

Post by Tim Chow

Post by m***@compuplus.net
Tim Chow, the best person (in fact a genuine math
phd) to discuss such matters is depriving stick's
forum from his wisdom and instead wastes his time
here in RGB discussing positions... What gives??

Isn't it obvious? It's because only on RGB can I
savor the delicious vituperative that you lovingly
heap on me!

First of all, why are you answering questions that I
had asked from someone else?

Just a question. Not a complaint. Any answer from
anyone is a better answer than no answer from the
person who was asked (i.e. michael).

Okay, so I understand why you are here in RGB instead
of stick's forum.

But what about the other half of the question? Why
aren't you and michael discuss "math axioms" about
the "cube" here in RGB?

MK

[Michael]

Post by m***@compuplus.net
But what about the other half of the question? Why
aren't you and michael discuss "math axioms" about
the "cube" here in RGB?
MK

There are 4 reasons I posted it there
a) the main one: this forum does not support tables. I could of course upload the tables as images somewhere. But that wouldn't help as one day they would be deleted. I wanted my questions to stay for as long as possible for reference to someone who might think of designing a new bot revising 20+ year old things that might be wrong.
b)that forum is known to have a lot of mathematicians of about equal caliber to Tim's. Unfortunately no one so far expressed any opinion to my question.
c)Tim is already a member there and could participate. So far he didn't. I was planning to invite him to participate giving me his opinion on the subject matter. Unfortunately from the very first response the topic has been diverted to other issues and got totally destroyed.
d)Other than Tim no one else in this forum would care.
the topic is this:
http://www.bgonline.org/forums/webbbs_config.pl?noframes;read=203042#203042

[Tim Chow]

Post by Michael
http://www.bgonline.org/forums/webbbs_config.pl?noframes;read=203042#203042

Thanks for the pointer.

I did not read through the whole thread. However, I gather that your main
question is whether the standard "live cube takepoint" calculation is
"mathematically correct."

The term "mathematically correct" is more subtle than I think you recognize.
Let's start by stating something that is perhaps obvious but that needs to
be emphasized in any careful discussion of this type: Backgammon is not
mathematically solved. So if by "mathematically correct" we mean that the
live cube takepoint calculation yields an absolutely mathematically correct
rule for perfect backgammon play, then it is certainly not "mathematically
correct," for the simple reason that nothing is mathematically correct in
this very strong sense.

In fact, if you really want to speak precisely, even the concept of a "take
point" is not "mathematically correct." If we spell out the assumptions that
are usually left unstated, the term "take point" presupposes that if you are
given the following information---

1. the match score and the cube location, and
2. the percentages for wins, gammons, and backgammons for both sides

---then there exists a number, called the "take point," with the property
that if the win rate exceeds the take point then the correct cube action
is to take, and if the win rate falls below the take point then the correct
cube action is to drop. But this assumption is not "mathematically correct."
Whether the correct action is a take or a pass depends on the details of
the position, and not just on the items listed above. Mathematically speaking,
therefore, there is *no such thing as a take point*.

Take point calculations are *heuristics*. By a "heuristic" I mean a rule of
thumb that does not always hold, but that is relatively easy to apply and
that generally provides good advice about how to play well. We all know a
lot of backgammon heuristics---making the 5pt is usually better than making
the bar point, race when ahead in the race, double in a straight race when
you're 10% ahead in the pip count, etc. With the exception of some heuristics
for non-contact positions, none of these are mathematically proven to be 100%
correct. They're just rules of thumb.

Because take point calculations involve more arithmetic than most heuristics,
they may give the impression of being "mathematically correct." But don't
let the arithmetic fool you. Take point calculations are nothing more than
sophisticated rules of thumb, like JOH's rule of 9/36 market losing sequences.

My point is that asking whether a heuristic is "correct" is the wrong question.
The only appropriate question is whether the heuristic works well, and whether
the assumptions underlying the calculation are believable.

O.K., so let's say we understand all this, and we understand that the take
point calculation can't be "correct." We can still ask, who came up with
the calculation and why did they think it would yield a good heuristic?
The way people have approached this sort of question in the past is that
they have developed *models* of backgammon. That is, they invent some kind
of mathematical process that is simpler than real backgammon but that
resembles backgammon in certain ways. They then analyze the simplified
mathematical model, and use that to obtain a heuristic for real backgammon.

The simplest model for backgammon involves a *continuously* varying equity
that moves up and down randomly. In this model you get a 20% take point:

http://www.bkgm.com/articles/KeelerSpencer/OptimalDoublingInBackgammon/

As Keeler and Spencer discuss, this model is oversimplified in many ways.
So others have tried to come up with more complicated models that are closer
to real backgammon but that can still be analyzed. For a recent attempt of
this sort, see:

https://arxiv.org/pdf/1203.5692.pdf

I'm not sure where the particular formula that you asked about on BGO comes
from originally. It might be from Janowski---

http://www.bkgm.com/articles/Janowski/cubeformulae.pdf

---but I haven't read the paper carefully enough to tell whether it describes
exactly the formula you're interested in.

---
Tim Chow

[m***@compuplus.net]

...........
http://www.bkgm.com/articles/KeelerSpencer/OptimalDoublingInBackgammon/
As Keeler and Spencer discuss, this model is oversimplified
in many ways. So others have tried to come up with more
complicated models that are closer to real backgammon but
that can still be analyzed. For a recent attempt of this

While expecting posts from you two that I could mock at, I
was pleasantly surprised that instead I liked reading what
you wrote.

Thanks for taking the time. Comparing it to what was posted
in Stick's forum helps prove once more that RBG is still
what other forums will have to measure up to.

Let me surprise you back by saying that, in return, I take
back the last five names that I called you... ;)

I also read and also liked Keeler and Spencer's paper. What's
good about their approach is that it's not limited to BG only.
It suits and supports my long going argument that "cube skill"
doesn't add anything to BG itself, as it wouldn't add anything
to hopscotch if you played "cubeful hopscotch".

Let me quote a few paragraphs that I liked and identified with:

----------------------
if the great mass of players is doubling at around 0.6, a player
does better by doubling at 0.7 (rather than 0.8) and keeping....
----------------------

I had made similar arguments based on a "great mass of players"
being trained to play a certain way, both regarding checke and
cube decisions. I love this! :)

----------------------
(A more cynical view is that experts can make more money by
playing their readers than by selling books.)
----------------------

Hah haa! I love this even better. With today's top experts
being bots, nobody who learns BG from bots and try to play
like bots (i.e. with the lowest ER/PR's as judges by the
bots) will be able to beat the bots. You need to play like
a human to beat the over hyped "primitive" bots.

Remember: bots are like trains on tracks but humans can be
like ATV's... You don't need to follow the tracks, idiots!

----------------------
More important is the fact that backgammon is not really
continuous, especially near the end of the game...... The
farther from the end of the game, the less the probability
of winning changes with each move, and the closer to each
other the doubling point and the folding point become.
----------------------

This is just what I have been arguing to you, Chow, never
even having an idea that I didn't know what I was talking
about... ;)

----------------------
What effect does all this have on the practical player?
Our answer is: “Essentially none.” There are some critical
differences between the “pure” game and any real game.
----------------------

This is saying "cube skill is bullshit" in polite English.

----------------------
However, there are “reactive” strategies for A that avoid
infinite expectations. Essentially, one accepts “absurd”
doubles but does not redouble.
----------------------

This brings back so much fond memories, from as far as the
Jellyfish days... :) Anyone else remember those "jacking
up the cube" threads??

Once again, I'll act unlike myself and say "Thanks Chow",
(and Michael;) for allwing me to enjoy RGB this much in a
long time.

Cocksuckers... :)

MK

[Grunty]

I take back the last five names that I called you... ;)
"Thanks Chow", (and Michael;) for allwing me to enjoy RGB this much in a long time.
MK

Never thought I'd live long enough to see this. The guy's getting senile.

[Michael]

Post by Grunty

I take back the last five names that I called you... ;)
"Thanks Chow", (and Michael;) for allwing me to enjoy RGB this much in a long time.
MK

Never thought I'd live long enough to see this. The guy's getting senile.

The villagers over here have another saying:
At old age even the Devil wears the veil :-)

[Michael]

Post by Tim Chow

Post by Michael
http://www.bgonline.org/forums/webbbs_config.pl?noframes;read=203042#203042

Thanks for the pointer.
I did not read through the whole thread. However, I gather that your main
question is whether the standard "live cube takepoint" calculation is
"mathematically correct."
The term "mathematically correct" is more subtle than I think you recognize.
Let's start by stating something that is perhaps obvious but that needs to
be emphasized in any careful discussion of this type: Backgammon is not
mathematically solved. So if by "mathematically correct" we mean that the
live cube takepoint calculation yields an absolutely mathematically correct
rule for perfect backgammon play, then it is certainly not "mathematically
correct," for the simple reason that nothing is mathematically correct in
this very strong sense.
In fact, if you really want to speak precisely, even the concept of a "take
point" is not "mathematically correct." If we spell out the assumptions that
are usually left unstated, the term "take point" presupposes that if you are
given the following information---
1. the match score and the cube location, and
2. the percentages for wins, gammons, and backgammons for both sides
---then there exists a number, called the "take point," with the property
that if the win rate exceeds the take point then the correct cube action
is to take, and if the win rate falls below the take point then the correct
cube action is to drop. But this assumption is not "mathematically correct."
Whether the correct action is a take or a pass depends on the details of
the position, and not just on the items listed above. Mathematically speaking,
therefore, there is *no such thing as a take point*.
Take point calculations are *heuristics*. By a "heuristic" I mean a rule of
thumb that does not always hold, but that is relatively easy to apply and
that generally provides good advice about how to play well. We all know a
lot of backgammon heuristics---making the 5pt is usually better than making
the bar point, race when ahead in the race, double in a straight race when
you're 10% ahead in the pip count, etc. With the exception of some heuristics
for non-contact positions, none of these are mathematically proven to be 100%
correct. They're just rules of thumb.
Because take point calculations involve more arithmetic than most heuristics,
they may give the impression of being "mathematically correct." But don't
let the arithmetic fool you. Take point calculations are nothing more than
sophisticated rules of thumb, like JOH's rule of 9/36 market losing sequences.
My point is that asking whether a heuristic is "correct" is the wrong question.
The only appropriate question is whether the heuristic works well, and whether
the assumptions underlying the calculation are believable.
O.K., so let's say we understand all this, and we understand that the take
point calculation can't be "correct." We can still ask, who came up with
the calculation and why did they think it would yield a good heuristic?
The way people have approached this sort of question in the past is that
they have developed *models* of backgammon. That is, they invent some kind
of mathematical process that is simpler than real backgammon but that
resembles backgammon in certain ways. They then analyze the simplified
mathematical model, and use that to obtain a heuristic for real backgammon.
The simplest model for backgammon involves a *continuously* varying equity
http://www.bkgm.com/articles/KeelerSpencer/OptimalDoublingInBackgammon/
As Keeler and Spencer discuss, this model is oversimplified in many ways.
So others have tried to come up with more complicated models that are closer
to real backgammon but that can still be analyzed. For a recent attempt of
https://arxiv.org/pdf/1203.5692.pdf
I'm not sure where the particular formula that you asked about on BGO comes
from originally. It might be from Janowski---
http://www.bkgm.com/articles/Janowski/cubeformulae.pdf
---but I haven't read the paper carefully enough to tell whether it describes
exactly the formula you're interested in.
---
Tim Chow

Thanks Tim for your detailed answer. I will have to study your links during the upcoming weekend before discussing this further.
In the meantime, I totally agree with you that in reality there is no fixed value for a Cubeful Doubling or Take point, based on winning percentages alone. You (in plural) still need the position for other factors.(market analysis, recube potential etc). If you notice when referring to my tables I no longer call them DPs, I call them "reference DPs".

I am sure you will also agree with me on the reasons as to why Murat liked the topic so much, that he retracted his latest 5 huge insults ;-)

[m***@compuplus.net]

Post by Michael
I am sure you will also agree with me on the reasons
as to why Murat liked the topic so much, that he
retracted his latest 5 huge insults ;-)

At first I was going to retract my last 10 name callings
but I didn't want to spoil him too much all at once. ;)

And why do you take my "terms of endearment" as "insults"
is beyond me...??

About my reason for liking what Chow wrote, did it sound
like I was trying to hide it? So, why are you feeling so
smart (and expecting Chow to feel the same and agree with
you) about something you don't specify...?

Chow is learning and getting better at understanding it
just like the rest of us...

In the past, I offered Chow to cross over to my side of
the line and ally with me, believing that the two of us
would make a good team to destroy all the folk-science
bullshit of the "small circle of incestuous gamblers of
backgammon" (copyright Patti Beadles).

If he doesn't, it's okay. I've been defending my views
here for the past 21 years and I don't need any help to
continue until all of you thick skulled idiots can catch
up.

It's just a matter of time...

MK

[Michael]

Post by m***@compuplus.net
"small circle of incestuous gamblers of
backgammon" (copyright Patti Beadles).
MK

Well the reason you liked it so much was clear: Because of the possibility that bot's cube decisions might deviate from reality.
However my aim is totally different from yours. My aim is to make them better, your aim is to prove they are "bullshit", hence everybody should get rid of them.

I agreed with some of your opinions in the past as e.g. that the cube is a game within a game turning a relatively easy game to an extremely difficult one. I also agreed with you that using BG for gambling is bad in many ways, including turning a lot of people away from it.

Could you give me the link where Patti said what you quoted?

[Tim Chow]

Post by Michael
Could you give me the link where Patti said what you quoted?

I think Patti did use the term "incestuous" but I don't think she used exactly
the phrase Murat put in quotation marks. This was a long time ago (pre-2000?)
so it may not be archived in Google Groups.

---
Tim Chow

[Michael]

Post by Tim Chow

Post by Michael
Could you give me the link where Patti said what you quoted?

I think Patti did use the term "incestuous" but I don't think she used exactly
the phrase Murat put in quotation marks. This was a long time ago (pre-2000?)
so it may not be archived in Google Groups.
---
Tim Chow

So it was in this group huh?
I did a little search and here's what actually Patti said:
https://groups.google.com/d/msg/rec.games.backgammon/1yNQqn_G-5M/W9D1xkxsmGEJ

[m***@compuplus.net]

Post by Michael
So it was in this group huh?
https://groups.google.com/d/msg/rec.games.backgammon/1yNQqn_G-5M/W9D1xkxsmGEJ

I liked the expression so much that I probebly used it 100
time since then and every now and then I tried to give her
"credit" :) by mentioning her even if I could no longer
remember her exact words. I never tried to or ever have
unintentionally distorted what she meant too badly.

MK

[Michael]

OK I have studied the 3 links. I am (rather was) just an Engineer not a Mathematician.

From the first 2 I think I understood how those models are used, however I cannot express an opinion in how close they might be to reality. One needs a Phd in Maths to follow such studies in detail.
Certainly I don't qualify.
The 3rd one by Janowski was much easier for me to follow, at least upto page 6 where from he starts studying the effect of Jacoby, beavers etc that don't interest me anyway.

The model is for money games. Still the conditions deviate from reality.
In reality the equity does not change continuously neither has a smooth sinusoidal type of shape.
The swings don't occur smoothly, but rather suddenly.The possibility to double to 4 is not the same as the possibility to double to 8.
The redoubles are seldomly optimal. A backgammon game is too short for redoubles to occur ad infinitum.
Although he states that "The infinite model is a good approximation to any of the finite models" you don't need to be a rocket scientist to guess how far off this approximation could be.

It looks to me Janowski concentrated in defining the boundaries within which the cubeful TP should exist.
As he states using the Dead cube model and the Live cube model "When considered together, they provide an envelope in which correct cube action decisions are to be found."

Imo we don't need this envelope. This envelope assumes correct calculation of both the Dead TP and The Fully Live TP and further calculations between the 2 to find the real (cubeful) TP. The last 2 are done by questionable heuristic types of calculations. The Fully Live TP is just an invention to reflect the value of owning the cube.

However the only thing that is indisputable is the Dead TP and the fact that cube ownership has a value.

I don't really qualify neither as Mathematician or statistician or programmer or top Bg player to form a team that would revise all these things.
The following is just a suggestion for such a group of people who might in the future undertake the difficult task of doing such a revision.

Imo we should scrap all these concepts that are based on the fully Live TP and equity.
If we are to exclude getting any other information from the actual position other than the winning percentages, then the Real TP imo should depend on only a few factors:

a)The score and the Dead TP Ats given the winning/losing percentages
b)The number of roll at which the cube decission occured.
That would determine how much game is left to provide for the possibility of a redouble/s
c)On whether the game is with or without contact.
d)Some other factors that I cannot think of.

All these can be plugged in to bots taking data from statistics. If we don't want to rely so heavily on precalculated statistics, "b" which is
the most important one, can be found by setting the bot to run say 1000 cubeless trials. (it will only need a second).
The real TP would then just be a % factor away from the already known and indisputable Dead TP.
Humans could also do an approximate estimate using the same criteria otb more easily too.

Just a suggestion as I said...

NB. I still have 2 graphs from the times I was trying to understand the effect of luck.
They are for 2 separate games, both indicative of how the equity changes and how redoubles (if any) might occur.
https://drive.google.com/open?id=0B7L6em2ChhMOS0hJZ2NOMjd5VXM

[m***@compuplus.net]

Post by Michael
However the only thing that is indisputable is the Dead TP

Am I understanding that this means when neither side can
lose more by doubling/taking?

Post by Michael
and the fact that cube ownership has a value.

Do you mean only a value in having control of the future
cube actions or also a value in being able to exploit a
different strategy in checker decisions?

In any case, there arises problems of calculating these
values.

In the past, I had proposed that if cube ownership value
is known, it can be "traded" to eliminate stagnant cube
ownerships/actions. I had punned by using "in-cube-ating"
to describe the action of of sitting on a bad cube,
especially in match play where there are no beavers, etc.

So, the player who doesn't own the cube would have access
to the cube, with or without taking ownership of the cube,
by paying the cube owner the full or partial value of the
cube ownership.

I was, of course, mocking at the "cube skill" thing but
at the same time I was proposing a logical way of adding
even more "cube skill" to gambling backgammon... :)

Post by Michael
a)The score and the Dead TP Ats given the winning/losing
percentages
b)The number of roll at which the cube decission occured.
That would determine how much game is left to provide for
the possibility of a redouble/s
c)On whether the game is with or without contact.
d)Some other factors that I cannot think of.

These are very similar to what I had written in the past,
explaining how I was making cube decisions, with a few
differences:

a) I only consider the score in combination of my own gut
feeling on whether I am more or less likely to win.

b) "How much game is left" is the other very important in
my cube decisions but mostly based on "how much checker
play left" and hardly any on "redouble possibilities left",
which I worry about less because I believe that checker
skill is more important in the long run, (also meaning with
enough checker play left in the game).

c) This is the hardest for me since winning chances can't
be calculated (until towards the very end of the game) and
there is no use left for checker skill/strategy. If it's
not very obvious to me, I tend to chance it and take.

d) An example for me would be playing against an opponent
based on equal or prorated odds (according to ER or rating
difference).

Post by Michael
All these can be plugged in to bots taking data from
statistics. If we don't want to rely so heavily on
precalculated statistics, "b" which is the most important
one, can be found by setting the bot to run say 1000
cubeless trials.

I don't quite understand this part. It sounds like you may
be contradicting what you said above. Whether precalculated
or not, your statistics/trials need to be not only cubeless
but also based on random play. Otherwise you end up injecting
back bias of an undeterminable magnitude.

Referring to bots as AI by the ilks of Marc Brockmann is
way too pretentious, when they are not even neural-nets
in the real meaning. Unbiased neural-net learning (from
cubeless/random moves) for bots stopped after TD-Gammon.

Since then bots have become elaborate folk science...

MK

[Michael]

Post by m***@compuplus.net

Post by Michael
However the only thing that is indisputable is the Dead TP

Am I understanding that this means when neither side can
lose more by doubling/taking?

That's one of the 2 cases. e.g at score 4-5 in a match of 7 if the leader owns the cube then he owns a "dead/unusable" thing. The other case is when we calculate it with the assumption that the game is cubeless. (refer to my thread at BGO for how it's calculated. This calculation is 100% correct however because of the assumption it doesn't represent reality. Therefore the concept of the Fully live TP (with all it's implications) is introduced , in order to calculate the real thing (called cubeful) from the Cubeless or Dead TP.

Post by m***@compuplus.net

Post by Michael
and the fact that cube ownership has a value.

Do you mean only a value in having control of the future
cube actions or also a value in being able to exploit a
different strategy in checker decisions?

The value in having control of future cube actions.This value arises from eventually "stealing" some winning percentages. Suppose someone doubles at 65% game winning chances and it's a D/T. The game might swing and the guy who received the cube start increasing his percentages from 35 to 50,60,70, 80%. When he reaches 80% he Re-Doubles and gets a Pass. At that point he "trashed/stole" the other guy's 20% chances of winning. That's the value of owning the cube.

Post by m***@compuplus.net
In any case, there arises problems of calculating these
values.

The Dead TP has no problems getting calculated. As for the rest yes I agree there are problems in calculating them.

Post by m***@compuplus.net
In the past, I had proposed that if cube ownership value
is known, it can be "traded" to eliminate stagnant cube
ownerships/actions. I had punned by using "in-cube-ating"
to describe the action of of sitting on a bad cube,
especially in match play where there are no beavers, etc.
So, the player who doesn't own the cube would have access
to the cube, with or without taking ownership of the cube,
by paying the cube owner the full or partial value of the
cube ownership.

OK but isn't it enough that we already have such a difficult thing to deal with, should we add more difficult things to it?

Post by m***@compuplus.net
I was, of course, mocking at the "cube skill" thing but
at the same time I was proposing a logical way of adding
even more "cube skill" to gambling backgammon... :)

All you have to do to realize how difficult it is to handle the cube is split the Top players' PRs in checker play error rate and cube play error rate.
The cube play error rate is usually 3-5 times more. Mine is 3 times more :-)

Post by m***@compuplus.net

Post by Michael
a)The score and the Dead TP Ats given the winning/losing
percentages
b)The number of roll at which the cube decission occured.
That would determine how much game is left to provide for
the possibility of a redouble/s
c)On whether the game is with or without contact.
d)Some other factors that I cannot think of.

These are very similar to what I had written in the past,
explaining how I was making cube decisions, with a few
a) I only consider the score in combination of my own gut
feeling on whether I am more or less likely to win.
b) "How much game is left" is the other very important in
my cube decisions but mostly based on "how much checker
play left" and hardly any on "redouble possibilities left",
which I worry about less because I believe that checker
skill is more important in the long run, (also meaning with
enough checker play left in the game).
c) This is the hardest for me since winning chances can't
be calculated (until towards the very end of the game) and
there is no use left for checker skill/strategy. If it's
not very obvious to me, I tend to chance it and take.
d) An example for me would be playing against an opponent
based on equal or prorated odds (according to ER or rating
difference).

It's because you insult everybody behaving like a teenager who tries to make his classmate girls laugh. It always ends with a punch in the nose, it doesn't leave any room for anyone to consider your points seriously.

Post by m***@compuplus.net

Post by Michael
All these can be plugged in to bots taking data from
statistics. If we don't want to rely so heavily on
precalculated statistics, "b" which is the most important
one, can be found by setting the bot to run say 1000
cubeless trials.

I don't quite understand this part. It sounds like you may
be contradicting what you said above. Whether precalculated
or not, your statistics/trials need to be not only cubeless
but also based on random play. Otherwise you end up injecting
back bias of an undeterminable magnitude.

I think we should abandon the idea of precalculated stats. And yes the 1000 stats will be cubeless but the bot will take note of the number of times there should have been a cube action within those cubeless results. For an example see the 2nd graph I posted. It is a cubeless game but clearly there should have been 2 cube actions within it.

[m***@compuplus.net]

Post by Michael

Post by m***@compuplus.net

Post by Michael
However the only thing that is indisputable is the Dead TP

Am I understanding that this means when neither side can
lose more by doubling/taking?

That's one of the 2 cases. e.g at score 4-5 in a match of
7 if the leader owns the cube then he owns a "dead/unusable"
thing.

This, I understand.

Post by Michael
The other case is when we calculate it with the assumption
that the game is cubeless.

This, i don't understand.

At this point, I wonder if I should even further discuss this
for fear that I may end up getting infected with the "cube
skill virus" (i.e. "understanding it"), after years of making
a deliberate effort to not do so...

Post by Michael
This calculation is 100% correct however because of the
assumption it doesn't represent reality. Therefore the
concept of the Fully live TP (with all it's implications)
is introduced , in order to calculate the real thing
(called cubeful) from the Cubeless or Dead TP.

I have no idea about what you are talking about and feel
safe. :) But I get a feeling enough to ask "so, then, why
do you do it"?

Post by Michael

Post by m***@compuplus.net

Post by Michael
and the fact that cube ownership has a value.

Do you mean only a value in having control of the future
cube actions or also a value in being able to exploit a
different strategy in checker decisions?

The value in having control of future cube actions.

Okay, then my follow up question is why nobody else other
than me talks about the "value ow cube ownership" in terms
of "checker play dollars"...?

If a human player is really of a world-class caliber, he
surely would adjust his checker player decisions based on
whether he own the cube or not, no...?

Or is this too complicated for you all (not you personally)
to understand...??

Post by Michael
This value arises from eventually "stealing" some winning
percentages.... That's the value of owning the cube.
OK but isn't it enough that we already have such a
difficult thing to deal with, should we add more
difficult things to it?

We can't add anything more to "the game of backgammon" even
if we wanted to.

But if there is such a thing as "cube skill", why not allow
more skillful trading of "cube value"?

The cube owner thinks his cube ownership is worth $10 and
he is sitting on his investment.

His opponent thinks it's worth $20 and wants to acquire the
cube ownership by paying him $10, in believing that he can
make a $10 profit from it?

Why not allow this more dynamic "gambling market activity"
instead of letting the cube owner get away with sitting on
his bad investment to avoid further losses?

We are playing "backgammon for blood!" aren't we...?

Post by Michael
All you have to do to realize how difficult it is to
handle the cube is split the Top players' PRs in checker
play error rate and cube play error rate. The cube play
error rate is usually 3-5 times more. Mine is 3 times more :-)

Your observation, info is correct but your conclusion that
it's due to the difficulty of cube handling is wrong. It's
rather proof that the so called cube-skill is way too much
of an inflated financial skill, and the cube value is like
unstable currency or stock...

Post by Michael
It's because you insult everybody behaving like a
teenager who tries to make his classmate girls laugh.
It always ends with a punch in the nose, it doesn't
leave any room for anyone to consider your points seriously.

Come on Michael, you should have understood me better by
now. I have been posting in RGB since 1996!

Nobody here, including you, will ever break ranks with the
flock to consider my points seriously, god forbit admit that
I am ever right.

If they (you) did, they couldn't look at each other in the
face anymore after that. They would be excommunicated and
become another me...

Post by Michael
I think we should abandon the idea of precalculated stats.
And yes the 1000 stats will be cubeless but the bot will
take note of the number of times there should have been a
cube action within those cubeless results.

This is oxymoron to me... However, if the puppies will have
fun chasing their tails, let them enjoy life... :))

MK

[Michael]

Post by m***@compuplus.net

Post by Michael
The other case is when we calculate it with the assumption
that the game is cubeless.

This, i don't understand.

OK I will just correct only this statement to avoid a never ending discussion:
"The other case is when we calculate it with the assumption that the game AFTERWARDS is cubeless." In other words that we Double but the receiver gets a Dead Cube. This calculation is 100% correct however because of the assumption it doesn't represent reality.... etc"\
When I say "we" I actually mean the bots. I doubt there's any human who can memorize the numbers from the MET to do such calculations, let aside estimate wins/losses/gammons precisely.


I noted your idea to buy the cube at a negotiated value.I think it's applicable. Probably those who like gambling so much might raise you a statue for the idea. Now imagine that !! :-)

[Tim Chow]

Post by m***@compuplus.net
But if there is such a thing as "cube skill", why not allow
more skillful trading of "cube value"?

I think the main reason people don't do this (in games played for money) is
that it slows down the action. Gamblers don't want the game paused while
two people squabble over the price, perhaps failing to reach a deal.

---
Tim Chow

[m***@compuplus.net]

Post by Tim Chow

Post by m***@compuplus.net
But if there is such a thing as "cube skill", why not allow
more skillful trading of "cube value"?

I think the main reason people don't do this (in games played
for money) is that it slows down the action. Gamblers don't
want the game paused while two people squabble over the price,
perhaps failing to reach a deal.

No need for reaching a deal each time. Like the approximate
"doubling window" and predetermined simple, fixed doubling
values (i.e. 2, 4, 8), the value of cube ownership can also
be predetermined and simplified.

For example, current cube owner is given half of the current
cube value in points and also gets to keep the cube, or he
is given the full cube value in points but the cube ownership
passes to the opponent.

Let's say in a 9-point match Player-A is 3-2 ahead and owns
the cube at 2. Player-B can raise the cube to 4 and either
pay Player-A 1 point and let him keep the cube, changing the
score to 4-2, or pay him 2 points and take ownership of the
cube, changing the score to 5-2.

Whatever the definition of "cube skill" (that I personally
look down upon), this would require more of it but I don't
think you chimps have brains evolved enough to take the step... :(

MK

[m***@compuplus.net]

Post by Michael

Why aren't you and michael discuss "math axioms" about
the "cube" here in RGB?

a) the main one: this forum does not support tables.

Yes, unfortunately this a limitation of USENET by nature.

Post by Michael
I could of course upload the tables as images somewhere.

It may be a good idea to find a place for this purpose,
which is likely to be around for a long while.

Post by Michael
But that wouldn't help as one day they would be deleted.

Stick's forum can be deleted any day also and in fact, it
is more likely that Google's USENET archives will outlast
forums operated by individuals.

Post by Michael
b)that forum is known to have a lot of mathematicians of
about equal caliber to Tim's. Unfortunately no one so far
expressed any opinion to my question.

Their interests seem to be more on turneys, rules, rankings,
equipmet, etc. (none of mine) and interestingly, while they
complain about the lack of position discussions there, RGB
is almost all but position discussions... :(

Post by Michael
c)Tim is already a member there and could participate.

He used to particiapate heavily and sometimes cross post
to here but apparently he doesn't anymore (because of a
mystery reason;).

Post by Michael
d)Other than Tim no one else in this forum would care.
http://www.bgonline.org/forums/webbbs_config.pl?noframes;read=203042#203042

Nice table :) I would be interested in the subject but
not in such elaborate detail which find ridiculously
pretentious.

If you remember from looking at the games I had posted
on Youtube, I have a very different way of making cube
decisions, (that can perhaps be likened to playing music
"by ear"). Only towards the end of the game, I may come
closer to making "correct" cube decisions. Early game
cube decisions based on calculated take/drop poinsts is
fantasy...

MK