Page 2 of 3 FirstFirst 123 LastLast
Results 14 to 26 of 37

Thread: Fuzzy Math: Cacarulo

  1. #14
    Cacarulo
    Guest

    Cacarulo: See my "Moreover" post please *NM*


  2. #15
    Cacarulo
    Guest

    Cacarulo: Re: Insurance

    > If you were dealt those two hands at the
    > same time, you would not ever stand on the
    > first one but then hit the second one --
    > each will play out exactly the same. Why
    > should insurance be any different?

    This is the problem. You play each hand independently of the other. You could hit one and stand on the other based on the CD indices.

    > Let's look at what insurance really is: You
    > get to place a side bet which pays 2:1 if
    > the dealer has a ten in the hole. If he
    > does, then the money you lost on your hand
    > is then paid back to you for winning the
    > insurance betn You lose your initial bet if
    > the dealer has a BJ regardless of insurance
    > -- insurance is simply a side bet whose
    > winning payout happens to be the same as the
    > amount of your losing bet. If the dealer has
    > a BJ less than 1 time in 3 that you insure,
    > then you are losing money. A counter
    > attempts to use his knowledge of the
    > remaining cards to determine if this is a
    > positive-EV bet to take. You are advocating
    > that it will be a positive-EV bet on one of
    > your hands, but negative-EV on another hand
    > -- at the same time!

    I'm not advocating that. I say that insuring one hand but not the other (based on the CD indices) will make more money (EV) in the long run. The other alternatives are:

    2) Insure both hands
    3) Don't Insure any hand

    In the long run option 1) will come ahead.

    OTOH, when the index shows positive-EV it has a variance associated. The positive-EV is an average in which sometimes we have positive-EV and sometimes we have negative-EV depending on the composition of the pack.

    Sincerely,
    Cacarulo

  3. #16
    Cacarulo
    Guest

    Cacarulo: Your puzzle

    > And finally, I leave you with a puzzle: You
    > are playing doubledeck and put out two bets
    > with a RC of 7. You get dealt A/A on one
    > hand, and T/T on the other. You are asked if
    > you want insurance. There are exactly 52
    > cards remaining to be dealt/seen. The count
    > has now fallen to 2. Your indices for 2D are
    > .46 for A/A and 3.16 for T/T. What would you
    > do to reconcile this large difference --
    > Does the insurance side bet yield
    > positive-EV?

    Let's forget about CD indices for a moment and get
    back to the traditional use of a single generic index (+2.38). Since the "updated" TC is now +2 we don't insure any of the hands. Correct?

    Now, say we know the indices for each particular hand (+0.46 for AA and +3.16 for TT). According to what I advocate for we should insure AA but not TT.
    Your position is that we should either insure both hands or don't insure any. Correct?
    Well, if your answer is YES then what do you propose to do based on the info that we have?

    1) Should we go by the generic index and DON'T INSURE any of the hands
    2) Should we go by the TT index and DON'T INSURE any of the hands
    3) Should we go by the AA index and INSURE both hands?

    Which decision is MORE correct? We don't have the index for AATT v A so we have to decide only on what we have (a pair of indices or a generic index).

    The index that we don't have is +1.11. So, if we knew this index apriori we would insure BOTH hands being that the correct decision.
    Aren't we better then insuring at least one hand rather than any?

    Sincerely,
    Cacarulo

  4. #17
    Fuzzy Math
    Guest

    Fuzzy Math: Re: Your puzzle

    I believe I understand your point now -- insuring only one hand in certain situations (that is, always following the C-D indices) would be better in the long run than simply going by the generic index in those situations.

    My point has simply been that this play is not optimal -- accurately considering both of your hands (or for that matter, any/all other hands since the last shuffle) would yield better long-term results than just using the single-hand CD indices.

    You could say that the generic index has an EV of X, your single-hand CD indices have an EV of Y, and considering the composition of more than one hand has an EV of Z. So X < Y < Z. You are discussing Y. I am discussing Z. Hopefully we each understand each other now.

  5. #18
    Cacarulo
    Guest

    Cacarulo: Re: Your puzzle

    > I believe I understand your point now --
    > insuring only one hand in certain situations
    > (that is, always following the C-D indices)
    > would be better in the long run than simply
    > going by the generic index in those
    > situations.

    Going by the CD indices is always better than going by the generic.

    > My point has simply been that this play is
    > not optimal -- accurately considering both
    > of your hands (or for that matter, any/all
    > other hands since the last shuffle) would
    > yield better long-term results than just
    > using the single-hand CD indices.

    And I agree that it's not optimal but as I said we don't have those indices (Z) and we don't want them either. We have Y that is better than X.

    > You could say that the generic index has an
    > EV of X, your single-hand CD indices have an
    > EV of Y, and considering the composition of
    > more than one hand has an EV of Z. So X < Y < Z. > You are discussing Y. I am discussing Z.
    > Hopefully we each understand each other now.

    We do now

    Sincerely,
    Cacarulo

  6. #19
    Cacarulo
    Guest

    Cacarulo: Re: A few sd data

    > For single deck play I?ve got these figures
    > off the top:
    > Generic index = 1.4
    > T, 5 vs A Insurance if TC=>2.31254
    > 9, 6 vs A Insurance if TC=>0.82297

    > The problem is that when you removed the 5
    > cards at the same time you get:
    > Insurance if TC=>1.73515 so the dreams
    > for an ?early insurance? for the 9, 6 hand
    > seem to evaporate.
    > Given that both hands belong to the same
    > player the independence of effects is
    > dubious at least.

    My figures are:

    Generic = 1.416667
    T5vA = 2.357333
    96vA = 0.840136
    T596vA = 1.770213

    which are very close to yours.
    Now, say the current TC is +1.5. If we go by the generic, we insure both hands (note that the correct choice would be to don't insure any of the hands).
    Going by the CD indices we would insure only one hand (96). So we lose less than with the generic.

    If the count were +2 the generic would be doing a good job but overall the CD indices are better.

    Sincerely,
    Cac

  7. #20
    Fuzzy Math
    Guest

    Fuzzy Math: Re: A few sd data

    Since insurance calculations are purely linear based simply on the ratio of tens, it occurs to me that each card known to the player can have a constant value used in calculating the proper index.

    I'll use the 2 deck data:

    A,A vs A = +0.46
    A,Z vs A = +1.02
    A,L vs A = +1.53
    T,T vs A = +3.16
    T,A vs A = +1.76
    T,Z vs A = +2.36
    T,L vs A = +2.86
    Z,Z vs A = +1.58
    Z,L vs A = +2.09
    L,L vs A = +2.60

    Each ace that the player sees lowers the index by about 1.0 (which I will refer to as -1.0). Z is -.4, T is +.4, L is about +.1

    We can then look at the chart this way (with all numbers rounded for simplicity -- we'd like to be able to use this at the table):

    Generic Index = +2.38 (+2.4)

    A,A vs A = +0.46 (2.4 - 1.0 - 1.0 = +0.4)
    A,Z vs A = +1.02 (2.4 - 1.0 - 0.4 = +1.0)
    A,L vs A = +1.53 (2.4 - 1.0 + 0.1 = +1.5)
    T,T vs A = +3.16 (2.4 + 0.4 + 0.4 = +3.2)
    T,A vs A = +1.76 (2.4 + 0.4 - 1.0 = +1.8)
    T,Z vs A = +2.36 (2.4 + 0.4 - 0.4 = +1.4)
    T,L vs A = +2.86 (2.4 + 0.4 + 0.1 = +2.9)
    Z,Z vs A = +1.58 (2.4 - 0.4 - 0.4 = +1.6)
    Z,L vs A = +2.09 (2.4 - 0.4 + 0.1 = +2.1)
    L,L vs A = +2.60 (2.4 + 0.1 + 0.1 = +2.6)

    One immediate application of this is that we can learn just those four numbers as opposed to all 10 indices.

    Secondly, I would infer that this formula would continue to hold true with more than two cards. If the player had A/A and T/T, he would calculate an index of (2.4 - 1.0 - 1.0 + 0.4 + 0.4 = 1.2). The index that you had calculated for this scenario was 1.1, so it seems that this method has worked fairly well. The player in my example would then correctly insure both hands with the count being 2.

    The main problem I see with continuing this past 2 cards is that rounding could skew the results (this insurance-count is a balanced count though, so the rounding errors should mostly cancel out). If there are any other issues that I didn't think of, please correct me.

  8. #21
    Cacarulo
    Guest

    Cacarulo: Very good idea!

    > Since insurance calculations are purely
    > linear based simply on the ratio of tens, it
    > occurs to me that each card known to the
    > player can have a constant value used in
    > calculating the proper index.

    > I'll use the 2 deck data:

    > A,A vs A = +0.46
    > A,Z vs A = +1.02
    > A,L vs A = +1.53
    > T,T vs A = +3.16
    > T,A vs A = +1.76
    > T,Z vs A = +2.36
    > T,L vs A = +2.86
    > Z,Z vs A = +1.58
    > Z,L vs A = +2.09
    > L,L vs A = +2.60

    > Each ace that the player sees lowers the
    > index by about 1.0 (which I will refer to as
    > -1.0). Z is -.4, T is +.4, L is about +.1

    > We can then look at the chart this way (with
    > all numbers rounded for simplicity -- we'd
    > like to be able to use this at the table):

    > Generic Index = +2.38 (+2.4)

    > A,A vs A = +0.46 (2.4 - 1.0 - 1.0 = +0.4)
    > A,Z vs A = +1.02 (2.4 - 1.0 - 0.4 = +1.0)
    > A,L vs A = +1.53 (2.4 - 1.0 + 0.1 = +1.5)
    > T,T vs A = +3.16 (2.4 + 0.4 + 0.4 = +3.2)
    > T,A vs A = +1.76 (2.4 + 0.4 - 1.0 = +1.8)
    > T,Z vs A = +2.36 (2.4 + 0.4 - 0.4 = +1.4)
    > T,L vs A = +2.86 (2.4 + 0.4 + 0.1 = +2.9)
    > Z,Z vs A = +1.58 (2.4 - 0.4 - 0.4 = +1.6)
    > Z,L vs A = +2.09 (2.4 - 0.4 + 0.1 = +2.1)
    > L,L vs A = +2.60 (2.4 + 0.1 + 0.1 = +2.6)

    > One immediate application of this is that we
    > can learn just those four numbers as opposed
    > to all 10 indices.

    > Secondly, I would infer that this formula
    > would continue to hold true with more than
    > two cards. If the player had A/A and T/T, he
    > would calculate an index of (2.4 - 1.0 - 1.0
    > + 0.4 + 0.4 = 1.2). The index that you had
    > calculated for this scenario was 1.1, so it
    > seems that this method has worked fairly
    > well. The player in my example would then
    > correctly insure both hands with the count
    > being 2.

    > The main problem I see with continuing this
    > past 2 cards is that rounding could skew the
    > results (this insurance-count is a balanced
    > count though, so the rounding errors should
    > mostly cancel out). If there are any other
    > issues that I didn't think of, please
    > correct me.

    I will give it some more thought but I think you've elegantly solved the problem of playing with more hands.
    I've to run now.

    Sincerely,
    Cacarulo

    PS: Please Viktor, archive FM's post!

  9. #22
    Zenfighter
    Guest

    Zenfighter: Re: I figured that already

    Btw, MathProf in his last Gambling Conference paper, give us 2.38 and 0.82 as the cd indexes for both stiff hands. As you can see we have here a slightly triple discrepancy with the T-5 stiff. :-)

    Sincerely

    Z

  10. #23
    Cacarulo
    Guest

    Cacarulo: Trying with more accurate figures

    > I'll use the 2 deck data:

    > A,A vs A = +0.46
    > A,Z vs A = +1.02
    > A,L vs A = +1.53
    > T,T vs A = +3.16
    > T,A vs A = +1.76
    > T,Z vs A = +2.36
    > T,L vs A = +2.86
    > Z,Z vs A = +1.58
    > Z,L vs A = +2.09
    > L,L vs A = +2.60

    > Each ace that the player sees lowers the
    > index by about 1.0 (which I will refer to as
    > -1.0). Z is -.4, T is +.4, L is about +.1

    > We can then look at the chart this way (with
    > all numbers rounded for simplicity -- we'd
    > like to be able to use this at the table):

    > Generic Index = +2.38 (+2.4)

    > A,A vs A = +0.46 (2.4 - 1.0 - 1.0 = +0.4)
    > A,Z vs A = +1.02 (2.4 - 1.0 - 0.4 = +1.0)
    > A,L vs A = +1.53 (2.4 - 1.0 + 0.1 = +1.5)
    > T,T vs A = +3.16 (2.4 + 0.4 + 0.4 = +3.2)
    > T,A vs A = +1.76 (2.4 + 0.4 - 1.0 = +1.8)
    > T,Z vs A = +2.36 (2.4 + 0.4 - 0.4 = +1.4)
    > T,L vs A = +2.86 (2.4 + 0.4 + 0.1 = +2.9)
    > Z,Z vs A = +1.58 (2.4 - 0.4 - 0.4 = +1.6)
    > Z,L vs A = +2.09 (2.4 - 0.4 + 0.1 = +2.1)
    > L,L vs A = +2.60 (2.4 + 0.1 + 0.1 = +2.6)

    > One immediate application of this is that we
    > can learn just those four numbers as opposed
    > to all 10 indices.

    Totally agree.

    > Secondly, I would infer that this formula
    > would continue to hold true with more than
    > two cards. If the player had A/A and T/T, he
    > would calculate an index of (2.4 - 1.0 - 1.0
    > + 0.4 + 0.4 = 1.2). The index that you had
    > calculated for this scenario was 1.1, so it
    > seems that this method has worked fairly
    > well. The player in my example would then
    > correctly insure both hands with the count
    > being 2.

    > The main problem I see with continuing this
    > past 2 cards is that rounding could skew the
    > results (this insurance-count is a balanced
    > count though, so the rounding errors should
    > mostly cancel out). If there are any other
    > issues that I didn't think of, please
    > correct me.

    Could be. That's why I thought that we could use more accurate figures:

    Generic = +2.38

    A = -0.96
    T = +0.38
    L = +0.11
    Z = +0.40

    A,A vs A = +0.46 (2.38 - 0.96 - 0.96 = +0.46)
    A,Z vs A = +1.02 (2.38 - 0.96 - 0.40 = +1.02)
    A,L vs A = +1.53 (2.38 - 0.96 + 0.11 = +1.53)
    T,T vs A = +3.16 (2.38 + 0.38 + 0.38 = +3.14)
    T,A vs A = +1.76 (2.38 + 0.38 - 0.96 = +1.80)
    T,Z vs A = +2.36 (2.38 + 0.38 - 0.40 = +2.36)
    T,L vs A = +2.86 (2.38 + 0.38 + 0.11 = +2.87)
    Z,Z vs A = +1.58 (2.38 - 0.40 - 0.40 = +1.58)
    Z,L vs A = +2.09 (2.38 - 0.40 + 0.11 = +2.09)
    L,L vs A = +2.60 (2.38 + 0.11 + 0.11 = +2.60)

    A,A,T,T vs A = +1.11 (2.38 - 0.96 - 0.96 + 0.38 + 0.38 = +1.22)

    It seems that we don't get a better performance with the more accurate figures.

    Will try later with a few more examples.

    Sincerely,
    Cacarulo

  11. #24
    Cacarulo
    Guest

    Cacarulo: Re: I figured that already

    > Btw, MathProf in his last Gambling
    > Conference paper, give us 2.38 and 0.82 as
    > the cd indexes for both stiff hands. As you
    > can see we have here a slightly triple
    > discrepancy with the T-5 stiff. :-)

    Try with the following EORs instead of the ones that appear in TOB.

    A to 9 =  1.809954751131222 
    Tens = -4.072398190045249
    Mean = -7.692307692307693
    SS = 95.821133883417630


    Sincerely,
    Cac

  12. #25
    Cacarulo
    Guest

    Cacarulo: More examples

    > Generic = +2.38 (+2.4)

    > A = -0.96 (-1.0)
    > T = +0.38 (+0.4)
    > L = +0.11 (+0.1)
    > Z = -0.40 (-0.4)

    The approximation is within parenthesis.

    AATTvA = +1.11 (+1.22)
    AAAAvA = -1.46 (-1.46)
    ZZZZvA = +0.76 (+0.78)
    LLLLvA = +2.84 (+2.82)
    TTTTvA = +4.05 (+3.90)
    TLZLvA = +2.58 (+2.58)
    TTLLvA = +3.37 (+3.36)

    TLZLTLvA = +3.09 (+3.07)
    TLZLTLAAvA = +1.05 (+1.15)

    Pretty good eh?

    Of course, Fuzzy Math rounded figures will do the job as well and besides his numbers are easier to remember.

    Sincerely,
    Cac

  13. #26
    Cacarulo
    Guest

    Cacarulo: Typo

    > Z = +0.40

    Should be

    Z = -0.40

    Sincerely,
    Cac

Page 2 of 3 FirstFirst 123 LastLast

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  

About Blackjack: The Forum

BJTF is an advantage player site based on the principles of comity. That is, civil and considerate behavior for the mutual benefit of all involved. The goal of advantage play is the legal extraction of funds from gaming establishments by gaining a mathematic advantage and developing the skills required to use that advantage. To maximize our success, it is important to understand that we are all on the same side. Personal conflicts simply get in the way of our goals.