Page 2 of 3 FirstFirst 123 LastLast
Results 14 to 26 of 35

Thread: AsZehn: KO vs Red 7

  1. #14
    John Auston
    Guest

    John Auston: Re: Jumping in here

    > An interesting point, John, but the fact
    > that K-O is weaker away from the pivot is
    > K-O's problem. It isn't UBZII's problem,
    > which should be happy no matter what index
    > it is at.

    Does this sound reasonable?:

    Given: unbalanced counts are less accurate at departures and amount-to-bet, the further they are
    from pivot.

    Consider the following hi-lo trues:

    <-5 UBZII always 2 "better" than KO
    -5: -7 below UBZII pivot, -9 below KO
    -4: -6 below UBZII pivot, -8 below KO
    -3: -5 below UBZII pivot, -7 below KO
    -2: -4 below UBZII pivot, -6 below KO
    -1: -3 below UBZII pivot, -5 below KO
    0: -2 below UBZII pivot, -4 below KO
    +1: -1 below UBZII pivot, -3 below KO
    +2: right at UBZII pivot, -2 below KO
    +3: 1 above UBZII pivot, -1 below KO
    +4: 2 above UBZII pivot, right at KO
    +5: 3 above UBZII pivot, 1 above KO
    >5: UBZII always 2 "worse" than KO

    Play-all, it all evens out, and UBZII level 2 has the edge.

    But restrict to Wong only:

    +1: -1 below UBZII pivot, -3 below KO
    +2: at UBZII pivot, -2 below KO
    +3: 1 above UBZII pivot, -1 below KO
    +4: 2 above UBZII pivot, at KO
    +5: 3 above UBZII pivot, 1 above KO
    >5: UBZII always 2 "worse" than KO

    UBZII only has the "accuracy advantage" at +1 and +2,
    KO has the rest, which is probably just enough to overcome the
    level 1 versus level 2.

    Another observation. KO's disadvantage at negative counts would hurt it more than it does were it not for the fact that only around 1 unit is bet there anyway.

    The reverse happens to UBZII at high counts, since we are at max anyway.

    So, when we Wong, the key counts are +2, +3, +4, and +5, and KO has the accuracy edge there.

    Or so it seems to me, just trying to "thought experiment" it.

    Also, if what I am thinking is right, then it should hold that the larger the Wonging spread, the more KO should inch ahead, and the reverse.

    But I could be wrong. Just trying to think what could explain those results.

    John


  2. #15
    Don Schlesinger
    Guest

    Don Schlesinger: Re: Jumping in here

    > Does this sound reasonable?:

    > Given: unbalanced counts are less accurate
    > at departures and amount-to-bet, the further
    > they are
    > from pivot.

    > Consider the following hi-lo trues:

    > -5: -7 below UBZII pivot, -9 below KO
    > -4: -6 below UBZII pivot, -8 below KO
    > -3: -5 below UBZII pivot, -7 below KO
    > -2: -4 below UBZII pivot, -6 below KO
    > -1: -3 below UBZII pivot, -5 below KO
    > 0: -2 below UBZII pivot, -4 below KO
    > +1: -1 below UBZII pivot, -3 below KO
    > +2: right at UBZII pivot, -2 below KO
    > +3: 1 above UBZII pivot, -1 below KO
    > +4: 2 above UBZII pivot, right at KO
    > +5: 3 above UBZII pivot, 1 above KO

    > Play-all, it all evens out, and UBZII level
    > 2 has the edge.

    > But restrict to Wong only:

    > +1: -1 below UBZII pivot, -3 below KO
    > +2: at UBZII pivot, -2 below KO
    > +3: 1 above UBZII pivot, -1 below KO
    > +4: 2 above UBZII pivot, at KO
    > +5: 3 above UBZII pivot, 1 above KO

    > UBZII only has the "accuracy
    > advantage" at +1 and +2,
    > KO has the rest, which is probably just
    > enough to overcome the
    > level 1 versus level 2.

    > Another observation. KO's disadvantage at
    > negative counts would hurt it more than it
    > does were it not for the fact that only
    > around 1 unit is bet there anyway.

    > The reverse happens to UBZII at high counts,
    > since we are at max anyway.

    > So, when we Wong, the key counts are +2, +3,
    > +4, and +5, and KO has the accuracy edge
    > there.

    > Or so it seems to me, just trying to
    > "thought experiment" it.

    > Also, if what I am thinking is right, then
    > it should hold that the larger the Wonging
    > spread, the more KO should inch ahead, and
    > the reverse.

    > But I could be wrong. Just trying to think
    > what could explain those results.

    It's my thinking that's probably wrong here. I sometimes confuse UBZII with Zen. I couldn't see any sense that K-O would outperform a level-two balanced count, but all the while we were speaking about two unbalanced counts. I think you're right.

    Don

  3. #16
    Cacarulo
    Guest

    Cacarulo: Re: Jumping in here

    > UBZII only has the "accuracy
    > advantage" at +1 and +2,
    > KO has the rest, which is probably just
    > enough to overcome the
    > level 1 versus level 2.

    Correct.

    > Also, if what I am thinking is right, then
    > it should hold that the larger the Wonging
    > spread, the more KO should inch ahead, and
    > the reverse.

    You don't gain that much with a larger spread.

    1-12:
    KO = $65.30 (-6)
    UBZII = $63.85 (-5)

    1-20:
    KO = $65.46 (-6)
    UBZII = $63.93 (-5)

    Even with a smaller spread KO outperforms UBZ2.

    1-6:
    KO = $64.69 (-5)
    UBZII = $63.30 (-3)

    Sincerely,
    Cacarulo

  4. #17
    Don Schlesinger
    Guest

    Don Schlesinger: BJRM 2002

    I just can't seem to get K-O to outperform UBZII with any of the canned BJRM sims.

    Cac, why do you think using four more indices would change anything?

    Try optimal Wonging 1-6, 5/6, das, s17, LS, and let me know which count wins. I'm getting UBZII by about $2 or so.

    Don

  5. #18
    Cacarulo
    Guest

    Cacarulo: Re: BJRM 2002

    > I just can't seem to get K-O to outperform
    > UBZII with any of the canned BJRM sims.

    Try 1-6 with S17,DAS,5/6 and you will find that KO is a little better than UBZII. BTW, I have BJRM2000.

    KO = $62.88 (-1)
    UBZII = $62.73 (-2)

    > Cac, why do you think using four more
    > indices would change anything?

    I don't think. It just happens Those four indices are very important for KO. Besides, in BJRM, KO is not using the I18 but the KO preferred. An apple to apple comparison would require the same set of indices which is what I do (C22 against C22).

    > Try optimal Wonging 1-6, 5/6, das, s17, LS,
    > and let me know which count wins. I'm
    > getting UBZII by about $2 or so.

    LS is another thing. In that scenario I agree that
    UBZII is better than KO.

    Sincerely,
    Cacarulo

  6. #19
    Don Schlesinger
    Guest

    Don Schlesinger: Re: BJRM 2002

    > Try 1-6 with S17,DAS,5/6 and you will find
    > that KO is a little better than UBZII. BTW,
    > I have BJRM2000.

    > KO = $62.88 (-1)
    > UBZII = $62.73 (-2)

    The latest version, BJRM 2002, gives KO (-1) = $62.88, and UBZII (-2) = $62.73. So, there is no change, and you are correct. However, for K-O, it says BOTH K-O preferred AN I18 and Fab4. What does that mean??

    > I don't think. It just happens Those four
    > indices are very important for KO. Besides,
    > in BJRM, KO is not using the I18 but the KO
    > preferred. An apple to apple comparison
    > would require the same set of indices which
    > is what I do (C22 against C22).

    Why would those four indices be any more important to K-O than to UBZII? Most of them are actually closer to the UBZII pivot than to the K-O pivot, no?

    > LS is another thing. In that scenario I
    > agree that UBZII is better than KO.

    So it seems.

    Don

  7. #20
    Cacarulo
    Guest

    Cacarulo: Re: BJRM 2002

    > The latest version, BJRM 2002, gives KO (-1)
    > = $62.88, and UBZII (-2) = $62.73. So, there
    > is no change, and you are correct. However,
    > for K-O, it says BOTH K-O preferred AN I18
    > and Fab4. What does that mean??

    I don't know. John? Is it KO preferred or I18?

    > Why would those four indices be any more
    > important to K-O than to UBZII? Most of them
    > are actually closer to the UBZII pivot than
    > to the K-O pivot, no?

    You're right. The only index in KO that is exactly at the pivot is 8v5. The others are closer in UBZII. But, insurance is closer in KO as well as ten-splittings. I think the difference comes from these indices. Also, my guess is that KO using I18 should be more powerful than UBZII using I18. What do you think?

    Sincerely,
    Cacarulo

  8. #21
    John Auston
    Guest

    John Auston: Re: BJRM 2002

    > However,
    > for K-O, it says BOTH K-O preferred AN I18
    > and Fab4. What does that mean??

    Off the top of my head, I think I remember that KO Preferred is a technique whereby you do not use an individual and unique RC for each index you use, but rather you group the indices into one of 3 (I think) categories, and use the same single index value for each member of the group.

    So, I used the Perferred indices, and assigned one of the group's RC to each of the I18 and F4.

    "Preferred" and "I18 F4" can exist together.

  9. #22
    Cacarulo
    Guest

    Cacarulo: Preferred does not

    include ten-splittings. Besides, because of the rounded matrix issue, KO using the correct sweet-16 should be better than KO preferred.

    Sincerely,
    Cacarulo

  10. #23
    John Auston
    Guest

    John Auston: Re: Preferred does not

    > include ten-splittings. Besides, because of
    > the rounded matrix issue, KO using the
    > correct sweet-16 should be better than KO
    > preferred.

    Right. But my canned sims probably did split 10's, I just assigned them to one of the Preferred Groups (probably pivot).

    John

  11. #24
    Cacarulo
    Guest

    Cacarulo: Re: Preferred does not

    > Right. But my canned sims probably did split
    > 10's, I just assigned them to one of the
    > Preferred Groups (probably pivot).

    Ok. but then we agree that those "Preferred I18" are inferior to the "Correct I18". For example, correct split 10's are two points over the pivot.
    On the other hand, UBZII was calculated using the "Correct I18". The conclusion is that we are not comparing apples to apples and that somehow explains why BJRM doesn't show KO as a winner in Wonging mode.

    Sincerely,
    Cacarulo


  12. #25
    Norm Wattenberger
    Guest

    Norm Wattenberger: Comparison methodologies


    Choosing the number and value of indexes obviously affects results. In creating the CVCX canned sims, I chose to use the indexes in the respective books. This is clearly not a fair comparison of the base tag values. But, it is the way the vast majority of people play. If we wish to compare KO and UBZII, we need to decide first are we comparing the power of the tag values ? or the entire strategies as presented in the respective books. One of the KO indexes isn?t even the correct sign. But it is still correct in my mind given the aim of KO because it is easier to remember and has only a small impact on results. UBZII also makes compromises. I chose to compare strategies ? not tag values. Just my choice, not right or wrong. Of course the user has the ability to run his/her own sims.



  13. #26
    Don Schlesinger
    Guest

    Don Schlesinger: Re: Comparison methodologies

    > I chose to compare
    > strategies ? not tag values. Just my choice,
    > not right or wrong. Of course the user has
    > the ability to run his/her own sims.

    Personally, I don't choose either, as fair comparisons are impossible that way.

    Tag values are a PART of an overall approach for a count system. Clearly, other factors come into play. But, it makes little sense to say that since the author suggests we do the following, this is now how we ought to "consider" such and such a count system. If Carlson gives us 75 indices and Fuchs gives us 16, does that mean that when doing side-by-sides, this is how we decide which system is "better"?

    For example, who could possibly care what Lance Humble suggested, over 25 years ago, for Hi Opt II? He didn't have at his disposal the tools that we do today. Ditto for Revere, Uston, etc.

    So, we take their tag values, and then we use today's capabilities to build an approach that optimizes the value of the tags. And then, we compare -- apples-to-apples.

    You can't take a vote and ask people how they play and then decide that this is the "power" of a system. Because, you'll get 50 differing votes, so what good is it? And, frankly, as well-intentioned as the authors of the systems themselves might have been in their books or Web sites, etc., who really cares what they think? If you don't standardize your report and comparisons, what's the point?

    Don

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.